Formal expression of ship behavior is the basis for developing autonomous navigation systems, which supports the scene recognition, the intention inference, and the rule-compliant actions of the systems. The Convention on the International Regulations for Preventing Collisions at Sea (COLREGs) offers experience-based expressions of ship behavior for human beings, helping the humans recognize the scene, infer the intention, and choose rule-compliant actions. However, it is still a challenge to teach a machine to interpret the COLREGs. This paper proposed an ontological ship behavior model based on the COLREGs using knowledge graph techniques, which aims at helping the machine interpret the COLREGs rules. In this paper, the ship is seen as a temporalspatial object and its behavior is described as the change of object elements in time spatial scales by using Resource Description Framework (RDF), function mapping, and set expression methods. To demonstrate the proposed method, the Narrow Channel article (Rule 9) from COLREGs is introduced, and the ship objects and the ship behavior expression based on Rule 9 are shown. In brief, this paper lays a theoretical foundation for further constructing the ship behavior knowledge graph from COLREGs, which is helpful for the complete machine reasoning of ship behavior knowledge in the future.@en

Collison between ships is one of the major contributors to maritime accidents. To reduce ship collision accidents, the research on collision avoidance decision-making has been drawing much attention from various parties. In this research, extensive literature and expert knowledge are collected and analyzed to identify the common sense and discrepancies between collision avoidance decision-making for theoretical research and navigation practices. The key factors that are considered in the two perspectives are identified and discussed, based on which, the knowledge structures that can represent the development of the process in the two perspectives are established. A series of comparisons between the knowledge structure based on theoretical research and navigation practices are conducted. The comparisons indicate clear common sense and discrepancies between the theoretical research and navigation practices regarding collision avoidance decision-making. The potential causes of them are also analyzed. The research results would be beneficial for the development of collision avoidance decision-making for both autonomous and conventional manned ships in maritime traffic.

@en

Collision prevention is critical for navigation safety at sea. At early ages, researchers aimed at developing navigational assistance systems for enhancing situational awareness of human operators as human is at the core of collision avoidance. Recently, autonomous vehicles have gained a remarkable amount of attention with a focus on solving collision problems by machines. This results in two groups of studies, both working on preventing collisions but with different focuses: one aims at conflict detection, and the other focuses on conflict resolution. This paper offers a comprehensive overview of collision prevention techniques based on the three basic processes of determining evasive solutions, namely, motion prediction, conflict detection, and conflict resolution. The strengths and weaknesses of different methods for these three fundamental processes are discussed. Limitations and new challenges are highlighted. Moreover, this review points out the differences between the research for manned and unmanned ships and how the research in the two domains can learn from each other. A potential roadmap for the transition from existing manned ships to fully unmanned ships is provided in the end.

@en

This article provides a comprehensive overview on cooperative control methods for waterborne transport. We first proposed a hierarchical architecture of cooperation in the waterborne transport systems. Three layers of cooperation are identified according to the range of communication and cooperation, i.e., the individual layer, the local layer, and the network layer. The individual layer is the basis layer where a controller controls the dynamics of an individual vessel. The local layer considers the vessel-to-vessel (V2V) and vessel-to-infrastructure (V2I) interactions. The network layer considers not only V2V and V2I interactions but also the interdependence of the interconnected infrastructures, i.e., infrastructure-to-infrastructure interactions. Existing research for cooperation at each layer is reviewed, and the main research gaps are provided.

@en

Urban waterways have great potential in cargo transport to relieve the congestion in the overloaded road networks. This paper explores the potential of applying cooperative multi-vessel systems (CMVSs) to improve the safety and efficiency of transport in urban waterway networks. A framework consisting of vessel train formation (VTF) and cooperative waterway intersection scheduling (CWIS) is proposed. Two types of controllers are introduced. Intersection controllers solve the CWIS problems and assign each vessel a desired time of arrival and vessel controllers are responsible for the VTF in waterway segments and the timely arrival at the intersections. An alternating direction method of multipliers (ADMM)-based negotiation framework is proposed for the cooperation among the controllers. The simulation experiments involving the scenarios in which up to 50 vessels sailing in the canal network in Amsterdam are carried out to illustrate the effectiveness of the proposed approach. In the simulation of an isolated intersection, rescheduling is triggered when some vessels cannot arrive on time. Although some ASVs arrive later, the time that is needed for all the ASVs to pass through is the same after rescheduling. Moreover, we compare the cooperative situation with the proposed CMVSs with a baseline situation. In the baseline situation, vessels avoid collisions using the generalized velocity obstacle (GVO) method and cross the intersection with a first in, first out rule. The CMVSs show better path following performance, while the GVO method needs fewer velocity changes. From the perspective of efficiency, the CMVSs help to reduce the total time to pass through the intersection.@en

