A Review of Risk Analysis Research for the Operations of Autonomous Underwater Vehicles
Introduction
Autonomous underwater vehicles (AUVs) are effective platforms for navigating underwater or under ice to provide automated measurements without human intervention [1, 2]. The high level of autonomy of AUVs makes them an ideal tool for multiple data-gathering applications in scientific [3], [4], [5], commercial [6], military [7], and geopolitical areas [8]. In recent research, AUVs are increasingly deployed in harsh environments such as under sea ice or ice shelves in the Antarctic [4, 9, 10] and the Arctic regions [3, 11, 12]. Operating in such extreme conditions, including thick ice cover, permafrost, fragile material integrity, unpredictable climatic changes, and poor visibility, will inevitably pose a higher risk to both the physical vehicle and the onsite AUV supervisors compared to open water missions [13]. Hence, it is essential to conduct effective risk analysis before a mission to ensure the safe deployment of AUVs.
Table 1, which is adapted from a former study [14], summarizes potential accident types of AUV operations and their severity according to the level of damage to the vehicle itself, where AUV loss could be regarded as the most severe accident. AUV loss usually refers to the complete loss of the physical vehicle or an AUV being damaged and unrepairable for future missions. It is not only financially costly due to the higher insurance premium and acquisition costs of the vehicle [15]. Furthermore, it may also cause time delays or even the termination of research projects, lead to the loss of valuable gathered data, and potentially harm fragile polar environments [16, 17].
Over the years, there have been a number of formally reported accidents of AUV losses during deployment, as shown in Fig. 1. For example, the AUV Autosub2 was lost under the Fimbulisen ice shelf in Antarctica in February 2005. A formal accident inquiry concluded that this accident was equally likely to have been caused by an abort command or a loss of power. These technical failures was most likely introduced during the manufacturing and assembly phases [18]. Another lost vehicle, SeaBED, which was designed to scan the seafloor below overhanging sea ice, became trapped under the Antarctic ice during a mission and was almost crushed by an iceberg before it was rescued [19]. The Autonomous Benthic Explorer (ABE) was lost in March 2010, during its 222nd research dive off the coast of Chile. Researchers believed that the loss of the ABE was also caused by a technical failure. More specifically, the ABE may have suffered a catastrophic implosion of a glass sphere used for providing buoyancy, causing instant destruction of the on-board systems. Consequently, the ABE failed to send fail-safe commands for helping itself float to the surface for recovery [20]. An underwater glider, Seaglider SG522, lost communication in the Antarctic in February 2012 after having completed 156 dives. The inquiry panel identified that the root cause was an erroneous command, which resulted in this glider continuously diving and eventually being lost [21]. In April 2014, the Autosub Long Range AUV lost communication during a mission near the Irish coast. Luckily, it re-transmitted its position signal and was recovered after three months. More recently, a Hugin AUV was lost during its first under ice mission in the Antarctic in January 2019, and it was recovered four days later. Pre-dive checks had been reviewed for this vehicle without any irregularities. Technicians believed the vehicle was trapped below an ice floe, causing the Iridium signal for the AUV position failing to be received [22].
From the overview of historical accidents of AUV loss, it can be observed that the potential causes of historical accidents show a wide variety, which confirms the unpredictable and uncertain features of AUV related accidents. This non-uniform accidental pattern and relatively severe consequences imply the vulnerability of AUV operations and reinforce the necessity of implementing effective risk analysis before an AUV mission.
Risk analysis is a proactive approach for hazard identification, consequence analysis, and risk estimation for potential accidents [23]. There is a long history of the development of risk analysis techniques that have been applied in multiple fields, including nuclear power, chemical process, aerospace, and offshore oil and gas industries [24], [25], [26], [27]. Currently, with the booming development of the maritime industry, applications of risk analysis methods are also stimulated in this area [28], [29], [30], [31], [32]. Since marine systems are becoming more autonomous, using the AUV is an ongoing trend in the maritime industry for ocean research, ocean monitoring, military and commercial data-gathering, and so on [2, 29]. As AUV technologies have gradually matured, risk analysis for AUVs has rapidly become essential to ensure safer operations and assist decision making. A number of past efforts regarding risk analysis have been undertaken to improve the safety performance of AUVs. However, to the best knowledge of the authors, a systematic review and analysis of past studies has not yet been done. As a thorough review will enable domain researchers to gain a better understanding of AUV risk analysis and benefit future development, the authors believe that a critical review article is timely.
In light of the above, the objective of this article is to provide a structured review of risk analysis research regarding AUV operations. It aims to answer four key questions arising from historical developments and to highlight future trends in this domain. The four key questions as the focus of this review are listed in Table 2, which shows the overall process of this literature review from analyses of past studies. The main contribution of this study is to help researchers and AUV stakeholders obtain comprehensive insights about fundamental concepts and evolving methods for the risk analysis of AUVs. Meanwhile, it is expected to indicate directions for future research to bridge existing gaps.
