A method of inferring collision ratio based on maneuverability of own ship under critical collision conditions

유영준,이기표,안경수 대한조선학회 2013 International Journal of Naval Architecture and Oc Vol.5 No.2

In constructing a collision avoidance system, it is important to determine the time for starting collision avoidance maneuver. Many researchers have attempted to formulate various indices by applying a range of techniques. Among these indices, collision risk obtained by combining Distance to the Closest Point of Approach (DCPA) and Time to the Closest Point of Approach (TCPA) information with fuzzy theory is mostly used. However, the collision risk has a limit, in that membership functions of DCPA and TCPA are empirically determined. In addition, the collision risk is not able to consider several critical collision conditions where the target ship fails to take appropriate actions. It is therefore necessary to design a new concept based on logical approaches. In this paper, a collision ratio is proposed, which is the expected ratio of unavoidable paths to total paths under suitably characterized operation conditions. Total paths are determined by considering categories such as action space and methodology of avoidance. The International Regulations for Preventing Collisions at Sea (1972) and collision avoidance rules (2001) are considered to solve the slower ship’s dilemma. Different methods which are based on a constant speed model and simulated speed model are used to calculate the relative positions between own ship and target ship. In the simulated speed model, fuzzy control is applied to determination of command rudder angle. At various encounter situations, the time histories of the collision ratio based on the simulated speed model are compared with those based on the constant speed model.

A New Analytical Representation to Robot Path Generation with Collision Avoidance through the Use of the Collision Map

Seung-Hwan Park,Beom-Hee Lee 대한전기학회 2006 International Journal of Control, Automation, and Vol.4 No.1

A new method in robot path generation is presented using an analysis of the characteristics of multi-robot collision avoidance. The research is based on the concept of the collision map, where the collision between two robots is presented by a collision region and a crossing curve TLVSTC (traveled length versus servo time curve). Analytic collision avoidance is considered by translating the collision region in the collision map. The 4 different translations of collision regions correspond to the 4 parallel movements of the actual original robot path in the real world. This analysis is applied to path modifications where the analysis of collision characteristics is crucial and the resultant path for collision avoidance is generated. Also, the correlations between the translations of the collision region and robot paths are clarified by analyzing the collision/non-collision areas. The influence of the changes of robot velocity is investigated analytically in view of collision avoidance as an example.

중소형선박의 충돌회피지원모델에 관한 연구

김원욱,강성진,김영롱,김창제 한국수산해양교육학회 2019 水産海洋敎育硏究 Vol.31 No.4

This study aims to establish a collision avoidance support model applicable to small and medium-sized ships. Among the core technologies of autonomous vessels, it is most important to provide a route to avoid collision between ships. However, there has not been much research in this regard, and various alarm systems which give a warning against an approach of any ships in advance have been developed as the first stage of autonomous vessel. This research derived collision avoidance levels to choose a optimal route for collision avoidance, applying a collision judgement model to dynamic ship’s domains reflected the characteristics of length, speed, and maneuverability of a ship. of a ship. As a result, the collision avoidance level(CAL) proposed in this study expressed navigation situations of a ship properly. The collision avoidance level will play a great role in developing an autonomous ship and selecting the route to avoid a collision.

A method of inferring collision ratio based on maneuverability of own ship under critical collision conditions

You, Youngjun,Rhee, Key-Pyo,Ahn, Kyoungsoo The Society of Naval Architects of Korea 2013 International Journal of Naval Architecture and Oc Vol.5 No.2

In constructing a collision avoidance system, it is important to determine the time for starting collision avoidance maneuver. Many researchers have attempted to formulate various indices by applying a range of techniques. Among these indices, collision risk obtained by combining Distance to the Closest Point of Approach (DCPA) and Time to the Closest Point of Approach (TCPA) information with fuzzy theory is mostly used. However, the collision risk has a limit, in that membership functions of DCPA and TCPA are empirically determined. In addition, the collision risk is not able to consider several critical collision conditions where the target ship fails to take appropriate actions. It is therefore necessary to design a new concept based on logical approaches. In this paper, a collision ratio is proposed, which is the expected ratio of unavoidable paths to total paths under suitably characterized operation conditions. Total paths are determined by considering categories such as action space and methodology of avoidance. The International Regulations for Preventing Collisions at Sea (1972) and collision avoidance rules (2001) are considered to solve the slower ship's dilemma. Different methods which are based on a constant speed model and simulated speed model are used to calculate the relative positions between own ship and target ship. In the simulated speed model, fuzzy control is applied to determination of command rudder angle. At various encounter situations, the time histories of the collision ratio based on the simulated speed model are compared with those based on the constant speed model.