Collision risk assessment is essential for supporting collision avoidance, which is the core of various collision alert/avoidance systems. One main task of the systems is setting off alarms for taking evasive actions. The alarms need to be triggered before the conflict has no collision-free solution. However, most of existing collision risk measures are independent of conflict resolution. That means the collision alert does not indicate that the collision is avoidable or not. This article proposes an improved time-varying collision risk (TCR) measure, bringing in a new measure. The measurement of TCR reflects not only the dangerous level of the approaching ships but also the difficulty of avoiding collisions. By comparing with traditional measures, e.g., Collision Risk Index (CRI), we found that (1) the TCR can distinguish changes of risk that have identical CRI level, (2) the TCR measure offers a reasonable tool to evaluate the collision risk of entire traffic, and (3) it reflects the influence of maneuverability improvement on collision risk. Based on those results, this article reaches two conclusions: the collision risk is monotonically increasing when introducing more ships, and ignorance of ship maneuverability results in an underestimation of collision risk.@en

Ship collision is one of the major contributors of maritime accidents. Quantitative risk analysis of such accidents is an effective tool for maritime safety administrations to understand the current risk level and propose risk mitigation measures. In this paper, an improved Time Discretized Non-linear Velocity obstacle (TD-NLVO) algorithm is proposed to detect multiple ship encounter situations using historical AIS data. Boolean operation on the individual NLVO is integrated with TD-NLVO using the union of the velocity-obstacle sets to determine a dangerous encounter situation according to the pre-set criteria. Two case studies are implemented to illustrate the capability of the proposed algorithm. A comparison is conducted between the previous and the improved methods. The results indicate that the improved method can effectively identify a multiple ship encounter which satisfies the pre-set criteria. The improved method has the potential to provide more detailed information for stakeholders e.g. maritime safety administration, etc. to propose risk mitigation measures as well as to improve the accuracy of geometric probability analysis for ship collision risk.

@en

The lack of situational awareness is a major cause of ship collisions. Thus, enhancing the situational awareness of the stand-on ship is a key for navigational safety, where the intention estimation of the give-way ship is crucial. According to COLREGs, the stand-on ship is not allowed to take evasive actions until the give-way ship does not take proper actions timely. The stage that needs the stand-on ship to take actions plays as the second protective layer for the ship, which is named as ‘Stand-on Ship as Second Line of Defense’ (SLoD). A method to estimate the intention of the give-way ship and to trigger SLoD is proposed in this article. Four modules of the proposed method include: “data pre-processing” collects all traffic information and determines the ships' obligations; “action identification” pinpoints the turning points; “action uncertainty” generates a bounded reachable velocity considering the give-way ship's maneuverability; “conflict assessment” judges potential collision by using non-linear velocity obstacle algorithm. Several typical encounter scenarios are simulated to demonstrate the feasibility of the proposed method. The results show that intention estimation of the give-way ship improves the situational awareness of the stand-on ship, which can support the stand-on ship to make collision avoidance decisions.