The scope of this study is restricted to risk analysis for AUV operations. According to the objective and scope of this review, the literature retrieval was performed based on keywords searching including AUVs with the combination of risk identification, risk analysis, risk assessment, risk management, risk mitigation, risk modeling, and safety measures. A total of forty-two articles with significant relevance to the research purpose and scope were retrieved. In addition, to better answer the research questions and facilitate further statistical analysis, the selected publications were classified into various aspects, including the type of identified risk factors, the type of adopted risk analysis methods, the type of mission forms, the area of operations, and the type of potential consequences. The dataset of selected literature is classified and summarized in the Appendix.
The article is structured as follows. In section 2, critical risk factors of AUV operations are analyzed by categorizing them into technical factors, environmental factors, and human factors. Section 3 compares the evolving methods or models applied for AUV risk analysis by classifying them as three types: qualitative methods, semi-quantitative methods, and quantitative methods. Section 4 outlines current research gaps and future directions. The summary and conclusion of this study are given in Section 5.
Section snippets
Analysis of risk factors of AUV operations
Risk factors identification is defined as the process of identifying potential risk factors, which is the first step of the risk analysis phases [23]. Based on past studies, risk factors related to AUV operations are identified and analyzed in this section by categorizing them into technical factors, human factors, and environmental factors. Fig. 2 presents the number and distribution of publications of risk analysis of AUVs regarding three types of risk factors. As mentioned in the
Risk analysis methods of AUV operations
This section provides an overview of existing methods for risk analysis of AUV operations. It aims to outline the evolution of the developed methods and models, critically analyze the progress and limitations of past research, and highlight future research trends in this domain. This section is expected to help researchers gain a better understanding of historical developments for AUV risk analysis methods and bridge the existing research gaps in future work. In this section, the reviewed
Future challenges for risk analysis of AUV operations
Based on the above analysis of past progress, section 4 identifies current research gaps and discusses future challenges in the domain of AUV risk analysis.
Summary and conclusion
The main objective of this study is to provide a systematic review of past progress of risk analysis research for AUV operations. This review answers key questions including fundamental concepts and evolving methods in the domain of risk analysis for AUVs, and it highlights future research trends to bridge existing gaps. The scope of this article is restricted to the research questions. Based on the aim and scope of this study, a total of forty-two articles with significant relevance to
CRediT authorship contribution statement
Xi Chen: Conceptualization, Methodology, Visualization, Writing – original draft. Neil Bose: Conceptualization, Supervision, Validation, Funding acquisition, Writing – review & editing. Mario Brito: Supervision, Validation, Visualization, Writing – review & editing. Faisal Khan: Conceptualization, Supervision, Methodology, Funding acquisition, Writing – review & editing. Bo Thanyamanta: Resources, Validation, Writing – review & editing. Ting Zou: Supervision, Validation, Writing – review &
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work is funded by Fisheries and Oceans Canada through the Multi-partner Oil Spill Research Initiative (MPRI) 1.03: Oil Spill Reconnaissance and Delineation through Robotic Autonomous Underwater Vehicle Technology in Open and Iced Waters. Coauthor, Faisal Khan, wishes to acknowledge the financial support provided by the Canada Research Chair (Tier 1) program on Offshore Safety and Risk Engineering.
References (123)
- et al.
A Bayesian approach for predicting risk of autonomous underwater vehicle loss during their missions
Reliability Engineering & System Safety
(2016) - et al.
Corrosion induced failure analysis of subsea pipelines
Reliability Engineering & System Safety
(2017) - et al.
Analysis of causation of loss of communication with marine autonomous systems: A probability tree approach
Methods in Oceanography
(2014) Application of dynamic Bayesian network to risk analysis of domino effects in chemical infrastructures
Reliability Engineering & System Safety
(2015)- et al.
Risk information for operational decision-making in the offshore oil and gas industry
Safety Science
(2016) - et al.
Multi-unit nuclear power plant probabilistic risk assessment: A comprehensive survey
Reliability Engineering & System Safety
(2021) - et al.
Securing the operational reliability of an autonomous mini-submarine
Reliability Engineering & System Safety
(2000) - et al.
Safety performance monitoring of autonomous marine systems
Reliability Engineering & System Safety
(2017) - et al.
Review and analysis of methods for assessing maritime waterway risk based on non-accident critical events detected from AIS data
Reliability Engineering & System Safety
(2020) - et al.
An analysis of factors affecting the severity of marine accidents
Reliability Engineering & System Safety
(2021)