Vol. 37 32 No. No. 132 2020 2014 Vol.
UNDERWATER TECHNOLOGY
1
A Personal View... It’s time to reveal a long love affair
Judith Patten
29
Technical Briefing A simple in situ labelling approach and adequate tools for photo and video quadrats used in underwater ecological studies
3
Bernabé Moreno
Zhigang Deng, Mohammed Tousif Zaman and Zhenzhong Chu
35
Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle
13
Case studies in estimating subsea systems’ readiness level
ISSN 1756 0543
Book Review Image-Based Damage Assessment for Underwater Inspections
Sirous F. Yasseri and Hamid Bahai
www.sut.org
0A-SUT-37(1)-OFC.indd 1
3/17/2020 6:47:04 PM
UNDERWATER TECHNOLOGY Editor Dr MDJ Sayer Scottish Association for Marine Science Assistant Editor E Azzopardi SUT Editorial Board Chairman Dr MDJ Sayer Scottish Association for Marine Science Gavin Anthony, GAVINS Ltd Dr MA Atamanand, National Institute of Ocean Technology, India LJ Ayling, Maris International Ltd Commander Nicholas Rodgers FRMetS RN (Rtd) Prof Ying Chen, Zhejiang University Jonathan Colby, Verdant Power Neil Douglas, Viper Innovations Ltd, Prof Fathi H. Ghorbel, Rice University G Griffi ths MBE, Autonomous Analytics Prof C Kuo FRSE, Emeritus Strathclyde University Dr WD Loth, WD Loth & Co Ltd Craig McLean, National Ocean and Atmospheric Administration Dr S Merry, Focus Offshore Ltd Prof Zenon Medina-Cetina, Texas A&M University Prof António M. Pascoal, Institute for Systems and Robotics, Lisbon Dr Alexander Phillips, National Oceanography Centre, Southampton Prof WG Price FRS FEng, Emeritus Southampton University Dr R Rayner, Sonardyne International Ltd Roland Rogers CSCi, CMarS, FIMarEST, FSUT Dr Ron Lewis, Memorial University of Newfoundland Prof R Sutton, Emeritus Plymouth University Dr R Venkatesan, National Institute of Ocean Technology, India Prof Zoran Vukić, University of Zagreb Prof P Wadhams, University of Cambridge Cover Image (top): zoonar.com/syrist Cover Image (bottom): Steve Crowther Cover design: Quarto Design/ kate@quartodesign.com
Society for Underwater Technology Underwater Technology is the peer-reviewed international journal of the Society for Underwater Technology (SUT). SUT is a multidisciplinary learned society that brings together individuals and organisations with a common interest in underwater technology, ocean science and offshore engineering. It was founded in 1966 and has members in more than 40 countries worldwide, incIuding engineers, scientists, other professionals and students working in these areas. The Society has branches in Aberdeen, London and South of England, and Newcastle in the UK, Perth and Melbourne in Australia, Rio de Janeiro in Brazil, Beijing in China, Kuala Lumpur in Malaysia, Bergen in Norway and Houston in the USA. SUT provides its members with a forum for communication through technical publications, events, branches and specialist interest groups. It also provides registration of specialist subsea engineers, student sponsorship through an Educational Support Fund and careers information. For further information please visit www.sut.org or contact: Society for Underwater Technology 2 John Street, London WC1N 2ES e info@sut.org
Scope and submissions The objectives of Underwater Technology are to inform and acquaint members of the Society for Underwater Technology with current views and new developments in the broad areas of underwater technology, ocean science and offshore engineering. SUT’s interests and the scope of Underwater Technology are interdisciplinary, covering technological aspects and applications of topics including: diving technology and physiology, environmental forces, geology/geotechnics, marine pollution, marine renewable energies, marine resources, oceanography, salvage and decommissioning, subsea systems, underwater robotics, underwater science and underwater vehicle technologies. Underwater Technology carries personal views, technical papers, technical briefings and book reviews. We invite papers and articles covering all aspects of underwater technology. Original papers on new technology, its development and applications, or covering new applications for existing technology, are particularly welcome. All papers submitted for publication are peer reviewed through the Editorial Advisory Board. Submissions should adhere to the journal’s style and layout – please see the Guidelines for Authors available at www.sut.org.uk/journal/default.htm or email elaine.azzopardi@sut.org for further information. While the journal is not ISI rated, SUT will not be charging authors for submissions.
in more than 40 countries worldwide, including over 190 Corporate Members of the Society.
Disclaimer and copyright The Society does not accept responsibility for the technical accuracy of any items published in Underwater Technology or for the opinions expressed in such items. The copyright of any paper published in the journal is retained by the author(s) unless otherwise stated. All authors are supplied with a PDF version of their papers once published. Authors are encouraged to make the PDF version of their papers free to download from their own websites.
Open Access Underwater Technology is available as Open Access. PDF versions of all published papers from Underwater Technology may be accessed via ingentaconnect at www. ingentaconnect.com/content/sut/unwt. All issues from Volume 20 (1995) onwards are available as Open Access. The Society for Underwater Technology also encourages Underwater Technology authors to make their papers available online on their personal and/or institutional websites for Open Access. Through this arrangement, the Society supports the Open Access policy not only in the UK (the Research Councils UK (RCUK) policy) but also the drive towards Open Access in other countries.
Abstracting and indexing Underwater Technology is included in Emerging Sources Citation Index. Additional abstracting and indexing services include American Academy of Underwater Sciences (AAUS) E-Slate; Aquatic Sciences and Fisheries Abstracts (Biological Sciences and Living Resources; Ocean Technology, Policy and Non-Living Resources; and Aquatic Pollution and Environmental Policy); Compendex; EBSCO Discovery Service; Fluidex; Geobase; Marine Technology Abstracts; Oceanic Abstracts; Scopus; and WorldCat Discovery Services.
Subscription Subscription to the print version of Underwater Technology is available to non-members of the Society at the following rates per volume (single issue rates in brackets). Prices are given in GBP. Accepted methods of payment are cheque or credit card (MasterCard and Visa). Foreign cheques must be in GBP and drawn on a British bank otherwise a currency conversion surcharge is incurred. UK subscription Overseas subscription
£102.00 (£25.50 per issue) £108.00 (£27.00 per issue)
Underwater Technology is also available in electronic format via ingentaconnect as Open Access. To subscribe to the print version of the journal or for more information please email Elaine Azzopardi at elaine.azzopardi@sut.org
Publication and circulation Underwater Technology is published in March, July and November, in four issues per volume. The journal has a circulation of 2,400 copies to SUT members and subscribers
0B-SUT-36(2)-IFC.indd 1
Advertising To book an advert or for more information please contact Elaine Azzopardi at elaine.azzopardi@sut.org
7/22/19 9:02 PM
www.sut.org
A Personal View...
doi:10.3723/ut.37.001 Underwater Technology, Vol. 37, No. 1, pp. 1–2, 2020
It’s time to reveal a long love affair I am not an engineer (although I plan to be one in my next life), a scientist or a technologist, yet my life for almost 45 years has been heavily involved with the offshore oil and gas industry and oceanology, and also the ‘new kid on the block’, renewable and low-carbon energy. When asked to write a Personal View for Underwater Technology, true to form I jumped right in enthusiastically and accepted this invitation. I hoped that inspiration would hit me at an appropriate moment and indeed it has, for writing this on Valentine’s Day I have determined to come clean about a long, long love affair. I trained and worked in public relations for many years, firstly in a highly specialised consultancy, then in an advertising agency during the buzz and excitement of the 1960s. After I had my first child, I started my own PR business in 1969. However, it was in 1975 that my love affair truly began. I was asked to handle PR for a conference on China’s oil industry held in Glasgow and organised by the Sino-British Trade Council. There was one slight snag. I was due to give birth to my second child just six weeks before the conference. By the time my son arrived, I was looking after delegate registration and diving ever further into the programme planning for the conference. I gave up working for a day to give birth, and afterwards my husband brought the files into hospital (women were in hospital for up to 10 days then), plus the messages from my answering machine, and posted my hand-written notes out to people. I arrived in Glasgow
and discovered immediately how incredible conferences are: the warmth from people who had heard my answering machine message, ‘Sorry I’m in hospital having a baby’, the incredible buzz when journalists lined up to interview the Chinese Ambassador under the stairs, seeing the productive interaction between delegates. I was hooked! There, I met the inspirational David Stott, founder of Spearhead Exhibitions. He created Offshore Europe in Aberdeen, and shortly after the Glasgow conference asked if I would work on that mighty event. I did so for a quarter of a century, handling PR and marketing, and constantly marvelling at the ingenuity of engineers, learning more and more about the industry, and relishing its rapid development. During the 1975 show I was speaking with the late George Williams, then director of the United Kingdom Offshore Operators Association (UKOOA), which is now Oil & Gas UK, or OGUK) when two young men came over to ask how he, as a Shell Exploration manager, had found oil in Nigeria. I sensed their excitement at meeting him and listened to the story – it had just as much buzz and excitement as the advertising scene of the 1960s. I was already feeling the proverbial ‘round peg in a round hole’, and marvelling at how industry events bring people together. In around 1976 I undertook a survey to find out how many people worked in the exploration companies in Aberdeen. The total was something like 1200, and within a very short time after
Judith Patten has directed her PR company, JPPR, since 1969, with the exception of the time when she joined Spearhead Exhibitions for 18 months (from autumn 1988 to spring 2000). While her company started as a general PR consultancy, now her clients are all involved with some aspect of marine engineering. She was presented with the Scottish Green Energy Outstanding Contribution Award in 2012, awarded MBE in The Queen’s Birthday Honours ‘for services to renewable energy’ in 2014, and in the same year became a proud SUT Fellow.
this Shell and BP each employed that many in the city. It was exciting to witness the growth not only of the industry but of innovative solutions too – and engineers of all types and levels of seniority were only too glad to explain their projects and technical developments both at Offshore Europe and to me. David Stott acquired Oceanology International in the first half of the 1980s, creating opportunities to meet further inspirational people and learn about new solutions. I was ‘loaned’ to work in the press office at Stavanger’s Offshore Northern Seas (ONS) for two shows and found myself being interviewed by a journalist
1
01-SUT346-37(1)-ut69505_Color.indd 1
3/20/2020 1:02:12 PM
Judith Patten. It’s time to reveal a long love affair
who quoted me as saying, ‘I love exhibitions’ – the only words in English in his copy. Yes, I had well and truly got the bug! Then there were the niche events on inspection, repair and maintenance (IRM), platform weight, helicopters and more, and shows in Rio, Caracas and Azerbaijan. There was always the same incredible buzz on the show floor as people met face-to-face – there really is no substitute! My love for events grew, and as David planned to retire in 1999, I decided to make my PR business’ clients and team redundant in order to join him at Spearhead Exhibitions for his final 18 months so I could learn ‘at the feet of the master’. It was at this time that we both started talking about renewable energy: was it becoming commercial enough to have an exhibition of its own? The Board of Spearhead Exhibitions was not enthusiastic about this idea. After David retired, I continued working on Offshore Europe, and it was thanks to a phone call to the Economic Development Unit at Aberdeen City Council, when it was explained to me, ‘We are no longer the oil and gas capital, we are the Energy City’, and the answer of ‘yes’ to my question,
‘Does that include renewable energy?’, that the next, and perhaps most exciting, period of my working life began. In the early 2000s, David and I created the renewables event that we had wondered about, and which has now grown to become the UK’s largest renewable and low-carbon energy exhibition and conference: All-Energy. SUT became involved with AllEnergy right from the start after we bumped into Ian Gallett, SUT’s then CEO, at a show. He was there to see if there was scope for an SUT renewables conference. Over a cup of coffee, we decided to work together, and every year since, members of SUT’s Marine and Renewable Energies Committee (MREC) work with me on the marine renewables stream of the conference. The show was acquired by Reed Exhibitions in 2011. It stayed in Aberdeen until 2014 and then moved to Glasgow, where we will celebrate its 20th anniversary in May 2020. I have the fantastic challenge of creating a 600-speaker conference (yes, 600 speakers in just two days!), handle PR for the event each year, and act as its ‘ambassador’. I’m still on a learning curve. It is an event with
enormous purpose as we consider the climate emergency that will bring over 190 nations together at COP26 in Glasgow in November 2020; All-Energy will be a stepping-stone to this, and a legacy event. I believe that now, as the greatest beneficiary of industrialisation and fossil fuels, the UK has a responsibility to lead the world in the next great transformation: decarbonisation and tackling climate change. We must demonstrate that the challenge can be faced head-on and society can, and must, continue to thrive and prosper in order to create transformational changes to our economy. It is discussion of those changes and innovatively engineered solutions that will draw some 8000 people to All-Energy 2020. In almost 45 years of involvement with industry-led exhibitions and conferences, I have met the most incredible people, seen business being conducted first-hand, and experienced the value of people meeting each other. I have learned so much about innovative engineering, thrived on adrenalin coursing through my body, and loved every minute of it. Long may it continue!
