• 대한조선학회논문집
• /
• 제56권1호
• /
• pp.58-65
• /
• 2019

The stability, reliability and efficiency of a smart ship are important issues as the interest in an autonomous ship has recently been high. An automatic collision avoidance system is an essential function of an autonomous ship. This system detects the possibility of collision and automatically takes avoidance actions in consideration of economy and safety. In order to construct an automatic collision avoidance system using reinforcement learning, in this work, the sequential decision problem of ship collision is mathematically formulated through a Markov Decision Process (MDP). A reinforcement learning environment is constructed based on the ship maneuvering equations, and then the three key components (state, action, and reward) of MDP are defined. The state uses parameters of the relationship between own-ship and target-ship, the action is the vertical distance away from the target course, and the reward is defined as a function considering safety and economics. In order to solve the sequential decision problem, the Deep Deterministic Policy Gradient (DDPG) algorithm which can express continuous action space and search an optimal action policy is utilized. The collision avoidance system is then tested assuming the $90^{\circ}$intersection encounter situation and yields a satisfactory result.

• 선박안전
• /
• 통권38호
• /
• pp.69-77
• /
• 2015

If we don't know the intention of altering course of a target ship when being in a head-on or a crossing situation, we may be confused about our decision making to change our course for collision avoidance and be in a danger of collision. In order to solve these problems, we need to develop an automatic detection system on altering course of a target ship for efficient collision avoidance. In this paper, we proposed an early detection system on altering course of a target ship using the steering wheel signal. This system will contribute to the reduction of collision accidents and also be used to the VTS system and the analysis of marine accidents.

• 선박안전
• /
• 통권36호
• /
• pp.71-78
• /
• 2014

If we don't know the intention of altering course of a target ship when being in a head-on or a crossing situation, we may be confused about our decision making to change our course for collision avoidance and be in a danger of collision. In order to solve these problems, we need to develop an automatic detection system on altering course of a target ship for efficient collision avoidance. In this paper, we proposed an early detection system on altering course of a target ship using the steering wheel signal. This system will contribute to the reduction of collision accidents and also be used to the VTS system and the analysis of marine accidents.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2007년도 추계학술대회 및 제23회 정기총회
• /
• pp.79-80
• /
• 2007

An expert system of collision avoidance developed by CLIPS and Visual C++ is continuously introduced in this paper. Further, a simulation function of collision avoidance is added to the expert system, the function can simulate the avoidance actions of own ship and a specific target of a period of future time. This function can help navigators to estimate collision risk and make proper collision avoidance actions in dangerous situations for navigational safety of ships. Furthermore, navigational rules can also be evaluated during the process of simulation.

• 한국항해항만학회지
• /
• 제31권4호
• /
• pp.309-315
• /
• 2007

We've investigated the characteristics on mariner's behavior in the collision situation through a full-mission ship handling simulator and considered that it's necessary to model the standard avoiding behavior of mariners in order to apply the obtained results more widely and effectively. Thus we described the contents of standard avoiding behavior taken by mariners in the collision situation and established the concept of the standard model based on human factors for collision avoidance in a previous study. As a following study, this paper is to propose the method of modeling on mariners' standard behavior for collision avoidance by analyzing the contents of mariner's information processing and the related factors using regression analysis. As a result, we confirmed the influence of relating factors to avoiding behavior in mariner's deciding decisions and proposed the modeling method of mariners' standard behavior for collision avoidance with a example of recognition model.

• 한국지능시스템학회논문지
• /
• 제27권2호
• /
• pp.113-118
• /
• 2017

선박 충돌 사고를 예방하기 위해 충돌위험도를 미리 추정하여 알려주는 다양한 연구들이 꾸준히 소개되고 있고, 일부 항해장비에 적용되고 있다. 하지만, 충돌위험을 미리 알려주었을 때, 실제 운항자는 충분히 피항 가능한 상황인데 위험성을 알린다고 판단하여 위험 경고를 무시하거나 장비의 알람 기능을 꺼놓는 경우도 많은 것으로 알려져 있다. 선박 충돌 위험 예측이 운항자들에게 좀 더 유용해지기 위해서는 실제 선박들의 습관화된 충돌 회피 동작을 고려할 필요가 있다. 이 연구는 선박에서의 충돌 회피 조치가 어떻게 이루어지고 있는지 조우 유형별로 분석하고 이력을 관리하기 위한 시스템을 제안한다. 특히, 충돌회피를 위한 초기대응에 대한 적절성을 판단하는 핵심 모듈을 뉴로-퍼지 기반 추론 형태로 상세히 설계하고 테스트함으로써 제안하는 시스템의 타당성을 보인다.