@en

Maritime Autonomous Surface Ships (MASS) attract increasing attention in recent years. Researchers aim at developing fully autonomous systems that replace the role of human operators. Studies either focus on supporting conflict/collision detection (for manned ships) or solving conflict automatically (for unmanned ships). The cooperation between human and machine has been less focused on in existing studies. However, this type of cooperation is essential both in practice and in the future: firstly, demands on navigational assistance are still strong for supporting navigators in manned ships; secondly, MASS with different autonomy levels require increasing cooperation between human operators and machines, e.g. monitoring automation, remotely controlling the ship, etc.; thirdly, the intelligence of human and the machines is highly complementary. Moreover, fully autonomous ships cannot replace all the manual ships overnight. Therefore, the future waterborne transport system will be a system in which both human-operated vessels and autonomous vessels exist. In this article, we firstly provide an overview of existing modes of human-machine interaction (HMI) during ship collision avoidance. Then, we propose a framework of HMI oriented Collision Avoidance System (HMI-CAS) whose decision-making process is interpretable and interactive for human operators. The HMI-CAS facilitates automatic collision avoidance and enables the human operators to take over the control of the MASS safely. Moreover, the proposed framework acknowledges the under-actuated feature of ships. Simulations are carried out to demonstrate the proposed HMI-CAS. The results show that the proposed HMI-CAS can not only control the under-actuated MASS to avoid collision automatically but also share the decision-making with human operators and support the operators to control the MASS.

@en

In this research, a hybrid approach for global path planning for Maritime Autonomous Surface Ship (MASS) is proposed, which generates the shortest path considering the collision risk and the proximity between path and obstacles. The collision risk concerning obstacles is obtained using Time-Varying Collision Risk (TCR) concept, taking into account the velocity constraint of the ship that can achieve during operation. The influence of proximity from obstacles is measured with the Fast Marching (FM) algorithm. A new cost function is proposed allowing to combine the influence of obstacle proximity and collision risk in the region. Finally, the Fast Marching Square algorithm is applied to generate the globally optimal path that can reach the pre-set destination. The contribution of this work is two-fold: 1) considering the velocity constraint of the own ship, together with its influences of collision risk into the global path planning stage of autonomous navigation. 2) measuring the collision risk induced by the obstacles from their comprehensive influences on the achievable velocity range using TCR concept, instead of numerical integration of risk measurement. The results of the case study indicate that the proposed approach can find an optimal path considering the collision risk and proximity from the obstacles.

@en

Collision risk measurement is an essential topic for ship collision prevention. Many risk measures, i.e. DCPA/TCPA, etc., decouple the ship traffic into several pairs of ships and then evaluate the risk in each pair. This kind of measurement loses some information of the entire traffic and might include some biases in risk measurement, especially in multiple‐ship scenarios. In this article, Imminent Collision Risk Assessment (ICRA) is extended, which formulates collision risk as a ratio of reachable maneuvers leading to a collision and all reachable maneuvers (velocities). Two groups of scenarios have been simulated to show the ICRA is suitable for assessing the collision risk in multiple‐ship scenarios. Moreover, two improvements have been introduced: (1) a generalized velocity obstacle algorithm is introduced to collect the maneuvers leading to collisions, which considers ship dynamics; (2) the constraints of forces are considered in the formulation of reachable maneuvers. As a result, the proposed measurement helps one ship assess the risk of approaching obstacles which are difficult to avoid the collision in terms of own‐ship’s dynamics and kinetic constraints.

@en

Numerous methods have been developed for ship collision prevention over the past decades. However, most studies are based on strong assumptions, such as the need for a constant velocity of the target-ship, the limitation to two-ship scenarios, the simplification of ships’ dynamics, etc. Generalized Velocity Obstacle (GVO) algorithm can bridge these gaps. This paper presents a GVO algorithm for ship collision avoidance and designs a collision avoidance system (GVO-CAS). The proposed system visualizes the changes of one ship's course and speed resulting in collisions, which can be used not only for supporting the officer on watch to prevent collisions, but also for collision prevention of Autonomous Surface Vessels (ASVs) and for human operators taking over the control of ASVs. Simulation experiments show that the proposed collision avoidance system can work properly in various maritime environments. Compared to the original Velocity Obstacle algorithm, the GVO algorithm is more reliable and suitable for close range ship collision avoidance. Moreover, the GVO-CAS can offer rule-compliant evasive actions with a minimum number of required actions for ships. These results show the great potential to use the GVO algorithm in both manned and unmanned ships at sea.