2
01-SUT346-37(1)-ut69505_Color.indd 2
3/20/2020 1:02:14 PM
www.sut.org
Technical Paper
doi:10.3723/ut.37.003 Underwater Technology, Vol. 37, No. 1, pp. 3–11, 2020
Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle Zhigang Deng1*, Mohammed Tousif Zaman2 and Zhenzhong Chu3 1 Information Engineering College, Shanghai Maritime University, Shanghai, 201306, China 2 The Computation Intelligence and Bionic Robotics Laboratory, Robotics Engineering, Worcester Polytechnic Institute, Worcester, MA, 01609, USA 3 Shanghai Engineering Research Center of Intelligent Maritime Search & Rescue and Underwater Vehicles, Shanghai Maritime University, Shanghai, 201306, China Received 8 September 2019; Accepted 6 January 2020
Abstract Unmanned underwater vehicles (UUVs) move in dynamic environments and need to avoid other non-cooperative obstacles while executing a task, such as tracking a target or a special trajectory. It is a challenge to avoid collisions with moving obstacles in the tracking process. The present paper describes the implementation of horizonplane adaptive cruise control, which follows a given desired trajectory using control Lyapunov functions while satisfying constraints specified by a control barrier function to avoid collision with obstacles. The Lyapunov function is treated as a soft constraint, and the barrier function as hard constraint for the UUV; both are satisfied simultaneously using quadratic programming. Finally, the present paper describes a simulation of avoiding moving obstacles while tracking a target, with the results showing this as effective and feasible. Keywords: unmanned underwater vehicles (UUV), trajectory tracking, collision avoidance, control Lyapunov function (CLF), control barrier function (CBF), quadratic programming
1. Introduction Unmanned underwater vehicles (UUVs) are used in the scientific, oil and gas, and military communities. Trajectory tracking control based on adequate position sensors and advanced control approaches is a fundamentally important issue for UUVs. The complexity and unpredictability of the underwater environment, underactuated nature of the system, and their strong coupling and nonlinear * Contact author. Email address: dzg1026@126.com
characteristics make tracking control of UUVs a challenging area of research. Barrier functions’ value on a particular point of interest increases to infinity as the point approaches the boundary of a given unsafe region. A barrier function can be considered as a wall that cannot be surpassed. The extension of barrier function to control theory results in control barrier function (CBF; Liu and Li, 2018). In dynamic systems, a Lyapunov function helps to determine the stability of the system. A control Lyapunov function (CLF) is a Lyapunov function for a system with control inputs. Quadratic programming can be used to combine CLF and CBF to achieve the control objectives specified by CLF and the constraint conditions specified by CBF (Romdlony and Jayawardhana, 2014). The present paper proposes a novel control method using control Lyapunov functions while satisfying constraints specified by a control barrier function to avoid collision with obstacles. Previous studies focusing on tracking control have been undertaken, for example: sliding-mode control (Elmokadema et al., 2017; Yan et al., 2019); backstep control (Wu et al., 2001; Lapierre and Jouvencel, 2008; Sun et al., 2014); neural-network control (Pepijn et al., 2005; Bagheri et al., 2010; Gao and Guo, 2018); fuzzy control (Zhang et al., 2009; Ishaque et al., 2010; Xiang et al., 2017; Deng et al., 2018); classical adaptive control (Kennedy et al., 2007; Ames et al., 2014; Nguyen and Sreenath 2016; Xu et al., 2018; Antonelli et al., 2003; Sahu and Subudhi, 2014; Li et al., 2012; Peng et al.,
3
02-SUT332-37(1)-ut69521_Color.indd 3
3/20/2020 12:38:47 PM
Deng et al. Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle
2019); and path planning and obstacle avoidance (Singh et al., 2018; Singh et al., 2017). The sliding-mode kinematic controller is one of the most commonly used approaches for mobile robot tracking control and tracking control of UUVs. Sliding-mode control has outstanding performance, including insensitivity to parameter variations and good rejection of disturbances, and is suitable for robust tracking control of UUVs. However, a major drawback of the sliding-mode approach is the high frequency of control action, or chattering (Elmokadema et al., 2017; Yan et al., 2019). This high-frequency control activity causes high heat loss in electrical power circuits and premature wearing of actuators. In addition, such high control activity may excite unmodeled high-frequency dynamics, which in turn causes controller performance degradation. The backstep controller has been adopted in mobile robot and UUV control systems (Wu et al., 2001; Lapierre and Jouvencel, 2008; Sun et al., 2014). The main objective of the backstep algorithm is to define a velocity controller that stabilises the closed-loop system. The design of the velocity controller is simple, and system stability can be strictly proven by Lyapunov stability theory. This approach can handle large initial state errors. However, the velocity control law is directly related to the state errors, and therefore large velocities can be generated in large initial error conditions, causing significant speed increases in sudden tracking errors. This can cause the control inputs to exceed their control constraints at velocity jump points. In order to address this disadvantage of classical model-based control, sliding-mode control, and the backstep method, some controllers generate appropriate control inputs by using neural networks (Pepijn et al., 2005; Bagheri et al., 2010; Gao and Guo, 2018). In the present study, an exact underwater vehicle model was not required, and the thruster and vehicle nonlinearities can be implemented. However, an important disadvantage of such tracking control systems is that the neural network model requires online and/or offline learning to make the UUV perform properly. This learning procedure is computationally expensive and may not be suitable for the real-time situation. Fuzzy tracking control approaches do not require online or offline learning, and their computation procedure is simple (Zhang et al., 2009; Ishaque et al., 2010; Xiang et al., 2017; Deng et al., 2018). They provide a solution to the problem of large initial velocities, but the major drawback of fuzzy control is the formulation of fuzzy rules, which are usually obtained by trial and error-based human knowledge.
In addition to the described control methods, model-based solutions have been proposed for motion control (Kennedy et al., 2007; Ames et al., 2014; Nguyen and Sreenath, 2016; Xu et al., 2018; Antonelli et al., 2003; Sahu and Subudhi, 2014; Li et al., 2012; Peng et al., 2019). In Ames et al. (2014), the control input to the ego vehicle is calculated by combining the control barrier function and the control Lyapunov function through quadratic programming for implementation of adaptive cruise control. The present paper assumes a straight-line trajectory, and one constraint defined by the barrier function of the vehicle in front. Nguyen and Sreenath (2016) introduces exponentials for nonlinear systems. The pole placement helps to balance the stability of the tracking system and the safety constraints. In Xu et al. (2018), a combination of control barrier functions and control Lyapunov functions, or black-box legacy controller, is used through quadratic programming, which serves two fundamental purposes. For the UUV, uncertainties stem from modelling errors, and unknown hydrodynamic parameters and forces caused by ocean currents in an underwater environment. To prevent the velocity constraint violation, Li et al. (2012) employ the barrier Lyapunov function in Lyapunov synthesis. By ensuring the boundedness of the barrier Lyapunov function, it is guaranteed that the velocity constraints are not transgressed. Peng et al. (2019) construct the Lyapunov function and formulate the command governor as a quadratically constrained quadratic programming problem. The efficacy of the proposed anti-disturbance constrained control method for autonomous underwater vehicles is substantiated via simulations and comparisons. The present study constructed a kinematic model of the UUV Sea-Kite II, and obtained the tracking error formulation in the body frame based on the plane UUV kinematic model. Based on the tracking control methods for UUVs discussed previously, it is difficult to choose a single method that addresses all the described problems. The adaptive cruise control algorithm in Ames et al. (2014) have two major limitations: the trajectory of the ego vehicle and front vehicle was restricted to one dimension, and the front vehicle was constrained to have a constant time-invariant velocity, which is rarely the case in real-life scenarios. Therefore, the present authors constructed a plane kinematic model of the UUV (Sea-Kite II), and relaxed the two constraints with reference to adaptive cruise control in the collision avoidance with a moving obstacle in the process of trajectory tracking. A plane trajectory tracking control law was designed by
4
02-SUT332-37(1)-ut69521_Color.indd 4
3/20/2020 12:38:48 PM
Underwater Technology Vol. 37, No. 1, 2020
Inertial frame O
φ
X
Body-fixed frame
Y
Q
θ
ψ Z
φ
pitch q
2. Kinematic model of UUV
sway
surge
Yo p roll
w heave r yaw
results are then presented for two cases: straightline and curve-line movement. In the last section, conclusions and discussion of future scope for improvement are presented.
Xo
Zo
Fig 1: UUV coordinate systems
using control Lyapunov functions, while satisfying constraints specified by a control barrier function to avoid obstacles. Two different kinds of movement of obstacle collisions were avoided while tracking a spline curve trajectory by the plane trajectory tracking controller. The present authors conclude the simulation results to be effective and feasible. Section 2 of the present paper describes the kinematic model for the unmanned underwater vehicle. Section 3 discusses the principle of adaptive cruise controller. Simulation experiments and their
It is necessary to obtain a model of the UUV system with an assigned frame of reference. It is possible to build up two frames: inertial frame and body-fixed frame, as shown in Fig 1. There are six degrees of freedom (DOF): surge, sway, heave, roll, pitch and yaw. The present study defines q = [u v w p q r ]T as the UUV spatial velocity state vector with respect to its body-fixed fame, and η = [x y z φ θ Ψ]T as the position and orientation state vector with respect to the inertial frame. The specifications of the vehicle are illustrated in Table 1. According to the thruster distribution as shown in Fig 2, Sea-Kite II is an under-actuated UUV, and there are only four DOFs: surge, heave, pitch and yaw. For horizon-plane motion control, without lateral force, there are two DOFs: surge and yaw, the nonholonomic constraints can be calculated as follows ( Jiang et al., 2000): . x sin Ψ − y. cos Ψ = 0 (1)
Table 1: Specifications of the UUV Parameter
Value
Parameter
Value
Dimensions Weight Power supply Thrusters
(1080 710 480) mm 75 kg 150 VDC, 1.5 Kwh 4 total: 2 horizontals and 2 verticals
Maximal speed Depth rating Video cameras Lighting
1.5 (m/s) 300 m 1/3” DSP CCD LED lights
Y
4TH
28 cm
X O
1. cm
3TH
54 cm 1TH
2TH
Fig 2: The structure and thruster distribution of Sea-Kite II
5
02-SUT332-37(1)-ut69521_Color.indd 5
3/20/2020 12:38:48 PM
Deng et al. Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle
move, so that the velocity q of the unmanned underwater vehicle can track a smooth velocity qd, and the trajectory η of the unmanned underwater vehicle follows a given reference trajectory ηd smoothly and accurately.
3. The adaptive cruise controller The objective of the present study was to combine barrier function (hard constraint) and Lyapunov constraint (soft constraint) for the goal of trajectory tracking of a UUV in the presence of a moving UUV and obstacles in its surroundings. According to Ames et al. (2014), a nonlinear affine control system can be defined by the equation: Fig 3: Tracking error in body coordinate frame of underwater vehicle
In the trajectory tracking task, the UUV must follow a desired path in Cartesian workspace with a specified timing law. The desired path satisfies the UUV nonholonomic constraint. Equivalently, it tracks a path generated by a reference UUV (Fig 3). For the horizontal-plane motion control of the UUV, the desired reference state of a vehicle is described as ηd = [xd yd Ψd]T, qd = [ud rd]T, where qd = [ud rd]T is the desired velocity in the body-fixed frame, ηd = [xd yd Ψd]T is the desired state of the UUV in the inertial frame, qc = [uc rc]T is the output of velocity controller in the body-fixed frame, and the actual state of the UUV is represented by η = [x y Ψ]T and q = [u r ]T. The state errors are denoted by e = [ex ey eΨ]T in the body frame. Here, ηd − η = [xd−x yd−y Ψd−Ψ]T is the tracking error in the inertial frame. J is the conversion function from the inertial frame to the body-fixed frame for the position and orientations. ⎡ cos Ψ sin Ψ ⎢ = ⎢− sin Ψ cos Ψ ⎢ ⎢ 0 0 ⎣
0⎤ ⎥ 0⎥ ⎥ 1 ⎥⎦
(2)
The tracking error formulation in the body frame can therefore be described as e = J ⋅ (ηd−η). For horizon-plane motion control of Sea-Kite II, the kinematic model of the UUV is then given by: ⎡x ⎤ ⎡ cos Ψ 0⎤ ⎢ ⎥ ⎢ ⎥ η = ⎢⎢ y ⎥⎥ = ⎢ sin Ψ 0⎥ q c ⎢ ⎥ ⎢ ⎥ ⎢ 0 1 ⎥⎦ ⎢⎣ Ψ ⎥⎦ ⎣
(3)
Therefore, the motion tracking problem of the unmanned underwater vehicle is as follows: design a control law for the thruster to drive the UUV to
x
f (x z ) + g (x z )u
z
q (x z ),
(4)
where x ∈ X are the output states, z ∈ Z are the uncontrolled states and U are the permissible values for u, the control input. Quadratic programming combines CLF and CBF into a single controller in the equation: u ∗ (x z )
a g
⎛1 i ⎜⎜⎜ ⎝2
T
⎞ T H (x z ) u + F (x z ) u⎟⎟⎟ ⎠
⎡u ⎤ u = ⎢ ⎥ ∈ R m +1 , ⎢⎣δ ⎥⎦
(5)
where δ is the relaxation factor for the soft constraint, such that Ψ0(x, z) + ΨT1 (x, z)u ≤ δ is the CLF γ and L f B (x z ) + Lg B (x , z )u ≤ is the CBF. B (x , z ) H(x, z)Rm+1*m+1 and F(x, z)Rm+1 are the arbitrarily chosen cost function based on the state-based weighting of control inputs; Lf B(x, z) is the lie derivative of V(x, z) along the vector field f(x, z); and LgB(x, z) is the lie derivative of V(x, z) along the vector field g(x, z). The objective of the tracking controller is to drive the UUV to track the set trajectory by controlling the surge and yaw motion speed, and the state errors e = [ex ey e Ψ]T converge to zero. The equivalent trajectory tracking error in the body-fixed fame can be expressed as: ⎡e x ⎤ ⎡ cos Ψ sin Ψ ⎢ ⎥ ⎢ e = ⎢⎢e y ⎥⎥ = ⎢− sin Ψ cos Ψ ⎢ ⎢ ⎥ ⎢ 0 0 ⎢⎣e Ψ ⎥⎦ ⎣
0 ⎤ ⎡x d − x ⎤ ⎥⎢ ⎥ 0⎥ ⎢ y − y ⎥ , ⎥⎢ ⎥ 1⎥⎦ ⎢⎣ Ψ d − Ψ ⎥⎦
(6)
where ex , e Ψ → 0, ey becomes uncontrollable. This can be avoided by introducing a change in the variable:
6
02-SUT332-37(1)-ut69521_Color.indd 6
3/20/2020 12:38:48 PM
Underwater Technology Vol. 37, No. 1, 2020
eΨ = e Ψ +
The Lyapunov candidate is given as follows:
ey π1
where, π1 = e x2
,
(7) V = π2 + k1π1 − (1 + k1),
e y2 + 1 and α > 0.
where π2 =
The system dynamics for the nonlinear affine control system can be defined by: x
f (x ) + g (x )u ,
(8)
where u = [u0u1] = [ucrc] and x = [e–Ψey−ex]T. Thus: 2 ⎡e ⎛ e ⎞⎤ ⎢ x rd + 1 + e x u d sin ⎜⎜ e Ψ − y ⎟⎟⎥ ⎢π ⎜⎝ α απ ⎟⎟⎠⎥ π13 1 ⎥ ⎢ 1 ⎢ ⎥ ⎛ ⎞ e ⎢ ⎥ e y ⎟ ⎟⎟ f x = ⎢ e x rd + u d sin ⎜⎜⎜ Ψ − ⎥, ⎟ ⎢ ⎥ α απ ⎝ 1⎠ ⎢ ⎥ ⎢ ⎥ − rde y ⎢ ⎥ ⎢ ⎥ ⎥⎦ ⎣⎢ ⎡ ⎤ − e e x y ⎢α − e x ⎥ 3 ⎥ ⎢ π π 1 1 ⎢ ⎥ g x = ⎢⎢ − e x 0 ⎥⎥ . ⎢ e 1 ⎥⎥ y ⎢ ⎢ ⎥ ⎢⎣ ⎥⎦
(9)
f
( , z ) + Lg B (x , z )u −
L f B (x , z ) =
γ ⎤⎥ ≤ 0 , (11) B (x , z )⎥⎦
−u o + u d cos (e ) h (x , z )(h (x , z ) + )
⎡ 1 Lg B (x , z ) = ⎢⎢ 0 10 h x , z ( ) (h (x , z ) + ⎢⎣ and u0 is the obstacle velocity.