COLREGs-compliant multiship collision avoidance based on deep reinforcement learning

Zhao, Luman,Roh, Myung-Il Elsevier 2019 Ocean engineering Vol.191 No.-

<P><B>Abstract</B></P> <P>Developing a high-level autonomous collision avoidance system for ships that can operate in an unstructured and unpredictable environment is challenging. Particularly in congested sea areas, each ship should make decisions continuously to avoid collisions with other ships in a busy and complex waterway. Furthermore, recent reports indicate that a large number of marine collision accidents are caused by or are related to human decision failures concerning a lack of situational awareness and failure to comply with the Convention on the International Regulations for Preventing Collisions at Sea (COLREGs). In this study, we propose an efficient method to overcome multiship collision avoidance problems based on the Deep Reinforcement Learning (DRL) algorithm by expanding our previous study (Zhao et al., 2019). The proposed method directly maps the states of encountered ships to an ownship's steering commands in terms of rudder angle using the Deep Neural Network (DNN). This DNN is trained over multiple ships in rich encountering situations using the policy-gradient based DRL algorithm. To address multiple encountered ships, we classify them into four regions based on COLREGs, and consider only the nearest ship in each region. We validate the proposed collision avoidance method in a variety of simulated scenarios with thorough performance evaluations, and demonstrate that the final DRL controller can obtain time efficient and collision-free paths for multiple ships. Simulation results indicate that multiple ships can avoid collisions with each other while following their own predefined paths simultaneously. In addition, the proposed approach demonstrates its excellent adaptability to unknown complex environments with various encountered ships.</P> <P><B>Highlights</B></P> <P> <UL> <LI> This study deals with a multiship collision avoidance problem in congested sea areas. </LI> <LI> A COLREGs-compliant method for multiship collision avoidance is proposed based on deep reinforcement learning. </LI> <LI> A novel strategy is used to solve the problem by categorizing target ships into four regions defined by COLREGs. </LI> <LI> The proposed method is applied to a wide range of problems, from single collision avoidance to multiple collision avoidance. </LI> </UL> </P>

Enabling Vessel Collision-Avoidance Expert Systems to Negotiate

Qinyou Hu,Chaojian Shi,Haishan Chen,Qiaoer Hu 한국항해항만학회 2006 한국항해항만학회 학술대회논문집 Vol.1 No.-

Automatic vessel collision-avoidance systems have been studied in the fields of artificial intelligence and navigation for decades. And to facilitate automatic collision-avoidance decision-making in two-vessel-encounter situation, several expert and fuzzy expert systems have been developed. However, none of them can negotiate with each other as seafarers usually do when they intend to make a more economic overall plan of collision avoidance in the COLREGS-COST-HIGH situations where collision avoidance following the International Regulations for Preventing Collisions at Sea(COLREGS) costs too much. Automatic Identification System(AIS) makes data communication between two vessels possible, and negotiation methods can be used to optimize vessel collision avoidance. In this paper, a negotiation framework is put forward to enable vessels to negotiate to optimize collision avoidance in the COLREGS-COST-HIGH situations at open sea. A vessel vector space is defined and therewith a cost model is put forward to evaluate the cost of collision-avoidance actions. Negotiations between a give-way vessel and a stand-on vessel and between two give-way vessels are considered respectively to reach overall low cost agreements. With the framework proposed in this paper, two vessels involved in a COLREGS-COST-HIGH situation can negotiate with each other to get a more economic overall plan of collision avoidance than that suggested by the traditional collision-avoidance expert systems.

DECISION MAKING FOR COLLISION AVOIDANCE SYSTEMS CONSIDERING A FOLLOWING VEHICLE

신성근,이혁기,유승한 한국자동차공학회 2023 International journal of automotive technology Vol.24 No.2

In vehicles with driving assistance systems, the responsibility of driving still lies with the driver. Therefore, active safety systems follow a design principle of avoiding intervention as long as possible. Decision making using this principle requires collision avoidance with surrounding objects during the evasive maneuvering of an ego vehicle. However, general decision-making methods for collision avoidance focus on preventing collisions with objects in front of the subject vehicle; the risk of collision with the following vehicle caused by emergency braking is not considered. This paper presents decision-making methods for such collision avoidance systems, wherein emergency braking and steering are considered as collision avoidance maneuvers. The risk of collision is predicted based on the braking model of a normal driver and several emergency braking strategies are designed to avoid collisions. A within-lane steering avoidance method to improve avoidance performance in small overlap collision scenarios and a general avoidance method of steering to change lanes are designed. Collisions with surrounding objects can be avoided using the designed evasive maneuvers; further, maneuvers that can be implemented at the last moment are determined. The results of computer simulations indicate an improved collision avoidance performance using the proposed methods.