@en

We propose an innovative time-varying collision risk (TCR) measurement for ship collision prevention in this article. The proposed measurement considers the level of danger of the approaching ships and the capability of a ship to prevent collisions. We define the TCR as the probability of the overlap of ships’ positions in the future, given the uncertainty of maneuvers. Two sets are identified: (1) the velocity obstacle set as the maneuvers of the own ship that lead to collisions with target ships, and (2) the reachable velocity set as the maneuvers that the own ship can reach regarding its maneuverability. We then measure the TCR as the time-dependent percentage of overlap between these two sets. Several scenarios are presented to illustrate how the proposed measurement identifies the time-varying risk levels, and how the approach can be used as an intuitively understandable tool for collision avoidance.

@en

Ship collision is a classical problem for maritime practitioners and researchers. Human error is a major cause of collision accidents, which motivates researchers to develop automation systems replacing navigators on board. However, before autonomous ships fully replace conventional ships, supporting the situational awareness of human operators is still a rigid demand. Moreover, how to prevent automation performing out of the human’s expectations (e.g., violation of regulations) is another challenge. Therefore, the design of human-machine interactions (HMIs) becomes crucial.

This dissertation developed the Human-Machine Interaction oriented Collision Avoidance System (HMI-CAS) that allows human operators and automation to share their intelligence. Specifically, the HMI-CAS not only offers one (optimal) solution to human operators but also visualizes the solution space with both dangerous solutions and feasible solutions. Thus, the decision process of automation becomes transparent for human operators. The human operators can not only read and understand the solutions offered by the machine but also validate and modify the solutions via the interface of the HMI-CAS. Without human interventions, the HMI-CAS also can work automatically. Moreover, to support the humans take evasive action in time, the measure of collision risk utilizing a concept called “room-for-maneuver” is proposed, which offers alerts before collisions become inevitable.

In brief, instead of replacing humans on board, the proposed HMI-CAS aims at bridging the intelligence of humans and machines, which enriches the choice of collision avoidance systems for supporting human operators and for developing autonomous ships.
@en

Traffic flow theory has been applied to the study of marine traffic flow and underpins the phenomena being studied. This paper is intended to demonstrate the uses of classical fundamental diagrams, which are developed previously on the basis of highway traffic, by conducting empirical tests on marine traffic data. Marine traffic data are extracted from the Automatic Identification System (AIS) from Shanghai port in 2010 and 2018. Interpolation methods are used to estimate the three primary variables of fundamental diagrams, namely, flow rate, traffic density, and local speed. The relationships among these three variables are discussed and marine traffic data are further compared and contrasted with classical traffic flow models. The importance of the finding is to create consistency, predictability and uniformity of uses of fundamental diagrams on marine traffic flow.

@en

Maritime transportation system has made a significant contribution to the development of the world economy. However, with the growth of quantity, scale, and speed of ships, maritime accidents still pose incrementing risk to individuals and societies in terms of multiple aspects, especially collision accidents between ships. Great effort is needed to prevent the occurrence of such accidents and to improve navigational safety and traffic efficiency. In this paper, extensive literature on probabilistic risk analysis on ship-ship collision was collected and reviewed focusing on the stakeholders which may benefit from the research and the methodologies and criteria adopted for collision risk. The paper identifies stakeholders, the modelling aspects (frequency estimation, causation analysis, etc.) in which the stakeholders are interested in. A classification system is presented based on the technical characteristics of the methods, followed by detailed descriptions of representative approaches and discussion. Areas for improvement of such risk analysis approaches are highlighted, i.e. identifying collision candidates, assessing the collision probability of multiple ships encounters, assessing the human and organizational factors. Three findings are concluded from this literature review: (1) Research on collision risk analysis and evaluation of ship encounters from individual ship perspective have facilitated the research in macroscopic perspective, and in turn, results from macroscopic research can also facilitate individual risk analysis by providing regional risk characteristics; (2) Current approaches usually estimate geometric probability by analysing data at certain intervals, which could lead to over/underestimation of the results; and (3) For causation probability induced by human and organisational factors in collision accidents, lack of data and uncertainty is still a problem to obtain accurate and reliable estimations. The paper also includes a discussion with respect to the applicability of the methods and outlines further work for improvement. The results in this paper are presented in a systematic structure and are formulated in a conclusive manner. This work can potentially contribute to developing better risk models and therefore better maritime transportation systems.@en