⎤ ⎥, )⎥⎥⎦
(13)
⎡ e k1e y k1e x ⎤ ⎥ f x + εV Ψ 0 (x , z ) = ⎢ Ψ and ⎢π π π ⎥ ( ) 1 1 ⎦ ⎣ 2 ⎡ e k1e y k1e x ⎤ ⎥ g x ; substituting δsc = 0 would Ψ1 (x , z ) = ⎢ Ψ ⎢π π π ⎥ ( ) 1 1 ⎦ ⎣ 2
where
make the constraint ‘hard’ and would force exact convergence at rate of ε. The constraints defined previously are then combined by using quadratic programming: ⎛1
(x z ) a g i ⎜⎜ ⎝2
T
⎞ H acc + Facc T u⎟⎟⎟ ⎠
but are subject to constraints:
(10)
where:
Ψ0 (x, z) + Ψ1T (x, z) u ≤ δsc,
⎡u ⎤ u = ⎢ ⎥ ∈ R m +1 , ⎢⎣δssc ⎥⎦
where h(x) = z − 0.5. In the formulation of h(x); z is the distance between the vehicle and the obstacle; 0.5 units is the safety distance between the vehicle and obstacle that the vehicle must not cross. The hard constraint defined by the control barrier function is defined as: ⎡ inf ⎢⎢ u ∈R ⎣
+ 1 and k1 is a constant.
The Lyapunov soft constraint is given by the equation:
∗
The next step is to determine the hard and soft constraints for the control system. The barrier function is defined as: ⎛ h (x ) ⎞⎟ ⎟, B (x ) = − log ⎜⎜⎜ ⎜⎝1 + h (x )⎟⎟⎠
2 Ψ
(12)
Ψ0 ( ⎡ inf ⎢⎢ u ∈R ⎣
) Ψ1T (x , z )u ≤ δsc f
x , z )u − ( , z ) + Lg B (x,
γ ⎤⎥ ≤0 B (x , z )⎥⎦
The output of the controller are the surge velocity uc and the yaw rate rc, which are then fed to the UUV for trajectory tracking. The QP (CLF-CBF QP) is guaranteed to have a solution because the control objective is relaxed (Ames et al.; 2014). Based on the findings presented, the overall stability is guaranteed. Therefore, a stable tracking control can be achieved using the proposed control strategy for the UUV.
4. Simulation In order to verify the effectiveness of the algorithms of the control Lyapunov barrier for underwater vehicles, the method was simulated in MATLAB R2018b on a computer with Intel Core i7 CPU and dominant frequency of 2.7 GHz with 8GB RAM. The experiments were undertaken with the Lyapunov barrier control method, including two cases: straight-line and curve-line movement for the obstacle while tracking a given curve trajectory. The aim of the simulation was to illustrate the advantages of the proposed controller in driving a UUV onto a
7
02-SUT332-37(1)-ut69521_Color.indd 7
3/20/2020 12:38:52 PM
Deng et al. Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle
Table 2: Parameter values used in simulation Value
ucmin rcmax k1
0 (m/s) umax 0.671 (rad/s) α 1 γ 1
ε
Parameter
Value
3.5
1.1 (m/s) 1 0.5
3 2.5 z (m)
Parameter
4
2 1.5
2.5 Desired trajectory Obstacle trajectory Actual trajectory
2
1
1.5
0.5
1
0
0
5
10
y (m)
0.5
15
20
25
30
time (s)
0
Fig 6: The distance between UUV and the obstacle in straight-line movement
-0.5 -1 -1.5 -2 -2.5 -14
-12
-10
-8
-6
-4
-2
0
x (m)
Fig 4: The tracking trajectory while avoiding an obstacle in straight-line movement
desired trajectory while avoiding obstacles. The parameters used in the simulation are given in Table 2, which are the parameters for Sea-Kite II used at Shanghai Maritime University, Shanghai.
4.1. Avoidance with obstacle in straight-line movement The present study simulated tracking a target moving in straight line where the UUV needs to avoid an obstacle in straight-line movement. The target moves with a cubic spline curve trajectory, with: points = [−12.5 −2.5; −10.75 −0.25; −8.75 2; −5.75 1; −3.5 −1; −0.5 −0.5]; t = [0 6 12 18 24 30]; the target initial velocity: ud0 = 0 m/s. The obstacle moves in straight line which is described as follows: points = [−13.5 0.5; −10.15 0.5; −8.75 0.5; −6.8 0.5; −5.0 0.5; −4.0 0.5;]; t = [0 6 12 18 24 30]; the obstacle initial velocity: uo0 = 0 m/s. The initial
1
uc (m/s)
UUV Target 0.5
0 0
5
10
15
20
25
30
20
25
30
20
25
30
time (s)
psai (rad)
1 0.5 0 -0.5 0
5
10
15
rc (rad/s)
time (s)
0.2 0 -0.2 0
5
10
15
time (s)
Fig 5: The state of UUV while avoiding an obstacle in straight-line movement
8
02-SUT332-37(1)-ut69521_Color.indd 8
3/20/2020 12:38:57 PM
Underwater Technology Vol. 37, No. 1, 2020
uses deceleration and acceleration to avoid crossing the obstacle to meet the requirement for the safe distance.
2.5 Desired trajectory Obstacle trajectory Actual trajectory
2 1.5 1
y (m)
0.5 0 -0.5 -1 -1.5 -2 -2.5
-14
-12
-10
-8
-6
-4
-2
0
x (m)
Fig 7: The tracking trajectory while avoiding an obstacle in curve-line movement
UUV state is (−13.75,−2,0), and time varies from 0 to 30s. The sampling time T is 0.05s. Figs 4 to 6 show the results of the simulation. As can be seen in Fig 4, the UUV approaches the target quickly. The displacement and direction angle errors gradually decrease with the increase of simulation time, and the UUV successfully avoids the moving obstacle. Fig 5 shows the state of the robot including surge velocity, heading and yaw velocity during the target tracking. The state of the UUV approaches the target’s state gradually. Fig 6 shows the changing distances between the UUV and obstacles while avoiding the moving obstacle. The present authors conclude that the UUV mainly
4.2. Avoidance with obstacle in curve-line movement A further simulation of tracking a target moving in a curve-line was performed where the UUV needs to avoid an obstacle in curve-line movement. The target moves with a cubic spline curve trajectory, as in the previous simulation. The obstacle moves in cubic spline curve, with: points = [−13.5 2; −10.75 2; −8.75 2; −6.9 1.75; −5.5 0.75; −5.5 −0.25]; t = [0 6 12 18 24 30]; and initial velocity of obstacle is uo0 = 0 m/s. The initial UUV state is (−13.75, −2,0), and time varies from 0 to 30 s. The sampling time T is 0.05 s. As shown in Fig 7, the UUV approaches the target quickly. The displacement and direction angle errors gradually decrease with the increase of simulation time, and the UUV successfully avoids the moving obstacle. Fig 8 shows the state of the robot including surge velocity, direction and yaw velocity during the target tracking. The state of the UUV approaches the target’s state gradually. Fig 9 shows the changing distances between the UUV and obstacles while avoiding the moving obstacle. The present authors conclude that the UUV mainly use deceleration and acceleration, and pre-plans the path to avoid obstacle. It is clear that the UUV keeps the safe distance with the obstacle during the tracking procedure.
1
uc (m/s)
UUV Target 0.5
0 0
5
10
15
20
25
30
20
25
30
20
25
30
time (s)
psai (rad)
1 0.5 0 -0.5 0
5
10
15
rc (rad/s)
time (s) 0.2 0 -0.2 0
5
10
15
time (s)
Fig 8: The state of UUV while avoiding an obstacle in curve-line movement
9
02-SUT332-37(1)-ut69521_Color.indd 9
3/20/2020 12:38:57 PM
Deng et al. Collision avoidance with control barrier function for target tracking of an unmanned underwater vehicle
5.5 5 4.5 4
z (m)
3.5 3 2.5 2 1.5 1 0.5 0
0
5
10
15
20
25
30
time (s)
Fig 9: The distance between UUV and the obstacle in curve-line movement
5. Conclusion and future work The present paper discussed the trajectory tracking control method of using control Lyapunov functions while satisfying constraints specified by a control barrier function to avoid obstacles. Firstly, control Lyapunov barrier function was described, and the method of adaptive cruise control was proposed. Next, the present study described the simulations with trajectory tracking while avoiding the obstacle under two scenarios: straight-line and curve-line movement of the obstacle. From the simulation results, it is clear that the designed control system satisfies the tracking performance, and displays stability and robustness. However, a real scenario might contain numerous underwater vehicles, or stationary or moving obstacles surrounding the UUV (each incorporating a barrier function constraint on the UUV), and therefore the existing formulation for barrier functions can be generalised for handling multiple obstacles and using weighted barrier functions in future.
Acknowledgement The project described in the present paper is supported by the National Natural Science Foundation of China (51839004).
References Ames AD, Grizzle JW and Tabuada P. (2014). Control barrier function based quadratic programs with application to adaptive cruise control. In: Proceedings of IEEE 53rd Annual Conference on Decision and Control, 15–17 December, Los Angeles, USA, 6271–6278. Antonelli G, Caccavale F, Chiaverini S and Fusco G. (2003). A novel adaptive control law for underwater vehicles. IEEE Transactions on Control Systems Technology 11: 221–232.
Bagheri A, Karimi T and Amanifard N. (2010). Tracking performance control of a cable communicated underwater vehicle using adaptive neural network controllers. Applied Soft Computing 10: 908–918. Deng X, Yuan F, Zhu D and Xue F. (2018). Design and implementation of a remotely operated vehicle testbed. Underwater Technology 35: 13–22. Elmokadema T, Zribia M and Youcef-Toumib K. (2017). Terminal sliding mode control for the trajectory tracking of underactuated autonomous underwater vehicles. Ocean Engineering 129: 613–625. Gao Z and Guo G. (2018). Adaptive formation control of autonomous underwater vehicles with model uncertainties. International Journal of Adaptive Control and Signal Processing 32: 1–15. Ishaque K, Abdullah SS, Ayob SM and Salam Z. (2010). Single input fuzzy logic controller for unmanned underwater vehicle. Journal of Intelligent and Robotic Systems 59: 87–100. Jiang XS, Feng XS and Wang DT. (2000). Unmanned underwater vehicle. Shengyang: Liaoning Science and Technology Publishing House, 404 pp. Kennedy J, Gamroth E, Bradley C and Proctor AA. (2007). Decoupled modelling and controller design for the hybrid autonomous underwater vehicle: MACO. Underwater Technology 27: 11–21. Lapierre L and Jouvencel B. (2008). Robust nonlinear pathfollowing control of an AUV. IEEE Journal of Oceanic Engineering 33: 89–102. Li Z, Yang C, Ding N, Bogdan S and Ge T. (2012). Robust adaptive motion control for underwater remotely operated vehicles with velocity constraints. International Journal of Control, Automation and Systems 10: 421–429. Liu Y-H and Li H. (2018). Adaptive asymptotic tracking using barrier functions. Automatica 98: 239–246. Nguyen Q and Sreenath K. (2016). Exponential control barrier functions for enforcing high relative-degree safety-critical constraints. In: Proceedings of 2016 American Control Conference (ACC), 6–8 July, Boston, USA, 322–328. Peng Z, Wang JS and Wang J. (2019). Constrained control of autonomous underwater vehicles based on command optimization and disturbance estimation. IEEE Transactions on Industrial Electronics 66: 3627–3635. Pepijn WJ, Colinvan VF and Daniel T. (2005). Neural network control of underwater vehicles. Engineering Applications of Artificial Intelligence 18: 533–547. Romdlony MZ and Jayawardhana B. (2014). Uniting control Lyapunov and control barrier functions. In: Proceedings of 53 rd IEEE Conference on Decision and Control, 15–17 December, Los Angeles, USA, 2293–2298. Sahu BK and Subudhi B. (2014). Adaptive tracking control of an autonomous underwater vehicle. International Journal of Automation and Computing 11: 299–307. Singh Y, Sharma S, Sutton R and Hatton D. (2017). Path planning of an autonomous surface vehicle based on artificial potential fields in a real time marine environment. In: Proceedings of 16th International Conference on Computer and IT Applications in the Maritime Industries, 15–17 May, Cardiff, UK, 48–54. Singh Y, Sharma S, Sutton R, Hatton D and Khan A. (2018). A constrained A* approach towards optimal path planning for an unmanned surface vehicle in a maritime environment containing dynamic obstacles and ocean currents. Ocean Engineering 169: 187–201.
10
02-SUT332-37(1)-ut69521_Color.indd 10
3/20/2020 12:38:57 PM
Underwater Technology Vol. 37, No. 1, 2020
Sun B, Zhu D and Yang SX. (2014). A bioinspired filtered backstepping tracking control of 7000-m manned submarine vehicle. IEEE Transactions on Industrial Electronics 61: 3682–3693. Wu W-G, Chen H-T and Wang Y-J. (2001). Global trajectory tracking control of mobile robots. Acta Automatica Sinica 27: 326–331. Xiang X, Yu C and Zhang Q. (2017). Robust fuzzy 3D path following for autonomous underwater vehicle subject to uncertainties. Computers & Operations Research 4: 165–177.
Xu X, Grizzle JW, Tabuada P and Ames AD. (2018). Correctness guarantees for the composition of lane keeping and adaptive cruise control. IEEE Transactions on Automation Science and Engineering 15: 1216–1229. Yan Z, Wang M and Xu J. (2019). Robust adaptive sliding mode control of underactuated autonomous underwater vehicles with uncertain dynamics. Ocean Engineering 173: 802–809. Zhang L-J, Qi X and Pang Y-J. (2009). Adaptive output feedback control based on DRFNN for AUV. Ocean Engineering 36: 716–722.
