e-Navigation based cooperative collision avoidance at sea: The MTCAS approach

Denker, Christian; Baldauf, Michael; Fischer, Sandro; Hahn, Axel; Ziebold, Ralf; Gehrmann, Elke; Semann, Mika

doi:10.1109/euronav.2016.7530566

Cited by 5 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following that close quarters have been defined in [38], where the author observed that COLREGS [39,40] do not precisely instruct navigators on specific distances in which the evasive action is necessary. Such distances were thus determined in [41] and recently applied in Maritime Traffic Alert and Collision Avoidance System [42,43]. Safe distances of evasive actions were also investigated in [44].…”

Section: Literature Reviewmentioning

confidence: 99%

A Ship Domain-Based Method of Determining Action Distances for Evasive Manoeuvres in Stand-On Situations

Szlapczynski

Krata

Szlapczynska

2018

Journal of Advanced Transportation

View full text Add to dashboard Cite

A ship encounter can be considered safe if neither of ships’ domains (defined areas around ships) is intruded by other ships. Published research on this includes optimising collision avoidance manoeuvres fulfilling domain-based safety conditions. However, until recently there was no method, using ship’s domain to determine exact moment when a particular collision avoidance manoeuvre can still be successfully performed. The authors have already proposed such method for give-way encounters. In the paper, documenting continuation of the research, another kind of scenarios is considered. This paper is focused on situations where the own ship is the stand-on one and the target is supposed to manoeuvre. The presented method uses a ship’s dynamics model to compute distance necessary for a manoeuvre successful in terms of avoiding domain violations. Additionally, stability-related phenomena and their impact on possible manoeuvres in heavy weather are taken into account. The method and applied models are illustrated in a series of simulation results. The simulations cover various examples of stand-on situations, including encounters in heavy weather conditions. Discussed manoeuvres may be limited to course alteration or may combine turns with speed reduction.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

A Ship Domain-Based Method of Determining Action Distances for Evasive Manoeuvres in Stand-On Situations

Szlapczynski

Krata

Szlapczynska

2018

Journal of Advanced Transportation

View full text Add to dashboard Cite

show abstract

“…In general, artificial ship collision avoidance mainly depends on ship position and motion relationship to determine the collision avoidance opportunity and make collision avoidance decisions using methods such as a ship domain-based approach [6], time to the closest point of approach (TCPA) and distance at closest point of approach (DCPA) [7]. Collision avoidance of an unmanned surface vessel (USV) depends on sensor fusion methods [8][9][10][11], which are capable of detecting static and dynamic obstacles.…”

Section: Literature Review 21 Ship Collision Avoidance Methodsmentioning

confidence: 99%

“…Therefore, lane deviation is not encouraged while navigating. Hence, this reward was denoted as a lane = λ lanein , in lane −λ laneout , lane deviation (7) where λ lanein denotes the reward when sailing in a waterway normally, and −λ laneout denotes the punishment when the ship has sailed out of the route.…”

Section: Markov Decision Process Of Multi-ship Cooperative Collision Avoidancementioning

confidence: 99%

A Novel Ship Collision Avoidance Awareness Approach for Cooperating Ships Using Multi-Agent Deep Reinforcement Learning

Chen

et al. 2021

JMSE

View full text Add to dashboard Cite

Ships are special machineries with large inertias and relatively weak driving forces. Simulating the manual operations of manipulating ships with artificial intelligence (AI) and machine learning techniques becomes more and more common, in which avoiding collisions in crowded waters may be the most challenging task. This research proposes a cooperative collision avoidance approach for multiple ships using a multi-agent deep reinforcement learning (MADRL) algorithm. Specifically, each ship is modeled as an individual agent, controlled by a Deep Q-Network (DQN) method and described by a dedicated ship motion model. Each agent observes the state of itself and other ships as well as the surrounding environment. Then, agents analyze the navigation situation and make motion decisions accordingly. In particular, specific reward function schemas are designed to simulate the degree of cooperation among agents. According to the International Regulations for Preventing Collisions at Sea (COLREGs), three typical scenarios of simulation, which are head-on, overtaking and crossing, are established to validate the proposed approach. With sufficient training of MADRL, the ship agents were capable of avoiding collisions through cooperation in narrow crowded waters. This method provides new insights for bionic modeling of ship operations, which is of important theoretical and practical significance.

show abstract

“…Taking into consideration the safe and secure navigation of the vessels, which is one of the main objectives of the "e-Navigation" strategy that is adopted by the IMO (IMO, 2007), additional efforts were pursued by researchers to develop a scheme that will allow the optimisation of course planning and the reduction of the traffic conjunction and collision accidents (Denker et al, 2016), and a technology that will identify the safe and dangerous navigational sectors and support the OOW in making correct decisions (TotemPlus, 2014). Furthermore, the NDSS-NAVDEC was developed, consisting of numerous modules that focus on the collection of data from shipboard and alternative resources, and the analysis assessment of the navigational situation of the own vessel (Pietrzykowski et al, 2012;Borkowski, 2014;Borkowski, 2017).…”

Section: Collision Avoidance Techniquesmentioning

confidence: 99%