A Study on Intention Exchange-based Ship Collision Avoidance by Changing the Safety Domain

Donggyun Kim 해양환경안전학회 2019 海洋環境安全學會誌 Vol.25 No.3

Even if only two ships are encountered, a collision may occur due to the mistaken judgment of the positional relationship. In other words, if an officer does not know a target ship’s intention, there is always a risk of collision. In this paper, the experiments are conducted to investigate how the intention affects the action of collision avoidance in cooperative and non-cooperative situations. In non-cooperative situation, each ship chooses a course that minimizes costs based on the current situation. That is, it always performs a selfish selection. In a cooperative situation, the information is exchanged with a target ship and a course is selected based on this information. Each ship uses the Distributed Stochastic Search Algorithm so that a next-intended course can be selected by a certain probability and determines the course. In the experimental method, four virtual ships are set up to analyze the action of collision avoidance. Then, using the actual AIS data of eight ships in the strait of Dover, I compared and analyzed the action of collision avoidance in cooperative and non-cooperative situations. As a result of the experiment, the ships showed smooth trajectories in the cooperative situation, but the ship in the non-cooperative situation made frequent big changes to avoid a collision. In the case of the experiment using four ships, there was no collision in the cooperative situation regardless of the size of the safety domain, but a collision occurred between the ships when the size of the safety domain increased in cases of non-cooperation. In the case of experiments using eight ships, it was found that there are optimal parameters for collision avoidance. Also, it was possible to grasp the variation of the sailing distance and the costs according to the combination of the parameters, and it was confirmed that the setting of the parameters can have a great influence on collision avoidance among ships.

Quantitative calculation method of the collision risk for collision avoidance in ship navigation using the CPA and ship domain

Ha Jisang,Roh Myung-Il,Lee Hye-Won 한국CDE학회 2021 Journal of computational design and engineering Vol.8 No.3

Collision risk (CR) assessment is necessary for avoiding collisions with other ships. The CR can be used to make decisions on collision avoidance. In this respect, the ship domain and the closest point of approach (CPA)-based methods have been proposed to assess the CRs. However, the ship domain method is limited in terms of the quantitative calculation of the CR, whereas the CPA-based method does not guarantee reliable collision avoidance. In this study, an improved method is proposed for the quantitative calculation of the CR in ship navigation that combines the advantages of two existing methods. The proposed method calculates the CR using the CPA and defines the ship domain as a critical value of the CR to ensure reliable collision avoidance. In this process, the CR value of another ship on the boundary of the ship domain is calculated as 1, which implies that collision occurs, and the coefficients for the CR calculation are adjusted considering the distance from the ship domain. Furthermore, the manoeuvring performance and the heading angle of the ships are considered in the calculation of the CR. To evaluate the proposed method, it is applied to various examples, including a comparison with previous methods. The results show that the proposed method can be used to obtain a quantitative CR for collision avoidance.

Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship

전도현,노명일,이혜원,유동훈 대한조선학회 2024 International Journal of Naval Architecture and Oc Vol.16 No.-

In this paper, we propose a collision avoidance method based on deep reinforcement learning (DRL) that simultaneously controls the path and speed of a ship. The DRL is actively applied in machine control and artificial intelligence. To verify the proposed method, we applied it to the Imazu problem. It provides benchmark scenarios for collision avoidance. In particular, we compared and analyzed the collision avoidance performance according to the level of learning and various parameters to ensure that the proposed method displays optimal avoidance performance. The results indicated that the proposed method can determine a safe avoidance path for a given situation. Finally, to compare the performance of the proposed method, we compared the collision avoidance method based on the path–speed control of the OS proposed in this study with the collision avoidance method that controls only the path of the OS (Chun et al., 2021). We observed that the proposed method failed in 6 out of 20 scenarios of the Imazu problem when only the path of the OS was controlled. However, it succeeded in collision avoidance in all the 20 scenarios when both path and speed were controlled simultaneously.