11
02-SUT332-37(1)-ut69521_Color.indd 11
3/20/2020 12:38:57 PM
CALL FOR PAPERS Underwater Technology: InternaƟonal Journal of the Society for Underwater Technology The Society for Underwater Technology is calling for papers for its internaƟonal journal, Underwater Technology. The journal publishes peer-reviewed technical papers on all aspects and applicaƟons of underwater technology, including: • • • • • • • • • • • • •
diving technology and physiology environmental forces geology/geotechnics marine polluƟon marine renewable energies marine resources oceanography subsea systems underwater acousƟcs underwater roboƟcs underwater science underwater vehicle technologies salvage and decommissioning
Original papers on new technology, its development and applicaƟons, and papers covering new applicaƟons for exisƟng technology, are parƟcularly welcome. Submissions should adhere to the journal’s guidelines available at www.sut.org/publicaƟons/underwater-technology/guidelines-for-authors/ For more informaƟon or to make a submission, please contact the Assistant Editor, Elaine Azzopardi, at Elaine.Azzopardi@sut.org
02-SUT332-37(1)-ut69521_Color.indd 12
3/20/2020 12:38:57 PM
doi:10.3723/ut.37.013 Underwater Technology, Vol. 37, No. 1, pp. 13–27, 2020
Technical Paper
www.sut.org
Case studies in estimating subsea systems’ readiness level Sirous F. Yasseri* and Hamid Bahai Brunel University London, Kingston Lane, Uxbridge, UB8 3PH, UK Received 15 October 2018; Accepted 6 January 2020
Abstract Systems readiness level (SRL) is a metric defined for assessing progress in the development of systems. The methodologies to estimate SRLs are built on the technology readiness level (TRL), originally developed by NASA to assess the readiness of new technologies for insertion into a system. TRL was later adopted by governmental institutions and many industries, including the American Petroleum Institute (API). The TRL of each component is mathematically combined with another metric, integration readiness level (IRL), to estimate the overall level of readiness of a system. An averaging procedure is then used to estimate the composite level of systems readiness. The present paper builds on the previous paper by Yasseri (2013) and presents case examples to demonstrate the estimation of SRL using two approaches. The objective of the present paper is to show how the TRL, IRL, and SRL are combined mathematically. The performance of the methodology is also demonstrated in a parametric study by pushing the states of readiness to their extremes, namely very low and very high readiness. The present paper compares and contrasts the two major system readiness levels estimation methods: one proposed by Sauser et al. (2006) for defence acquisition based on NASA’s TRL scale, and another based on API’s TRL scale. The differences and similarities are demonstrated using a case study. Keywords: subsea production systems; technology readiness level; integration readiness level; system readiness level; system maturity
1. Introduction An understanding of technology readiness is critical in making decisions about the use of new and/or existing components in a new system (Olechwski et al., 2015). The most widely used tool for readiness assessment is NASA’s technology readiness level (TRL) scale. NASA introduced TRLs (Mankins * Contact author. Email address: Sirous.Yasseri@Brunel.ac.uk
1995; 2009) in the 1970s, and in 1995 published a refined 9-level scale, along with with with the first detailed descriptions of each level (Azizian et al., 2009). Presently, the TRL approach is used in multiple industries and serves a similar purpose. Commercial implementations of TRLs are similar to NASA’s nine-level scale. Fig 1 shows a generic ninelevel TRL definition. Similar to the NASA scale, this generic scale begins with a technology in its very basic scientific form, and progresses to a proven technology in its actual operating environment. TRL is a measure of an individual technology at a point in its development cycle, and not of a system’s readiness. TRL on its own cannot indicate a system’s readiness. TRL 1 through to TRL 8 on NASA’s scale focus on the design, development and testing aspects of a system. TRL 4 and TRL 5 concern verification (not the validation of components). TRL 9 focuses on the ‘operational’ aspects of the components, as it is integrated with the system. Thus, the qualification of a component for TRL 8 and TRL 9 must be performed within the context of the system that uses it. TRLs were not intended to address systems integration, i.e. assuring various components to work together perfectly in a system, nor to indicate that the technology will result in the successful development of a system (Gove, 2007). The wrong technology, or even the right technology which is improperly implemented, can be ineffective. The TRL scale is also used as an evaluation and planning tool to assess the readiness for the insertion of individual components into a system and to simplify the communications on the status of all components. A system comprises core technology components and their linkages in accordance with the system architecture. According to Henderson and Clark (1990), two types of knowledge are needed: component and architectural (i.e. knowledge of how the 13
03-SUT309-37(1)-ut69510_Color.indd 13
3/24/2020 1:48:42 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
Fig 1: A generic NASA-type TRL scale (Nolte et al., 2004)
Fig 2: API (2009) TRL scale
components are linked together). Henderson and Clark (1990) emphasise that systems often fail because attention is given to the technology, while knowledge of the linkages and integrations is not fully addressed. They conclude that while the TRL provides the metric for describing component knowledge, it is important to find a metric that provides a description of architectural knowledge and integration. Once the technology is sufficiently proven, it can be incorporated into a system and subsystem. Smith (2005) makes a distinction between readiness and maturity by noting that a system considered mature in one context, may not possess sufficient readiness for operation in a different environment. Bilbro (2007) used ‘maturity’ as part of the definition of ‘readiness’ and thereby implies a relationship between the two terms. However, some authors use these terms interchangeably (e.g. Azizian et al., 2009). Any ready technology is mature, but not all mature technology (in some systems) is ready for a different use. The present paper uses the term ‘readiness’ to include maturity and suitability for deployment; the terms system ‘readiness’ and system ‘maturity’ are used in the present paper to signify different things. The two metrics of integration readiness level (IRL) and system readiness level (SRL) are architecture-based extensions to the TRL introduced by Sauser et al. (2007; 2008). Yasseri (2013) extended the American Petroleum Institute’s (API; 2009) TRL definitions by defining IRL and SRL metrics suitable for use in subsea system developments with API’s TRL. API (2009) adapted NASA’s scale into a seven-stage scale (Fig 2; see Yasseri, 2013). API
(2009) distinguishes three development levels: concept validation (TRL 0 to 2), technology validation (TRL 3 to 5), and system validation (TRL 6 to 7). Similar to NASA’s TRL, API’s TRL concentrates on individual technology being developed to be integrated with other components/sub-systems in a broader subsea system. Fig 3 compares the API TRL definition with a NASA-type scale. This adaptation fulfills the needs of the sanctioning authority for a harmonised scale to monitor the state of progress in a major investment project. Acceptable technology maturity has often been the principal driver, particularly in subsea systems, where availability is fundamental to customer requirements. Generally, technology and system development should follow the same timeline (evolution), or maturation paths. A technology is inserted into a system based on its readiness, functionality and environmental readiness, and its ability to successfully interact with other components in the system. Many factors governing the development of a successful system are not always effectively implemented, but by considering IRL such oversights can be substantially reduced. API’s TRL levels 0 to 6 follow NASA’s TRL levels 1 to 7, and API’s level 7 combines NASA’s levels 8 and 9. The dependencies between modules/components and the dependency of the subsystem/system to its environment are not explicitly addressed by API’s TRL.
2. Domain mapping A system is an aggregation of pieces of equipment and enabling products (including software)
14
03-SUT309-37(1)-ut69510_Color.indd 14
3/24/2020 1:48:51 PM
Underwater Technology Vol. 37, No. 1, 2020
Fig 3: Comparison of NASA-type TRL scale with API (2009) TRL scale
to perform a mission. This arrangement of components and equipment (or subsystems) is the system’s architecture. In the context of an oil and gas production system, system components include flowlines, Christmas trees, blowout preventers, actuated valves and pumps, that together enable a subsea production system to achieve its purpose. A system diagram or map shows the basic logical architecture, or abstraction, of a system and enables its visualisation. A system map shows every piece of functional equipment that is required to perform a function and how they are linked together. There are several ways to map a system. For example, a schematic diagram of a chemical process uses symbols to represent the vessels, piping, valves, pumps, and other system equipment, indicating their interconnecting paths while omitting physical details. The schematic shows the intent and how the parts are supposed to interact with each other, i.e. the flow of fluid along its path. A schematic usually omits all details that are not relevant to understand interdependencies. In contrast, a construction drawing shows equipment as they are actually laid out and to scale, and can be used to build the system. Two popular methods of representation of a system functional architecture (or alternatively, its structural elements) are the block diagram and the design structure matrix (DSM).
2.1. Block diagram A block diagram is a representation of a system in which the principal, functions, parts, and equipment are represented by blocks connected by lines that show the relationships between the blocks. The block diagram is especially focused on the input and output of a system and does not consider the internal workings of the equipment. This principle is referred to as the black box in engineering, whereby the paths that get from input to output are unknown or left to be defined at a later stage. Fig 4 shows the block diagram of a simple system comprising nine components, grouped into three modules at an advanced stage of the functional architecting. There are interfaces between the components within each module and the different modules. Interfaces between the two components are shown by double-headed arrows, implying that the readiness of two components to be integrated is interdependent. Single-headed arrows are used to show the direction of flow, not interdependencies. 2.2. Design structure matrix (DSM) A DSM is a square matrix used to represent the relationships and dependencies between individual components of a system (Browning, 2016; Eppinger and Browning, 2012). The block diagram in Fig 4 is shown in Table 1 as a DSM. The network shown in 15
03-SUT309-37(1)-ut69510_Color.indd 15
3/24/2020 1:49:06 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
C
F
and assemblies or components (Yasseri, 2015a). This hierarchical representation avoids problems related to presenting extremely large matrices by shifting the focus to various levels in the hierarchy, enabling the analysis of a system at different levels and details. In general, the DSM analysis only considers relationships across components (at the same level), and not within components (done at the next level of the hierarchy).
G
3. Sauser et al.’s method for system readiness level estimation
E
D A
B
I
H
Fig 4: A directed graph network showing components of a system and their connectivity
Table 1 gives a view of the system’s functional components, and how they must interface to achieve the required mission. The components are shown as rows and columns in the matrix, and the components are listed in the same order along both axes. Component interactions are represented by an ‘x’; components A and F do not require input from the other components, whereas B, C, D, E, G, H and I receive input. By reading down a column, it is possible to see that a component provides input to others in its associated row. When reading across a row in the matrix, it is possible to see that a component in the row receives input from the component in the corresponding column. A subsea system can be organised into a hierarchical structure, e.g. subsystem, sub-sub systems, Table 1: DSM of the functional block diagram shown in Fig 4 A
B
C
D
E
F
G
H
I
A B
X
C D
X X
X
E
X
X
X
X
X
X
X
F G
X
H I
X
X
TRL provides an indication of the components’ readiness status. However, it is useful to have a metric that provides a description of the components integrated into the system, i.e. how components relate to each other and work together. It is important that all stakeholders have the same understanding when evaluating the integration readiness, or system readiness, and how TRL relates to IRL and SRL. SRL based on the component integration and interoperability is more relevant for identifying lagging or critical technology, especially if a substitute must be sought. The present section describes a concept originally proposed by Sauser et al. (2006) for the development of an SRL scale that incorporates the readiness level of all components of the entire system without exception. The original Sauser et al.’s (2006) definition of SRL is based on the NASA-type TRL scale. They introduced a new metric for integration, namely the integration readiness level (IRL), and proposed a mathematical method for combing TRL and IRL for estimating SRL. The resultant SRL scale can provide an assessment of the progress of overall system development and identify potential areas that require further work. Gove (2007) and Sauser et al. (2010) identified the requirements for IRL as: 1) Provide an integration-specific metric to determine the integration maturity between two or more configuration items, components and/or subsystems. 2) Provide a means to reduce the uncertainty involved in maturing and integrating new technology into a system. 3) Provide the ability to meet system requirements during the integration assessment, so as to reduce the integration of obsolete technology over less mature technology. 4) Provide a common platform for the maturity assessment of new system developments and new technology insertion. Based on these requirements, Sauser et al. (2010) proposed a 9-level IRL as described in Table 2.
16
03-SUT309-37(1)-ut69510_Color.indd 16
3/24/2020 1:49:20 PM
Underwater Technology Vol. 37, No. 1, 2020
Table 2: Definitions of TRL and IRL for the US Department of Defense (Sauser et al., 2010) Technology (components)
Actual system completed and qualified through test and demonstration System prototype demonstration in relevant environment System/subsystem model demonstration in relevant environment
6
Demonstration
8
Execute a support program that meets operational support performance requirements and sustains the system in the most cost-effective manner over its total life cycle
8
Actual integration completed and mission qualified through test and demonstration in the system environment The integration of technologies has been verified and validated with sufficient detail to be actionable The integrating technologies can accept, translate and structure information for its intended application
7 6
5
Component and/or breadboard validation in relevant environment
5
4
Component and/or breadboard validation in laboratory environment Analytical and experimental critical function and/or characteristic proof of concept
4
Combine
3 Research
3
2
Technology concept and/or application formulated
2
1
Basic principles observed and reported
1
The introduction of an IRL to the assessment process provides a means of checking the technology’s position on an integration readiness scale. Since both the technologies and integration elements are assessed using a numerical scale, it is possible to combine the IRL and TRL of all components (Yasseri 2016) and generate a composite metric for the overall system readiness (Sauser et al., 2008). The SRL matrix comprises one element for each of the constituent technologies and quantifies the readiness level of a specific technology with respect to each technology in the system. TRL and IRL values were normalised from the original 1 to 9 levels by dividing each element by 9. When no integration is present between two technologies, an IRL value of 0 is entered. For integrations to itself, a non-normalised IRL value of 9 or normalised value of 1 is used. The integration with itself was introduced (the diagonal line in the DSM matrix) to perform the matrix multiplication. TRL is defined as a vector with n entries, as shown in equation 1, where TRLi is the TRL of technology i:
Description
9
Integrate
Integration is mission proven through successful mission operations
IRL Demonstration
9
7
Interrogation (interfaces)
Description
Construct
TRL
There is sufficient control between the technologies necessary to establish, manage and terminate the integration There is sufficient detail in the quality and assurance of the integration between technologies There is compatibility between technologies to orderly and efficiently integrate and interact
There is some level of specificity to characterise the interaction between technologies through their interface An interface between technologies has been identified with sufficient detail to allow the characterisation of the relationship
⎡TRL L1 ⎤ ⎢ ⎥ ⎢TRL L2 ⎥ ⎥ [TRL ]n ,1 = ⎢⎢ ⎥ ⎢ ⎥ ⎢TRL ⎥ n⎦ ⎣
(1)
Matrix IRL illustrates how the technologies are integrated with each other from a system perspective. For a system with n technologies, IRL is defined in equation 2, where IRLij is the IRL between technologies i and j. The hypothetical integration of a technology i to itself is denoted by [TRL]ii:
[TRL ]n×n
⎡TRL L11 TRL L12 ⎢ ⎢TRL L21 TRL L22 = ⎢⎢ ⎢ ⎢TRL TRL n1 n2 ⎣
TRL L1n ⎤ ⎥ TRL L2 n ⎥ ⎥ ⎥⎥ TR RLnn ⎥⎦
(2)
In these matrices, the standard TRLs and IRLs corresponding to values 1 to 9 should be normalised. In any system, each of the constituent technologies is connected to a minimum of one other technology
17
03-SUT309-37(1)-ut69510_Color.indd 17
3/24/2020 1:49:26 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
through a bi-directional integration. The manner in which each technology is integrated with other technologies is used to formulate an equation for calculating SRL. This SRL equation comprises the TRL and IRL values of the technologies and the interactions that form the system. In order to calculate a value of the SRL from the TRL and IRL values, TRL and IRL matrices are normalised. An SRL matrix is obtained from the product of the TRL and IRL matrices, as shown in equation 3: [SRL]n × 1 = [IRL]n × n × [TRL]n × 1
(3)
The SRL matrix comprises one element for each of the constituent technologies and quantifies the readiness level of a specific technology with respect to each technology in the system; it also accounts for the development state of each technology through the TRL. For a system with n technologies, SRL is as shown in equation 4: ⎡ SRL L11TRL L1 + IRL L12TRL L2 + +IRL I L1nTRLn ⎤ RL1 ⎤ ⎡ IRL ⎥ ⎢ ⎥ ⎢ ⎢SRL L21TRL L1 + IRL L22TRL L2 + +IRL L2nTRLn ⎥ RL2 ⎥ ⎢ IRL ⎥, ⎥=⎢ R ] = ⎢⎢ [SRL ⎥ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢SRL ⎥ ⎢ IRL TRL L + IRL R TRL L + + IRL TRL R n1 1 n2 2 nn n⎦ n⎦ ⎣ ⎣
(4)
where IRLij = IRLji. These values will be within the interval (0 and n); hence, for each technology, i, its corresponding SRLi
is divided by ni (ni is the number of integrations of technology I; each technology is dictated by the system architecture, including its integration to itself, to obtain its normalised value between 0 and 1. The SRL for the complete system is the average of all normalised SRL values, as shown in equation 5:
SRL RLestimate
RL 2 SRL R n SRL RL1 SRL + + + n1 n2 nn = n
(5)
Equation 5 gives an SRL assuming there is no modularisation. If a system consists of m modules, then: SRL RLestimated =
SRL RL M 1
where SRL RL M 1 =
SRL RLComp 1
SRLM 2 + …+ SRL SRL RLMn , m SRL SRL LCo Cop op 2 +
(6)
+ SRL R Comp m k
k
, and k is the number of components in module 1. SRL estimations for the remaining modules are similar. The SRL metric can be used to determine the readiness of a system and its status within a developmental lifecycle. Table 3 presents an example of how the various levels of the SRL scale can correlate to an acquisition life cycle. Table 4 presents an SRL scale
Table 3: Banding of SRLestimated according to Sauser et al. (2010) SRL
Phase
Definitions
0.10 to 0.39 0.40 to 0.59
Concept refinement Technology development
0.60 to 0.79
System development and demonstration
0.70 to 0.89 0.90 to 1.00
Production Operations and support
Refine initial concept; develop system/technology strategy Reduce technology risks and determine an appropriate set of technologies to integrate into a full system Develop system capability or (increments thereof): reduce integration and manufacturing risk; ensure operational supportability; reduce logistics footprint; implement human systems integration; design for production; ensure affordability and protection of critical program information; and demonstrate system integration, interoperability, safety and utility Achieve operational capability that satisfies mission needs Execute a support program that meets operational support performance requirements and sustains the system in the most cost-effective manner over its total life cycle
Table 4: SRL scale according to the US Department of Defense (2011) integrated defense acquisition, technology, and logistics life cycle SRL
Definition
9 8 7 6 5 4 3 2 1
The system has achieved initial operational capability and can satisfy mission objectives System interoperability should have been demonstrated in an operational environment System threshold capability should have been demonstrated at operational performance level Whether the system component can be integrated and should have been validated System high-risk component technology development should have been complete; low-risk system components System performance specifications and constraints should have been defined and the baseline allocated System high-risk immature technologies should have been identified and prototyped System materiel solution should have been identified System alternative material solutions should have been considered
18
03-SUT309-37(1)-ut69510_Color.indd 18
3/24/2020 1:49:31 PM
Underwater Technology Vol. 37, No. 1, 2020
Fig 5: A simple system comprising ten components arranged into three modules. No integration with the environment is required
according to the US Department of Defense (2011). The results of equation 5 can be multiplied by 9 to de-normalise it, and the resulting figure can be used in conjunction with entries from Table 4. It is important to note that many military and space systems cannot be verified (levels 8 and 9) in their operational environment until deployed. Likewise, many systems are part of an evolutionary life cycle in which the final maturity will be verified once deployed or in the next generation.