It is of critical importance to prevent collisions at sea for navigation safety. Some popular techniques have been proposed and been used in practice, e.g. closest point of approach, collision threat parameters area, etc. However, most of these techniques assume that the target ship keeps a constant velocity which is unrealistic and may easily lead to a false alarm. In this article, Velocity Obstacle (VO) algorithms are applied to support collision avoidance with target ships whose trajectories are non-linear (with time-dependent velocities) and (probabilistically) predictable. In particular, Linear-VO, Non-Linear VO, and Probabilistic VO algorithms are used for this purpose. Compared to traditional approaches, these algorithms are capable of detecting collision dangers with target ships sailing non-linearly and (probabilistically) predictably, finding collision-free velocities in multi-ship scenarios, and preventing collisions. Two scenarios are designed to demonstrate the performance of VO algorithms. The results show that the proposed VO algorithms can facilitate collision detection and offer proper collision-free solutions for ships.

@en

Maritime accidents have been imposing various risks to individuals and societies in terms of human and property loss, and environmental consequences. For probabilistic risk analysis and management, collision candidate detection is the first step. Therefore, it is of great importance to further improve methods to detect possible collision scenarios. This paper proposes a Time Discrete Non-linear Velocity Obstacle (TD-NLVO) method for collision candidate detection, which is based on the Non-linear Velocity Obstacle algorithm and tested on historical AIS data (Automatic Identification System). Collision candidates are detected based on the perspective which considers a ship encounter as a process, rather than analysing traffic data at certain time slices. Case studies on single encounters of ship traffic in waterways environments are conducted and presented in this paper. The results indicate that the TD-NLVO method can effectively detect collision candidates which satisfy pre-set criteria. A comparison between seven other popular AIS data-based collision candidate methods is performed, and the results indicate that the proposed method outperforms the other methods regarding its robustness towards the choice of parameter settings.@en

In order to quantitatively analyze and classify waterborne traffic flow complexity, a method of waterborne traffic flow complexity metric was proposed based on Lempel-Ziv algorithm and TOPSIS (technique for order preference by similarity to an ideal solution). Firstly, Lempel-Ziv algorithm was used to obtain the complexity eigenvalues of measured time sequences of waterborne traffic flow and other compared sequences (periodic sequence, logistic sequence, Henon sequence and random sequence). Secondly, the close degree of each sequence was calculated by using TOPSIS. The class of complexity was divided by those close degrees of compared sequences. At last, the complexity degree of each sequence was represented by close degrees of time sequences of waterborne traffic flow and the complexity class. This complexity metric was carried out on the waterborne traffic flow of south channel of the Yangtze River. Calculation result shows that the correlation coefficient of complexity close degree of waterborne ship traffic flow is 0.6981 with the number of ship traffic accidents, and 0.7692 with the downside traffic flow of standard ships, respectively. The change trend is basically consistent, so the complexity close degree can reflect the complexity of waterborne ship traffic flow. The close degrees of periodic sequence and random sequence are 0.0001 and 0.9999, respectively. Meanwhile, the close degrees of logistic sequence and Henon sequence are 0.4492 and 0.5377, respectively. The close degrees of logistic sequence and Henon sequence are greater than periodic sequence and and less than random sequence. The close degrees of waterborne traffic flow from July to November in 2013 are 0.8280, 0.8527, 0.8565, 0.8237 and 0.8107, respectively, the complexity of waterborne traffic flow is basically consistent. The values of waterborne traffic flow are far greater than those of periodic sequence, locate between the values of Henon sequence and random sequence, and are closer to the value of random sequence, which shows that the waterborne traffic system is neither periodic nor completely stochastic. The complexity class of time sequence of waterborne traffic flow is Level 1, showing high complexity.@en

Ships collision assessment and conflict resolution are two main research problems in the maritime domain. In this paper, a methodology of collision risk assessment is introduced which is based on Velocity Obstacle (VO). In this paper, VO is employed to distinguish the dangerous velocities and judge the collision probability. The ship maneuverability is also considered to find the reachable velocities (RV) in a given time window. The percentage of dangerous velocities in all RV is defined as collision risk. By this approach, the proposed assessment model not only can identify the collision dangers, but also measure the probability of surviving in the encounter situation, regarding to its maneuverability. Several simulated encounter situations are presented to demonstrate this model. @en