4. Example case 1: SRL calculation according to the approach by Sauser et al. Fig 5 shows a simple system comprising ten technologies (components): A1 to A10. It is not necessary for this system to be integrated into its operational environment, but it should be able to operate within its operational environment. Demonstration of TRL 8 and 9 is not possible until the system is deployed; therefore, proof that the system operates may be done in a simulated environment (e.g. weapon systems) or not at all (e.g. aerospace vehicles). This system can be considered as a single ten-component system or a system which comprises ten modules. Alternatively, some technologies may be bundled together to create subsystems so that more vendors can be involved, or for the purpose of parallel manufacturing.
Table 5 shows a DSM for this ten-component system as shown in Fig 5. The diagonal terms are the maximum TRL in the scale (9 in NASA scale). The off-diagonal terms are the IRL of two interfacing components. Columns 1 to 14 of Table 5 are part of a larger table employed for all calculations. Columns 15 to 29 of the larger table are shown in Table 6. The TRL levels are values 1 to 9; the IRL values also range from 1 to 9. Before the matrix calculation is performed, these values are normalised by being divided by 9. For example, an IRL of 9 has a normalised value of 9/9 = 1, and an IRL of 5 has a normalised value of 5/9 or 0.556. The values of the IRLs are specified in Fig 5 and have been inserted in the matrix (Table 5, columns 5 to 14). Each of the components of a system is connected to at least one other component. The TRL of each component is entered in the third column of Table 5. It is assumed that the integration is bi-directional, namely, that the IRL is the same in each direction, and thus that the DSM is symmetric. Maturity is differentiated from readiness. For example, component 7 in Fig 5 has been in an operational system for some years, but when it is used for a new system, it is not considered ready for the new situation and enters into the new system at level 6, or level 7 on NASA scale. Attainment of levels 8 and 9 must be proven within the environment of the new system. With the exception of mature technologies, all new technologies in the new system mature along the same timeline, though some may be ahead of others. However, the interface between two components may drag down the technology’s readiness, which otherwise matures faster in other aspects. Thus, it is not possible for a technology to be at level 7, and its linkage to other components in the system at level 2 or 3. The large difference is an indication that the mature technology must enter at a lower TRL, commensurate with the readiness of its linkage. In conclusion, if the IRL is low, then the TRL must be revised downward. The product [IRL]10-by-10 × [TRL]10-by-1 yields a resultant 10 × 1 column matrix (column 26) using equation 4. In order to calculate a composite SRL from the component TRL and IRL values, an SRL matrix is generated from the product of the IRL and TRL matrices, as per equation 4. For example, for row 4 of column 26, 1 × 0.444 + 0.56 × 0.566 + 0.44 × 0.444 +… = 0.951. The rest of column 26 is populated similarly. Each component’s SRL is calculated by dividing the value in column 26 by the total number of integrating components (including integration with itself, column 27). Results are then entered in
19
03-SUT309-37(1)-ut69510_Color.indd 19
3/24/2020 1:49:37 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
Table 5: DSM for the system shown in Figure 5 2
3
2
3
TRL
4
M1
4
A1
9
5
4
5
A2
5
9
4
4
B1
4
9
4
5 6
M2
4 Components
1
5
6
7
8
9
10
11
12
13
14
C1
C2
C3
C4
Components A1
A2
B1
B2
B3
B4
7
5
B2
5
9
4
4
8
5
B3
4
4
9
4
9
5
B4
4
9
4
4
C1
9
5
5
11
5
C2
4
5
9
5
4
12
5
C3
4
5
5
9
13
4
C4
4
9
10
M3
Table 6: Columns 15 to 29 of calculations for example case 1 (continuation of Table 5; refer also to figure 5) 15
16
17
18
19
20
21
22
23
24
Normalised IRL
25
26
27
28
29
Normalised TRL
Sum of TRLc times IRLc
Number of interfacing components
Component SRL
Module SRL
0.333
1
2
3
4
5
6
7
8
9
10
1.00
0.56
0.44
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.444
0.951
3
0.317
0.56
1.00
0.00
0.00
0.44
0.00
0.00
0.00
0.00
0.00
0.556
1.049
3
0.350
0.44
0.00
1.00
0.44
0.00
0.00
0.00
0.00
0.00
0.00
0.444
0.889
3
0.296
0.00
0.00
0.56
1.00
0.44
0.44
0.00
0.00
0.00
0.00
0.556
1.296
4
0.324
0.00
0.44
0.00
0.44
1.00
0.00
0.00
0.44
0.00
0.00
0.556
1.296
4
0.324
0.00
0.00
0.00
0.44
0.00
1.00
0.00
0.00
0.44
0.00
0.556
1.049
3
0.350
0.00
0.00
0.00
0.00
0.00
0.00
1.00
0.56
0.56
0.00
0.444
1.062
3
0.354
0.00
0.00
0.00
0.00
0.44
0.00
0.56
1.00
0.56
0.44
0.556
1.556
5
0.311
0.00
0.00
0.00
0.00
0.00
0.44
0.56
0.56
1.00
0.00
0.556
1.358
4
0.340
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.44
0.00
1.00
0.444
0.691
2
0.346
Composite SRL based on individual components
column 28. For example, component C2 requires five total integrations: components B3, C1, C3 and C4 and integration with itself. Thus, the component SRL for C2 is 1.556/5 = 0.311. Table 7 shows SRL for all components. The composite SRL is the average of the component SRLs (equation 8). As with any calculation involving an average, the analyst needs to be aware of the potential to mask an SRL that is significantly lagging or leading the average, reiterating the
0.324
0.338
0.3311
Composite SRL based on modules
0.3315
Multiply by 9 to reserve normalisation
2.980
2.983
importance of assessing and monitoring the individual component SRLs. Some modules have reached level 5 status on the TRL scale, but a few do not. The intention is to achieve the requirements of TRL 5 and assemble the system for integration testing. The SRL index according to Sauser et al. (2006; 2010) is 0.3311 (based on components), which is used in conjunction with Table 3. Multiplying this number by 9 gives a de-normalised value of 2.980, which can be
20
03-SUT309-37(1)-ut69510_Color.indd 20
3/24/2020 1:49:44 PM
Underwater Technology Vol. 37, No. 1, 2020
used with Table 4. Column 28 shows the same type of calculation based on modules. The SRL using modules’ SRL is 0.3315, which is slightly higher. This difference is a result of the two-tier averaging of the module-based SRL. The component SRLs provide an indication of the readiness of the individual component and its associated integrations. Examination of the individual component SRL values relative to each other identifies those components that are lagging. Composite SRL values are translated to whole numbers consistent with TRL and IRL scaling for ease of interpretation. The SRL shown in this section resulted in a composite SRL of 0.3315. Using the SRL translation model of Table 3, this system is at the technology development stage. Alternatively, multiplying 0.3315 by 9 gives 2.98, which translates to an SRL of 3 in conjunction with Table 4. The SRL calculated in this example is a snapshot in time, thus it is critical to measure the system readiness at multiple points along the life cycle.
4.1. System readiness definition based on API TRL scale Subsea production systems (SPS) are becoming more complex owing to the requirements of high availability and minimal intervention for repair. Many new technologies are inserted to achieve these goals. The software is also increasingly used to control SPS, thus adding to the level of complexity. It is not surprising that many designers are continually searching for the cause of unexpected failures and unacceptable behaviour in systems meant to be ‘ready’ for operation. There is a need to assess and measure, with high confidence, a system readiness level during the development life cycle. The readiness of equipment for use is assessed on its own merit. The aim of this and the next section is to demonstrate the notions of readiness proposed by Yasseri (2013; 2016) and Yasseri et al. (2018a; 2018b) through case studies. Achievement of API’s TRL 4 is one of several pieces of evidence that is used in the decisionmaking process for committing to major capital
Table 7: IRL and SRL definition compatible with API’s TRL (Yasseri, 2013)
System validation
Phase TRL 7 6
Technology validation
5
4
3
Concept validation
2
1
0
Development stage
IRL
Field-proven production system System installed. Production system installed and tested
7
System tested. Production system interface tested Environment tested. Pre-production system environment tested
5
Prototype tested. System function, performance and reliability tested Validated concept. Experimental proof of concept using physical model tests
3
Demonstrated concept. Proof of concept as desk study or research and development experimentation Unproven concept. Basic research and development in progress
1
6
4
2
0
Development stage
SRL
Development stage
Integration is field-proven through successful operations Integration is completed and qualified through sufficient and rigorous testing in the marine environment
7
The integration has been verified and validated with sufficient detail for the system to be deployable There are sufficient details to assure interoperability between technologies necessary to establish, manage and assure the integration There is sufficient detail in the control and assurance of the integration between technologies to deliver the required functionality There is sufficient evidence of compatibility between technologies within the system. Namely, they will work together and can be integrated with ease There is some level of specificity to the system functionality to allow identification of linkage between technologies
5
Manufacturing and installation in progress
4
Detail design and final procurement
3
Front-end engineering. Sourcing of long lead items
2
Concept selection. An optimal concept has emerged
1
Concept refinement. Two or more competing concepts are being considered
The interface, i.e. the linkage, between technologies can be identified and characterised with sufficient clarity
0
Concept definition. Various ideas are being considered or discounted
6
Field proven operational system The system is installed and tested. Commissioning in progress
21
03-SUT309-37(1)-ut69510_Color.indd 21
3/24/2020 1:49:44 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
Table 8: Definitions of the environmental readiness scales compatible with API’s TRL scale Environmental readiness level achieved
Activity
Description
Focus
7 6
Technology is field proven The system is installed, tested and commissioned System test is complete Environmental testing is complete
3
Proving technology over time Qualification of the installed system System qualification testing Environmental qualification testing Prototype qualification testing
Operation Assuring the integration of the system and the environment Integration of the system Enhancing reliability by reducing uncertainties Real size testing
2 1
Concept validation Conception
The prototype is tested. Technology is robust and usable Technology is validated Technology is demonstrated
5. Example case 2: SRL calculations for SPS Fig 6 shows a system that is similar to that shown in Fig 5 but is arranged into three modules, each of which must be integrated with the environment. The numerical values of TRL and IRL are kept the same; Example 2 refers to the API scale, while the NASA scale was used in example 1. The three modules in Fig 6 require integration with the operational environment. Table 9 details SRL calculations. Columns 4 to 15 in Table 9 show an asymmetric DSM matrix resulting
Scale testing Experimenting
Module 1 A1 TRL IRL 4
investment (Yasseri, 2014). Thus, TRL 4 is critical for making the decision on whether to go forward with the investment. TRL 5 is the most important technical stage during the subsea development process. At API’s TRL 5 stage readiness of all necessary components must be demonstrated; it also involves a demonstration that the components work together as a system. Thus, achieving a TRL of 5 is a prerequisite for the integration and installation of assemblies. Validation at this level must go beyond discrete components and must consider testing the assembled components (or subsystems), testing at the quayside, and possibly in shallow water (i.e. a relevant environment) and/or the operational environment (Table 7). Whereas an aerospace vehicle should be operated in its environment, a subsea system must be integrated with its environment and operate within it. Thus, the environment readiness must be included in the SRL assessment. All equipment must be supported and secured on the seabed using one of several means, e.g. gravity base, suction piles or ordinary piles, and preparing the seabed for the reception. While the technology readiness of the seabed is not meaningful, the integration of equipment with the seabed requires attention. The present paper uses a seven-level scale to assess the environmental readiness level (see Table 8).
IRL 5
A2 TRL
IRL 4 ENV IRLenv 6
Module 2 B1 TRL
IRL 5
B2 TRL
IRL 4
B3 TRL
IRL 5 B4 TRL
IRL 4
5 4
Module 3 IRL 5
C3 TRL
ENV IRLenv 6
IRL 4
C1 TRL IRL 5
IRL 5
C2 TRL
IRL 4
C4 TRL
ENV IRLenv 6
Fig 6: A simple system of a subsea system that must be integrated with the seabed
from the assumption that two interfacing components could have different TRLs, but their integration readiness levels are the same and equal to the least ready component owing to mutual dependency. In general, the matrix does not need to be symmetric, as the symmetry assumption is not necessary for the application of the method. Column 15 is the environment readiness index (Table 9). Entries in column 16 in Table 9 with the heading ‘Average IRL’ is the arithmetic average of all IRLs in that row, determined by summing up the IRLs of all interfacing components across the row and dividing it by the number of interfaces. For example, for the first row (4 + 5)/2 = 4.5, 2 is the number of components to be integrated, not integration with itself. Column 17 (with the heading ‘TRL*Average IRL’) gives the results of multiplication of the component’s TRL and the average of its
22
03-SUT309-37(1)-ut69510_Color.indd 22
3/24/2020 1:49:44 PM
Underwater Technology Vol. 37, No. 1, 2020
Table 9: Calculation of SRL for the system of Fig 6 according to the method of Yasseri (2013; 2016)
4
M1
5 6 7
TRL
A1
4
A2
5
5
B1
4
4
A1
B2
5 5
10
B4
4
11
C1
4
12
C2
5
C3
5
C4
4
M3
13 14
7
8
9
10
11
12
13
14
15
C2
C3
C4
ENV
A2
B1
B2
B3
B4
C1
4.50
22.50
4.74
4.50
18.00
4.24
4.33
21.67
4.65
4.00
20.00
4.47
4
4.00
16.00
4.00
5
5.00
20.00
4.47
4.50
22.50
4.74
4.67
23.33
4.83
4.00
16.00
4.00
4
5
5 5
4
4
4
4
4 5 4
5 4
5
4
15
5
5 5
18
4.24
4
4
17
18.00
IRL matrix 5
16
4.50
Components
B3
M2
8 9
6
4
5
SRL =
IRLs; for example, in row 1 4 × 4.5 = 18. Column 18 gives the root of the mean of squares (RMS) for each component. For example, the square root of column 17 is 4.24 (noted in column 18), giving a composite component readiness index. The root of the mean of the sum of squares for the first module is: M 1−R =
18.0 0 + 22 5 + 18.0 = 4.41 3
(7)
The number 3 in the denominator is the number of components of the module M1. Alternatively:
(4 24) + (4.74) + (4 24) 2
M1−R =
2
2
3
= 4.41 (8)
An estimate of the system readiness level is given in Yasseri (2013): SRL R estimate =
(
)×(
) +( 2
)×( 3
) +( 2
)×(
)
2
= 3.45 5,
(9)
where 7 is the highest score for the environmental readiness scale (Table 8) and is used for the normalisation purpose, and 3 is the number of modules. Using a value of 3.45 in Table 1, the system must be at the assembly and installation stage; if the project schedule dictates a different level then reasons must be given for this. From a metrics point of view SRLest and SRL should measure the same things on the same scale.
19 Module SRL
3
5
SQRT
Modules
2
4
TRL*Average IRL
3
Average IRL
2
Components
1
4.41
4.38
4.51
3.75
However, SRL is defined (Table 7), while SRLest is derived by aggregation of attributes of all components using calculations. The estimate of system readiness reaches its highest level from below, as it measures the system readiness as a whole, and not its elements. If all components mature simultaneously along the same path, then SRLest approaches SRL. More inter-dependencies will increase the gap between SRLest and SRL if the IRLs are not at the same level as the TRLs. The component readiness level and system readiness level should be distinguished. For example, in the case of a component being investigated when it achieves TRL 6, although the investigation is in the context of the intended system in its environment, the focus is on the individual component, not the whole system. This index informs management when and where to intervene if the system readiness is lagging behind schedule. The entries in each row identify which components require closer management attention. A tightly controlled project ensures that TRL, IRL, and SRL closely follow each other. While there are differences between the present paper’s methodology and that of Sauser et al. (2006, 2008 and 2010) method can be applied to this example. Columns 1 to 14 would be the same as shown in Table 9; Table 10 shows the rest of the larger table. Sauser et al. (2006, 2008 and 2010) include integration with itself (see column 29 in Table 10 for the count of interfaces), and do not consider that the module should be integrated with its environment.
23
03-SUT309-37(1)-ut69510_Color.indd 23
3/24/2020 1:49:54 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
Table 10: Calculation of SRL for the system shown in Fig 6 using the method of Sauser et al (2010). Columns 1 to 14 are the same as given in Table 9. 16
17
18
19
20
21
22
23
24
C1
C2
C3
C4
Component SRL
28
Number of interfacing components
27
Sum of TRLc × IRLc
26
Normalised TRL
Components
25
1.00
0.71
0.57
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.571
1.408
3
0.469
0.71
1.00
0.00
0.00
0.57
0.00
0.00
0.00
0.00
0.00
0.714
1.531
3
0.510
0.57
0.00
1.00
0.57
0.00
0.00
0.00
0.00
0.00
0.00
0.571
1.306
3
0.435
0.00
0.00
0.71
1.00
0.57
0.57
0.00
0.00
0.00
0.00
0.714
1.939
4
0.485
0.00
0.57
0.00
0.57
1.00
0.00
0.00
0.57
0.00
0.00
0.714
1.939
4
0.485
0.00
0.00
0.00
0.57
0.00
1.00
0.00
0.00
0.57
0.00
0.714
1.531
3
0.510
0.00
0.00
0.00
0.00
0.00
0.00
1.00
0.71
0.71
0.00
0.571
1.592
3
0.531
0.00
0.00
0.00
0.00
0.57
0.00
0.71
1.00
0.71
0.57
0.714
2.367
5
0.473
0.00
0.00
0.00
0.00
0.00
0.57
0.71
0.71
1.00
0.00
0.714
2.041
4
0.510
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.57
0.00
1.00
0.571
0.980
2
0.490
A1
A2
B1
B2
B3
B4
Normalised IRL matrix
Composite SRL based on component
29 Module SRL
15
0.490
0.479
0.501
0.4899
Composite SRL based on component
Reverse normalisation
3.429
0.4899 3.429
Table 11: DSM and SRL assessment of the system shown in Fig 7
Modules
3
4
3
4
5
6
7
8
9
10
11
Components TRL
2
2
Components
1
A
B
C
D
E
F
G
IRL matrix
12
13
14
15
16
ENV
Average IRL
TRL* Average IRL
SQRT
Module IRL
A
4
A
4
4
4
4.00
16.00
4.00
B
5
4
B
4
5
4.33
21.67
4.65
C
4
4
C
5
4
4.33
17.33
4.16
8
D
5
4
5
D
4
4.33
21.67
4.65
4.37
9
E
5
4
E
4
5
4.33
21.67
4.65
F
4
4
4
F
4
4.00
16.00
4.00
10
M2
7
M1
5 6
11 12
6
G
5
4
5
5
4
G
6
4.50
22.50
4.74
4.47
SRL=
4.09
Table 10 gives an SRL index of 0.4899, and when multiplied by 7 this yields 3.429, which may be rounded up to 4. These results are not substantially different from the earlier values. This is not the case, however, when the environmental readiness is very low. There is a distinction between critical and necessary technology. In principle, every piece of equipment in a system is necessary, since if it is not needed it can be eliminated. However, only a few pieces of equipment may be critical, since there
may be no substitutes for them, and without them, the intended system will not perform. In subsea practice, no subsystem, assemblies (or large components) are excluded from the assessment; all are considered necessary. The level of detail is decided by the assessor(s), drawing on help from the subject expert. This suggests that the purpose of TRL in the subsea industry is to ensure the readiness of the components for insertion into the system.
24
03-SUT309-37(1)-ut69510_Color.indd 24
3/24/2020 1:50:00 PM
Underwater Technology Vol. 37, No. 1, 2020
A TRL=4
IRL=4 IRL=4 C TRL=5
IRL=4
IR
5 L=
IRL=5
B TRL=5
IRL=4 D TRL=4
IRL=4
IRL=4
IRL=5
IRLENV=6
Fig 8: SRL as a function of TRL and TRL/IRL E TRL=5
IRL=4
F TRL=4
G TRL=5
IRLENV=6
Fig 7: A system comprising two modules
6. Discussion In designating interfaces between two components, the definition of the interactions between these interfaces is important (Yasseri, 2015b). In addition to the transmission of forces, moments and displacements between modules, there is also the transmission of fluid and energy, and data exchange. There are three possible forms of information exchange: • Internal interaction: Referring to information exchanged within a single module. • External interaction: Comprising information exchanged between two or more modules. • Boundary interaction: Occurring at the boundaries of the model, interacting with entities outside the project such as vendors, installation contractors, etc. According to Pimmler and Eppinger (1994), there are four types of interactions when integrating two technologies: physical connection, material flow, information flow, and energy exchange. The interaction strengths are the level or degree of interaction between the components. The DSM used shows the absence or presence of interaction, and it is assumed the numerical value of IRL will also account for the importance and strength of the interaction; otherwise, another layer needs to be added to the calculation. All major subsea equipment must be laid on the seabed and secured, e.g. by foundation slabs, suction
piles or conventional piles. There may also be a need for seabed preparation. With the exception of the shore approach and shipping channels, pipelines are not buried; however, the seabed may require sweeping or trenching. The need for a subsea system to be integrated with its environment differentiates it from other systems, such as aerospace vehicles or weapons. The present paper differs from the method originally presented by Sauser et al. (2006) as it includes the integration with the environment in addition to the method of estimating the SRL. Since the purpose, context and approach of the present study, and that of Sauser et al. (2006, 2008 and 2010) differ, a qualitative comparison of the two approaches provides limited insight. A limited parametric study of the current proposal may be useful. The simple system shown in Fig 7 is used to demonstrate the behaviour of the current method when the ratio of TRL: IRL changes. This system consists of two modules, both of which must be integrated with the environment. SRL estimation of modules and the system shown in Fig 7 are shown in Table 11. These calculations show that the system has achieved SRL level 4, which is the minimum TRL defined for some component. The environmental readiness keeps back the SRL until it reaches IRL env = 7. Fig 8 shows when the TRL and IRLEnv are kept at the same level, but the TRL: IRL ratio is changed; it also shows when the IRL is lagging behind the TRL. For a well-managed project, the TRL: IRL ratio should be close to one since the readiness of a technology to be integrated with other technologies will affect its TRL.
7. Conclusions TRLs provide a common understanding of the status of a technology during its development life cycle, which can also act as a means of assessing and managing risk, and making decisions concerning funding and implementation of technology. As
25
03-SUT309-37(1)-ut69510_Color.indd 25
3/24/2020 1:50:00 PM
Yasseri and Bahai. Case studies in estimating subsea systems’ readiness level
with any management tool, there are certain limitations to its use. Assigning a TRL rank is not a quick task. Two new metrics, namely IRL and SRL, introduced in previous papers by Yasseri (2013; 2016) for the subsea oil and gas production systems, were explored using example cases. Integration readiness level (IRL) indicates the readiness of two interfacing components to be brought together. The system readiness level (SRL) is a measure of the readiness of the entire system to be deployed, and combines TRL with IRL into a system readiness metric. Comparing the estimated SRL and values in the SRL table indicates the level of system readiness. These three indices provide part of the information required for project sanction to allow a project to move through the gate to the next phase of development in a stage-gate process. Technology, integration and systems development follow similar evolution, or maturation, paths. The SRL methodology provides decision-makers with a snapshot of a system’s state of readiness for deployment, and quantifies the level of componentto-component integration during system development, thus helping to improve system performance. Implementation of the SRL methodology aids decision-makers in identifying programmatic and technical risk areas. The proposed subsea system assessment framework was then compared and contrasted with the SRL methodology originally proposed by Sauser et al. (2006) for defense acquisition.
References American Petroleum Institute (API). (2009). API recommended practice for subsea production system reliability and technical risk management, first edition. API RP17N. Available at: https://www.apiwebstore.org/publications/ item.cgi?a123e3d2-c704-4f25-b1d5-3ed7ff448800, last accessed <22 February 2020>. Azizian N, Sarkami S and Mazzuchi T. (2009). A comprehensive review and analysis of maturity assessment approaches for improved decision support to achieve efficient defence acquisition. In: Proceedings of the World Congress on Engineering and Computer Science (WCECS) 2009, vol. 2, 20–22 October, San Francisco, USA. Available at: http://www.iaeng.org/publication/ WCECS2009/WCECS2009_pp1150-1157.pdf, last accessed <22 February 2020>. Bilbro JW. (2007). A suite of tools for technology assessment. In: Proceedings of AFRL Technology Maturity Conference, 11–13 September, Virginia Beach, USA. Available at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10. 1.1.475.9653&rep=rep1&type=pdf, last accessed <22 February 2020>. Browning TR. (2016). Design structure matrix extensions and innovations: A survey and new opportunities. IEEE Transactions on Engineering Management 63: 1–26.
Eppinger SD and Browning TR. (2012). Design structure matrix methods and applications. Cambridge: MIT Press, 352 pp. Gove R. (2007). Development of an integration ontology for systems operational effectiveness. Hoboken: Stevens Institute of Technology, 198 pp. Henderson RM and Clark KB. (1990). Architectural innovation: The reconfiguration of existing product technologies and the failure of established firms. Administrative Science Quarterly 35: 9–30. Mankins JC. (1995). Technology readiness levels – A white paper. Available at: http://www.artemisinnovation.com/images/ TRL_White_Paper_2004-Edited.pdf. last accessed <22 February 2020>. Mankins JC. (2009). Technology readiness assessments: A retrospective. Acta Astronautica 65: 1216–1223. Nolte WL, Kennedy BM and Dziegiel RJJ. (2004). Technology readiness level calculator. White Paper: Air Force Research Laboratory. Available at: https://ndiastorage.blob.core. usgovcloudapi.net/ndia/2003/systems/nolte2.pdf, last accessed <22 February 2020>. Olechowski A, Eppinger SD and Joglekar N. (2015). Technology readiness levels at 40: A study of state-of-the-art use, challenges, and opportunities. In: Proceedings of the 2015 Portland International Conference on Management of Engineering and Technology (PICMET), 2–6 August, Portland, USA. Available at: https://doi. org/10.1109/PICMET.2015.7273196, last accessed <22 February 2020>. Pimmler T and Eppinger S. (1994). Integration analysis of product decompositions. In: Proceedings of the ASME Design Theory and Methodology Conference, Minneapolis, USA. Available at: http://web.mit.edu/eppinger/ www/pdf/Pimmler_DTM1994.pdf, last accessed <22 February 2020>. Sauser B, Verma D, Ramirez-Marquez J and Gove R. (2006). From TRL to SRL: The concept of systems readiness levels. In: Proceedings of the Conference on Systems Engineering Research, 7–8 April, Los Angeles, USA. Sauser B, Ramirez-Marquez JE, Henry D and DiMarzio D. (2008). A system maturity index for the systems engineering life cycle. International Journal of Industrial and Systems Engineering 3: 673–691. Sauser B, Gove R, Forbes E and Ramirez-Marquez JE. (2010). Integration maturity metrics: Development of an integration readiness level. Information Knowledge Systems Management 9: 17–46. Smith JD. (2005). An alternative to technology readiness levels for non-developmental item (NDI) software. In: Proceedings of the 38th Annual Hawaii International Conference on System Sciences. 6 Jan, Hawaii, USA. Available at: https://www.computer.org/csdl/proceedings/ hicss/2005/2268/09/22680315a.pdf, last accessed <22 February 2020>. US Department of Defense (2011). Technology readiness assessment (TRA) guidance. Available at: https://apps.dtic. mil/dtic/tr/fulltext/u2/a554900.pdf, last accessed <22 February 2020>. Yasseri S. (2013). Subsea system readiness level assessment. Underwater Technology 31: 77–92. Yasseri S. (2014). Application of systems engineering to subsea development. Underwater Technology 32: 93– 109. Yasseri S. (2015a). Evidence-based practice in subsea engineering. Underwater Technology 32: 231–244.
26
03-SUT309-37(1)-ut69510_Color.indd 26
3/24/2020 1:50:12 PM
Underwater Technology Vol. 37, No. 1, 2020
Yasseri S. (2015b). Interface management of subsea field development. Underwater Technology 33: 41–57. Yasseri S. (2016). A measure of subsea systems’ readiness level. Underwater Technology 33: 215–228. Yasseri S, Bahai H and Yasseri R. (2018a). A systems engineering framework for delivering reliable subsea equipment.
In: Proceedings of the 28th International Ocean and Polar Engineering Conference, 10–15 June, Sapporo, Japan. Yasseri S. Bahai H and Yasseri R. (2018b). Reliability assurance of subsea production systems: A systems engineering framework. International Journal of Coastal and Offshore Engineering 2: 1–19.
27
03-SUT309-37(1)-ut69510_Color.indd 27
3/24/2020 1:50:12 PM
Society for Underwater Technology
Educational Support Fund Sponsorship for Gifted Students in Marine Science, Technology and Engineering to meet industry’s critical shortage of suitably qualified entrants.
SUT sponsors UK and overseas students (studying in the UK and abroad) at undergraduate and MSc level who have an interest in marine science, technology and engineering. Students are supported who are studying subjects such as:
Offshore and Ocean Technology Subsea Engineering Oceanography Marine Biology Ship Science and Naval Architecture Meteorology and Oceanography The SUT annual awards are £2,000 per annum for an undergraduate, and £4,000 for a one-year postgraduate course. (Part-time postgraduate studies funding available.) As one of the largest non-governmental sources of sponsorship, the SUT has donated grants totaling almost half a million pounds to over 270 students since the launch of the fund in 1990.
For further information please contact us t +44 (0)20 3405 9035
e info@sut.org
or visit our website
www.sut.org
03-SUT309-37(1)-ut69510_Color.indd 28
3/24/2020 1:50:13 PM
doi:10.3723/ut.37.029 Underwater Technology, Vol. 37, No. 1, pp. 29–33, 2020
Technical Briefing
www.sut.org
A simple in situ labelling approach and adequate tools for photo and video quadrats used in underwater ecological studies Bernabé Moreno Laboratorio de Ecología Marina, Universidad Científica del Sur, Lima 15067, Peru Received 23 December 2019; Accepted 25 February 2020
Abstract Ecological studies use quadrats to gather qualitative (1/0) and quantitative (density and surface coverage) information in terrestrial and marine sciences. Depending on the spatiotemporal scale of the assessment, this could be a pilot or a monitoring survey. For monitoring surveys, it is necessary to develop a code for the quadrat itself (in situ labelling), for the digital file (ex situ codification), and ideally, for both. The design of the quadrat used for these studies must accomplish ergonomics through certain specifications such as: made of highly resistant material; negative-buoyant but lightweight; anticorrosive (specially for marine environments); able to stay positioned on seafloor habitat; and compatible with the in situ labelling technique. The present paper is a comparison of quadrats of different materials and widths, including the implementation of an in situ and ex situ codification technique. Recommendations are made after several test hours sampling with quadrats. Keywords: scientific diving, NaGISA, underwater imaging, benthic ecology
1. Introduction Quadrats are widely used in terrestrial (Adler et al., 2007) and marine ecological studies (Iken and Konar, 2003) as they enable collection of standardised data at locations; comparison between sites subjected to (dis)similar environmental settings; and construction of extensive time series through monitoring efforts. Quadrat analysis includes assessing quantitative values of species or functional groups, which are key for punctual evaluations and in particular long-term assessments. In marine subtidal ecosystems, when functionality is prioritised over species identification, non-destructive methodologies * Contact author. Email address: 8ernabemoreno@gmail.com
are used to obtain information in a way that avoids collection of the whole macrobenthic community within the area (Peirano et al., 2016; Balazy et al., 2018). Incorporating a less invasive approach into classic photo and video transects has enabled efficient assessments of marine ecosystems (e.g. Beijbom et al., 2015; Bryant et al., 2017). Subtidal sampling protocols particularly designed for long-term evaluations require the ability for replicates and transects to be tracked, for which some sort of codification is necessary. This labelling can be done a) in situ, by using the quadrat itself to show certain codes, and b) during the data download by renaming files (Fig 1); however, ideally both methods are used. It is important to ensure an adequate labelling system for postprocessing image data in software such as VidAna (Hedley, 2003) or CPCe (Kohler and Gill, 2006). There are several standardised quadrat designs (e.g. Cook et al., 2013); some have been developed for specific camera attachments that allow the quadrat to be placed directly onto the camera for more automatic framing (Van Rein et al., 2011; Beijbom et al., 2015; Ashton et al., 2017). The camera-quadrat combination is useful for exhaustive monitoring; however, it is not adequate for environmental conditions such as strong currents, high turbidity and varying seafloor topography. When using a photo or video quadrat it is important to maintain the linearity on the straight lines (Fig 2). Good buoyancy is required from the diver, and the quadrat should not move once positioned over the benthos. The weight effect (opposite to the buoyant force) must be considered in order to manufacture a quadrat that is resistant and heavy enough to remain in place, but light enough to transport without additional help (e.g. other diver
29
04-SUT336-37(1)-ut69504_Color.indd 29
3/24/2020 1:45:23 PM
Moreno. A simple in situ labelling approach and adequate tools for photo and video quadrats used in underwater ecological studies
Fig 1: Two different codification options proposed for photo and video quadrat files labelling. These may vary depending on the timescale of the assessment (punctual or long-term evaluation), the experimental design (e.g. depth variable: stratified random sampling or transects) and the preference of the researcher. (IPA: Project Islas Pachacamac-Asia; AB: Antarctic Benthos Project)
or a lift-bag). Simplicity of the method and sampling gear (i.e. photography equipment) is fundamental as it should not be time consuming, nor require a high degree of expertise.
2. Materials and methods In order to determine a standardised method, quadrats were manufactured that followed NaGISA protocol measurements (Iken and Konar, 2003). Specifications such as the material and its width were considered to obtain a quadrat of adequate mass with anticorrosive and resistance properties. Two materials were tested: 1) polyvinyl chloride (PVC) pipes (∅ = 70 mm) that were cut, jointed and pierced to permit water leakage, with negative buoyancy (Fig 2a), and 2) A2 stainless steel (A2-SS) bars (∅1 = 4.7 mm, ∅2 = 9.5 mm) that were bent into quadrats of different sizes for different measurements. Inner quadrats or joining bars were connected by welding if necessary. Initially a Sony α6000 mirrorless camera with a Sea&Sea YS-01 strobe light was used (for Fig 2a). After several attempts, the photographic rig was changed to a GoPro HERO6 Black edition action camera with two SOLA 2000 video lights connected via flex arms to a dual handle and tray. Prior to any fieldwork activity, time (preferably using 00:00 GMT) and date (yyyymmdd) were synchronised between camera, dive computer, logbooks and any additional electronic gear in order to keep a chronological sequence and facilitate record keeping. Greenwich Mean Time was used, as these tests included dives and surveys within the Warm Temperate Southeastern Pacific (WTSP-MP) and the Scotia Sea (SS-MP) Marine Provinces (sensu Spalding et al., 2007). Diving operations were conducted through either surface-supplied air (‘hookah’) or SCUBA diving, depending on logistics and project
planning. Two types of in situ labelling systems were tested in order to choose the most practical and effective method. The first label method was adopted from Kohler and Gill (2006), using a medium-size clipper with an acrylic cell on which the code is written with a pencil and easily erased afterwards (Fig 2b, white arrow). The second method was obtained with cable ties of different colours, whose meaning is attributed by the diver depending on the sampling site and the experimental design. These labelling methods were used in both a) photo quadrats (in Pachacamac and Asia Islands, Peru) and b) video quadrats within transects (in Mackellar Inlet, Antarctica). The upper-left angle of the 0.25 m2 quadrat (white arrows on Fig 2e, f, g) was used as reading indicator. Black cable ties on the horizontal bar indicated a) depth (each one representing +5 m) or b) transect number (Fig 2d), and cable ties on the vertical bar indicated a) replicates (white) or b) transect point (orange). Cable ties were tightened to the A2-SS bar that enabled manual slide but avoided unwanted movements. NaGISA incorporates two levels of target sampling with increasing difficulty: 1) non-destructive (photography or videography of the 1, 0.5 and 0.25 m2 quadrats) and 2) destructive (for identification of collected specimens in the smaller area; Iken and Konar, 2003). If collection was required, labelled (A−E) collection bags (mesh size 500 μm) were placed inside the previous replicate bag (E < D < C < B < A) before entering the water in order to match the macrobenthic replicate samples with the corresponding photo quadrats. At the surface, a preliminary photograph was taken of a plate referring to the sampling station. The survey began with the deepest stratum (i.e. 15 m, three black cable ties towards the indicator angle) and the first replicate (i.e. one white cable tie towards the indicator angle). Quadrats were correctly placed according to NaGISA protocol. Photo and video graphs were taken of all sampling units (considering the visibility), and densities (abundance, estimated cover percentages) were annotated in acrylic slates. Macrobenthic organisms were then collected using the most accessible collection bag (i.e. replicate bag A > E). Cable ties were slid accordingly before changing to the next replicate and/or stratum. These steps were followed at different depths within stations. Once at surface, data was downloaded, renamed, arranged and doublecopied after every daily survey.
3. Results and discussion Features of the manufactured and tested sampling units are presented in Table 1. Despite achieving
30
04-SUT336-37(1)-ut69504_Color.indd 30
3/24/2020 1:45:35 PM
Underwater Technology Vol. 37, No. 1, 2020
Fig 2: Quadrats of different materials using different labelling systems for benthic assessments in various sites within the Warm Temperate Southeastern Pacific Marine Province (WTSP-MP), except for d) in maritime Antarctica, Scotia Sea Marine Province (SS-MP): a) 0.25 m2 PVC quadrat without in situ label; b) stainless steel (A2-SS) quadrat (0.5 m2 with an embedded 0.25 m2) where the light reflection on the acrylic cell (white arrow) avoids a proper in situ labelling; c) effective placement of the 0.5 m2 quadrat in a vertical wall (white arrows indicating anchorage points) enables a proper photography and density annotation; d) a dense aggregation of the Antarctic phaeophyte Adenocystis utricularis on boulders within a very shallow subtidal video-transect, cable ties were used to indicate transect-point (orange) within transect #1 (black) at Mackellar Inlet, King George Island, South Shetland Archipelago; e) a photo quadrat from the Pachacamac Island sampling station #5 (‘P05’ within the code below) showing 100 random points generated in the CPCe software; f) annotation with strong currents within a boulder environment of the photo quadrat in e); g) deepest stratum (15 m) of the station #2 in Pachacamac Island showing five echinoids Caenocentrotus gibbosus and large coverage of crustose coralline algae; h) sampling of the 0.25 m2 quadrat maintaining a peak performance buoyancy on station #1, Asia Island. Focal length f = 3 mm for all photographs, except in a) f = 16 mm.
31
04-SUT336-37(1)-ut69504_Color.indd 31
3/24/2020 1:45:37 PM
Moreno. A simple in situ labelling approach and adequate tools for photo and video quadrats used in underwater ecological studies
Table 1: Detailed features of the sampling units manufactured with different materials and widths (PVC: polyvinyl chloride; A2-SS: A2 stainless steel) Sampling unit
Material/density Total length Pipe/bar d (g cm−3) required (cm) diameter (∅) (mm)
Mass in air m (g)
Manufacture cost (GBP/PEN)∗
0.5 m2 w/internal 0.25 m2 quadrats 1 m2 quadrat with 1 m middle bar
PVC/ 1.38 A2-SS/ 8 A2-SS/ 8 A2-SS/ 8
800 1700 1100 3200
5/20 35/150 40/170 70/300
250 250 500 500
70 9.5 4.7 9.5
Sampling units were manufactured in Lima, Peru. PEN: Peruvian Nuevo Sol and its equivalent in GBP: British Pound Sterling
the main aim of surveying a specified area, PVC quadrats did not always remain still over the benthos as they did not have the required negative buoyancy, especially under strong currents. Additionally, the quadrats were not resistant enough as they seemed to be damaged after minor use. PVC quadrats have been widely used for subtidal studies (e.g. Kohler and Gill, 2006; Balazy et al., 2014) owing to cost (Table 1), accessibility and ability to be easily gridded with a cord (Marine Biodiversity Observation Network, 2019). As the PVC pipe diameter was the thickest of the materials (∅ = 70 mm) used in the present study, shadows were generated which affected the quality of the photo quadrat. The shadows caused issues in the post-processing as several plotted points in the CPCe software fell into shadowed surfaces, skewing the image analysis. However, when experimenting with different material (A2-SS bar) and width (∅1 = 4.7 mm) with the 1 m2 quadrat, shadows were unnoticeable, the quadrat remained still over the benthos and photo quadrats were effectively achieved. Nonetheless, given the length and the flexibility acquired with the width of the bar, uncomfortable vibrations occurred, and the angles were slightly bent while diving through strong currents. The A2-SS 0.5 m2 quadrat (∅2 = 9.5 mm) had a mass of 1.7 kg, and the 1m2 quadrat had a mass of 3.2 kg (Table 1). Both were reasonable weights to transport with bare hands, easily attach, or clip to a harness or lift-bag if necessary. Both quadrats were deployed in different substrate types (rugosity and inclination). These highly resistant, anticorrosive and negative-buoyant quadrats enabled a good framing over the transect line. Their stability on vertical walls were also tested, and they remained stationary for the duration of photography and annotation through anchoring to the substrate or attached macrofauna (Fig 2c, white arrows). Stainless steel has been effectively used in photo quadrats with ‘framers’ (0.35 m2; Beijbom et al., 2015) and gridded to generate a high-definition mosaic image by stitching the 25 sub-cells (0.5 m2; Cook et al., 2013). The utilisation of (non)framed
or (non)gridded photo quadrats will depend on the interest of the present study and the protocol that is being followed. For example, the arising application of artificial neural network (i.e. machine- and deep-learning tools) for automated classification of benthic groups (Beijbom et al., 2015; Ashton et al., 2017) would require an uninterrupted photo quadrat prior to analysis. Regarding the in situ labelling, clipped acrylic cells (Kohler and Gill, 2006) were ineffective as codes had to be erased and changed accordingly, which took an unnecessary amount of time. Additionally, the reflection off the white acrylic prevented the code visualisation, and owing to the cell size, a significant area of the 0.25m2 quadrat was obscured (Fig 2b, white arrow). This in situ labelling method was therefore discarded at the first stages of the research. In contrast, the cable tie method proved to be a very simple but highly effective approach for including relevant information within the photo and video quadrat. During the surveys, multiple dives were carried out together with divers who did not have scientific backgrounds. These non-scientific divers gave positive feedback on the technique, as they managed to set the quadrats correctly and slide the cable ties without problems. The training of parties in sampling techniques was also accomplished, thereby meeting the guidelines and goals of the project. The second photographic rig (1.1 kg on air) was easy to manoeuvre, with the flex arms facilitating the positioning and angulation of the video lights. Three intensities (2000 lumens maximum) were easily changed with the magnetic switch, depending on light requirement of the site. Data download, logbook digitalisation and electronic charge took place once at surface after finishing the daily sampling as programmed. Photo and video quadrats files were renamed and ordered in folders and subfolders following the proposed codification in Fig 1. At any depth, constant graphic register was highly recommended, especially if seasonal or infrequent records, or rare or unregistered animal behaviours were observed. This is achievable when ‘QuikCapture’
32
04-SUT336-37(1)-ut69504_Color.indd 32
3/24/2020 1:47:01 PM
Underwater Technology Vol. 37, No. 1, 2020
mode is activated on the camera, enabling instant recording when pressing the ‘on’ button. Ideally, wide FOV (field of view), and 1920 1080 pixels resolution for photographs, and 1080p–60fps (frames per second) for videos, is recommended. Labelling images with clear, self-interpretative, repetitive codification is achievable with the proposed approach. For successful labelling prior criteria needs to be defined before the sampling takes place. This may vary depending on research aims, researchers and individual preferences, but this must be specified in the metadata of each project. Recommendations from the present work are delivered through material, width and procedures with quadrats; in situ and ex situ labelling and codification techniques; and specifications of the photography rig and its operating modes. Maintaining a simple approach at each stage will enable the development and standardisation of an engaging extended network for benthic ecological assessments.
Acknowledgments The present work was carried out during fieldwork of two projects by the Universidad Científica del Sur. The first included the subtidal component GICS (Grupo de Investigación de Comunidades Submareales) from the project IPA (Pachacamac & Asia Islands), part of the GEF project (ID 4505) Strengthening Sustainable Management of the Guano Islands, Islets and Capes National Reserve System (RNSIIPG), in collaboration with the Peruvian Trust Fund for National Parks and Protected Areas (PROFONANPE) and the National Service of Protected Areas (SERNANP). The second included the hard-bottom component of the project Antarctic Benthos (Factores ambientales que rigen sobre la distribución del macrobentos en Bahía Almirantazgo y Ensenada Mackellar) with the Dirección de Asuntos Antárticos (Ministerio de Relaciones Exteriores del Perú). The author thanks the IPA, the Antarctic Benthos scientific teams and the staff from Naylamp Diving Dive Center for the surface and underwater assistance during the surveys and tests. Geanpierre Guzmán Urteaga is acknowledged as photographer of the image in Fig 2f. The author is grateful to Terri Souster and Báslavi Cóndor-Luján for revising and improving the manuscript with constructive comments, and Aldo Indacochea for feedback and support through the years.
References Adler PB, Tyburczy WR and Lauenroth WK. (2007). Longterm mapped quadrats from Kansas prairie: Demographic information for herbaceous plants. Ecology 88: 2673–2673. Ashton GV, Morley SA, Barnes DKA, Clark MS and Peck LS. (2017). Warming by 1°C drives species and assemblage level responses in Antarctica’s marine shallows. Current Biology 27: 2698–2705. Balazy P, Kuklinski P and Wlodarska-Kowalczuk M. (2014). Scientific diving in polar regions – the example of ecological studies at the Institute of Oceanology, Polish Academy of Sciences. Polish Hyperbaric Research 46: 65–84 Balazy P, Kuklinski P and Berge J. (2018). Diver deployed autonomous time-lapse camera systems for ecological studies. Journal of Marine Engineering & Technology 17: 137–142. Beijbom O, Edmunds PJ, Roelfsema C, Smith J, Kline DI, Neal BP, Dunlap MJ, Moriarty V, Fan T-Y, Tan C-J, Chan S, Treibitz T, Gamst A, Mitchell GB and Kriegman D. (2015). Towards automated annotation of benthic survey images: Variability of human experts and operational modes of automation. PLoS ONE 10: e0130312. Bryant DEP, Rodriguez-Ramirez A, Phinn S, GonzálezRivero M, Brown KT, Neal BP, Hoegh‐Guldberg O and Dove S. (2017). Comparison of two photographic methodologies for collecting and analyzing the condition of coral reef ecosystems. Ecosphere 8: e01971. Cook R, Fariñas-Franco JM, Gell FR, Holt RHF, Holt T, Lindenbaum C, Porter JS, Seed R, Skates LR, Stringell TB and Sanderson WG. (2013). The substantial first impact of bottom fishing on rare biodiversity hotspots: a dilemma for evidence-based conservation. PLoS ONE 8: e69904. Hedley J. (2003). VidAna 1.0 (Version 1.0). University of Exeter, UK. Iken K and Konar B. (2003). Natural Geography in Nearshore Areas (NaGISA): The nearshore component of the Census of Marine Life. Gayana 67: 153–160. Kohler KE and Gill SM. (2006). Coral Point Count with Excel extensions (CPCe): A visual basic program for the determination of coral and substrate coverage using random point count methodology. Computers & Geosciences 32: 1259–1269. Marine Biodiversity Observation Network. (2019). Sampling protocol for assessment of marine diversity on rocky shores. Available at: https://marinebon.org/p2p/ protocols/MBON_P2P_RS_Protocol.pdf, last accessed 22 February 2020. Peirano A, Bordone A, Marini S, Piazza P and Schiaparelli S. (2016). A simple time-lapse apparatus for monitoring macrozoobenthos activity in Antarctica. Antarctic Science 28: 473–474. Spalding MD, Fox HE, Allen GR, Davidson N, Ferdaña ZA, Finlayson M, Halpern BS, Jorge MA, Lombana A, Lourie SA, Martin KD, McManus E, Molnar J, Recchia CA and Robertson J. (2007). Marine ecoregions of the world: a bioregionalization of coastal and shelf areas. BioScience 57: 573–583. Van Rein H, Schoeman DS, Brown CJ Quinn R and Breen J. (2011). Development of benthic monitoring methods using photoquadrats and scuba on heterogeneous hardsubstrata: a boulder-slope community case study. Aquatic Conservation: Marine and Freshwater Ecosystems 21: 676–689.
33
04-SUT336-37(1)-ut69504_Color.indd 33
3/24/2020 1:47:01 PM
Society for Underwater Technology International multidisciplinary learned society This non-aligned membership-based organisation seeks to further the dissemination of knowledge and lessons learned in the underwater environment through networking, events and publications
Its membership covers the following activity areas: defence diving and manned submersibles environmental forces marine policy marine renewable energies ocean resources offshore site investigation and geotechnics salvage and decommissioning
For further information For events, membership, publications or general enquires, contact: e info@sut.org e events@sut.org
subsea engineering and operations underwater robotics underwater science underwater vehicles
04-SUT336-37(1)-ut69504_Color.indd 34
www.sut.org
3/24/2020 1:47:01 PM
Image-Based Damage Assessment for Underwater Inspections Michael Oâ&#x20AC;&#x2122;Byrne, Bidisha Ghosh, Franck Schoefs and Vikram Pakrashi Published by CRC Press, Taylor & Francis Group
Hardcover edition, 2018 ISBN 978-1-138-03186-9 230 pages In their introduction to this book, the authors highlight that any structure built or placed in the marine environment is prone to degradation over time and must, therefore, be inspected to ensure it remains functional, safe, and economically viable. This book concentrates on image-based inspection techniques and gives specific attention to the complexities of conducting such work under water. The point is made that basing structural inspection predominantly on image processing is still a young field. While its use is highly versatile and applied in many ways, the authors stress that given the type and significance of the decisions being made based on the surveys, there is an important requirement to ensure that the information being generated on the condition of the structures, using image-based approaches, is accurate. The book is divided into four main sections. The first section
places image-based surveys into the wider context of structural inspection. This section provides an overview of the recent history of underwater inspections and reviews the range of operational tools that are currently used to detect and diagnose damage. These methods include visual inspection, non-destructive testing, and conventional photography above and below water. For the latter method, the authors highlight that although images are taken in surveys routinely, they are used mostly as illustrations in reports, and not in imageprocessing techniques. This provides the introduction for the next section of the book. The second section examines the methods for acquiring the images in ways that make the results from any subsequent processing accurate and suitable for quantitative analysis. The authors emphasise the need for and importance of adopting methodical approaches for image collection, as this provides the basis for precise analyses. Much of this part of the book provides basic information on the principles of cameras, their settings and how to obtain good imagery in challenging conditions. Whereas experienced underwater photographers will know a lot of what is explained here, it is always worth being reminded of some of the competing processes that must be considered when generating images for quantitative analysis, rather than just their aesthetic quality. The authors rightly conclude that there is no point in inspectors basing surveys on image-processing techniques if the quality and consistency of the initial imagery is not high enough. The use of high-quality camera equipment
www.sut.org
Book Review
doi:10.3723/ut.37.035 Underwater Technology, Vol. 37, No. 1, pp. 35â&#x20AC;&#x201C;36, 2020
may be essential but the outputs will only be as good as the proficiency of the operator. The third section provides the focus of the book and examines a range of methods that analyse and interpret the imagery. The first chapter of this section describes many of the pre-processing techniques that can be employed to enhance, transform and/or calibrate the images. The reader is supported in this with a series of MATLAB algorithms to use. However, it is not explained why the authors use MATLAB code rather than employ the batchprocessing methods available in many of the proprietary photoediting programmes. The next chapter examines crack detection within the context that cracks and cracking are key factors in assessing the condition of a structure. An essential element here is how image processing that automatically quantifies the number and size of cracks can improve inspections and possibly make them more cost effective. Practical guidance on using information from the images to generate information on the physical properties of cracks is provided. Again, the reader is helped by the inclusion of relevant MATLAB procedures. This leads on to a chapter that examines image-based methods for detecting and measuring damage to the surface of structures. This is different from cracking as it relates to anomalies in surfaces rather than demarcated cracking. Being able to measure and quantify the area and volume of the damage permits temporal re-assessment that, in time, will support estimates of the rates of degradation. The detection techniques described are either
35
05-SUT347-37(1)-ut69506_Color.indd 35
3/20/2020 1:00:49 PM
Michael Oâ&#x20AC;&#x2122;Byrne et al. Image-Based Damage Assessment for Underwater Inspections
colour- or texture-based; there is a section that compares the two approaches with the overall conclusion that colour-based methods can be applied for a wider range of damage forms. The authors summarise this chapter by stating that the use of image-based techniques for assessing and measuring surface damage essentially makes them tools for ongoing non-destructive testing. There are elements of analysing the size and depth of cracks, and the area and volume of surface damage that have an obvious three-dimensional component. The authors acknowledge that 3D techniques greatly enhance the levels of information that can be obtained from inspection imagery, and the next chapter examines 3D imaging to follow on from this. Given the recent widespread adoption of structure from motion (SfM) photogrammetry for 3D image-based modelling, it is surprising that the authors quickly dismiss it based on some relatively weak criteria. Instead, the chapter concentrates on stereoimaging techniques and discusses the related aspects of calibration, rectification, correspondence, triangulation and surface reconstruction.
The following chapter addresses the possible need of engineers and researchers to have access to a database of common images from which to experiment with different image-processing methods to assess the best approaches for their own specific requirements. The chapter is based on the underwater lighting and turbidity image repository (ULTIR) and provides several examples of how to best use the repository for developing data-driven algorithms and evaluating the performance of the image-based methods described previously. Although described as open access, no images were available on the associated ULTIR website when tested at the time of this review. The final section explores the future applications in this emerging field. Examples include integrating image data with specialist engineering software, the application of virtual reality techniques, the use of spherical (360°) cameras, and applications for smartphone cameras. The section also examines the potential for applying deep learning techniques for automating some inspection roles, such as damage detection. Using smart technologies is seen as a central component for advancing
the field, as is the use of multicamera systems. The point is frequently made throughout the book that this field is developing rapidly and so it is not surprising that several techniques and technologies that were considered futuristic when it was written have already become mainstream. For a subject that is attracting a great deal of interest and generating exciting advances in several underwater industries, it was disappointing that overall the literature cited in many of the chapters was outdated. Most examples of editing and processing the image data used MATLAB scripts with little discussion of, or comparison with, other methods of photo-processing. However, these are minor criticisms of a book that is long overdue. It contains very appropriate and useful examples that enforce the need for good quality imagery within well-structured surveys for valuable subsequent analyses. This is a very informative book covering a wide range of topics, and is recommended for both engineers and researchers looking to discover more about the application of underwater image analysis. (Reviewed by Dr Martin Sayer, Tritonia Scientific Ltd)
36
05-SUT347-37(1)-ut69506_Color.indd 36
3/20/2020 1:00:50 PM
UT2 and UT3 The magazines of the Society for Underwater Technology
UT2 Issue O
UT2 Issue O
ne 2018
A Towed Vehic les CaSubsea Expo Round-up Un
1
A Ca Un THE
MAG
AZIN
E OF THE
SOCI
ETY
FOR
UNDE
UT2 Iss ue 1 Ma rch 2018
Decem
ber 20 15 Janu ary
ne 2016
OSIG Vess el Batteries Underwat er Vehicles
RWAT
1
ue 2OAp LOGY ril June 2017
Issue Tw
The magazine is represented at all the many exhibitions around the world at which the Society both co-organises and attends.
EUT R TE CHN 2 Iss
UT 2016
UT2 covers a focused range of underwater subjects including offshore, marine renewables, subsea engineering, ocean resources, diving and manned submersibles, underwater science and robotics.
Furthermore, the magazine is distributed at the many subsea training courses that are organised by the Society, ensuring it reaches tomorrowâ&#x20AC;&#x2122;s engineers and technologists.
o 2018
UT3 is the online magazine of the Society for Underwater Technology, and covers the subsea industry.
A Ca Un THE
MAG
AZIN
E OF THE
SUT OCI
1
ETY F 3 Au g OR U NDER Sept 2015W A
Subsea Mi ning Underwat er Vehicles
TER
TECH
NOLO
GY
AUVs Cables Unmanne d
Vehicle
1
UT2 Iss ue 2 Ma y 2018
It consists of the content of the print magazine UT2, greatly expanded with other information.
UT22 and UT33 are available online at http://issuu.com/ut-2_publication http://issuu.com/ut 2_publication www.sut.org 05-SUT-37(1)-IBC.indd 1
3/17/2020 7:12:11 PM
E EWDAT
NW O SH
1-3 DEC 2020,LONDON EXCEL
S
THE LATEST SOLUTIONS FOR YOUR BUSINESS NEEDS:
Register to attend for FREE oceanologyinternational.com
OFFSHORE ENERGY DEVELOPMENT
ASSET INTEGRITY & MONITORING
HYDROGRAPHY, GEOPHYSICS AND GEOTECHNICS
COASTAL ZONE & SHALLOW WATER
NAVIGATION AND POSITIONING
IMAGING AND METROLOGY
OCEAN OBSERVATION AND SENSING
MARINE POLLUTION AND ENVIRONMENTAL STRESSORS
DATA INTERPRETATION AND AI
UNMANNED VEHICLES AND VESSELS
CONNECTING THE GLOBAL OCEAN TECHNOLOGY COMMUNITY DISCOVER
CONNECT
SOURCE
OCEAN ICT
LEARN
FEATURING
Organised by:
06-SUT-37(1)-OBC.indd 1
In partnership with:
Endorsing associations:
3/17/2020 7:11:00 PM