• 한국정보통신학회논문지
• /
• 제11권1호
• /
• pp.181-190
• /
• 2007

본 논문에서는 지능형 선박의 실시간 장애물회피를 위한 충돌회피시스템을 논한다. 다른 무인자율항체들의 충돌회피시스템과는 달리 지능형 선박의 충돌회피시스템은 목적지까지의 합리적이고 안전한 경로를 통한 충돌회피뿐만 아니라 해양을 항해하는 모든 선박이 준수해야하는 국제해상충돌예방규칙(COLRECs, International Regulations for Preventing Collisions at Sea)을 반영해야한다. 목적지까지의 합리적이고 안전한 경로를 통한 충돌회피라는 일반적인 충돌회피시스템의 목표를 달성하기 위해 BK-곱에 기반한 휴리스틱 탐색기법을 채용하며, 또 다른 목표인 COLREGs의 준수를 위해 충돌회피를 위한 규칙을 적용하는 지식기반시스템을 채용한다. 제안된 충돌회피시스템의 성능검증을 위해 COLREGs에 명시된 여러 가지 조우상황을 설정한 시나리오를 이용하여 최적성과 안정성 관점에서 시뮬레이션을 수행하고 그 결과를 $A^{\ast}$ 탐색기법과 비교한다. 시뮬레이션을 통해 제안된 충돌회피시스템이 지능형 선박의 실용적이고 효율적인 실시간 충돌회피시스템으로 적합함을 확인하였다.

• 한국항해학회지
• /
• 제15권2호
• /
• pp.13-38
• /
• 1991

This paper intends to develop a Prototype Expert Collision Avoidance System by introducing expert system techniques into the decision block of anti-collision loop. The problem domain of this study is characterized and specified by combining the concepts of anti-collision loop and knowledge -based system for collision avoidance. Domain in knowledge which may originates from the appropriate sources such as the International Regulations for Preventing Collision at Sea 1972, Marine Traffic Laws, and many texts on the subject of anticollision navigation and good seamanship is acquired and formalized into the knowledge-base system using production rule. Finally, a Prototype Expert Collision Avoidance System is built by using the CLIPS, developed by AIS NASA written in and fully integrated with the C language, and some test-and-run results of the system are demonstrated and examined. The author considers the proposed system which is named PECAS to be meaningful as a test bed for a further refined Expert Collision Avoidance System on board the Automated Ship.

• 대한조선학회논문집
• /
• 제60권6호
• /
• pp.406-415
• /
• 2023

In this study, we focused on developing and verifying ship collision avoidance algorithms using Unity simulator and ROS(Robot Operating System). ROS is used to establish an environment where communication between different operating systems is possible, and a dynamic model of a ship is constructed within Unity simulator. The Lidar data collected in Unity environment is passed to the system based on python through ROS. In the system based on python, control command values were created through the logic of the collision avoidance algorithm using data, and the values were transferred back to Unity to control the movement of the virtual ship. Through the developed simulation system, the reliability of the collision avoidance algorithm of ships with two different forms in an environment similar to the actual physical world was confirmed. As a result, it was confirmed on the simulator that it could be avoided without collision even in an environment with various types of obstacles, and that the avoidance characteristics according to the dynamics of the ship could be analyzed.

• 한국항해항만학회지
• /
• 제28권6호
• /
• pp.475-480
• /
• 2004

본 연구에서는, 충돌회피지원을 위한 자동제어에 관하여 새롭게 얻은 몇 가지 성과에 대하여 소개하려고 한다. 그 내용의 첫 번째는 그 동안의 기존 연구에서 해결되지 못하였던 집단 선박의 충돌 회피 이론이다. 이 이론은 선박이 무리를 지어 있는 어선 군을 만났거나, 혹은 무리를 지어 접근하는 선박군에 대한 충돌회피 동작에 유용한 알고리즘이다. 피항동작을 계산하는 단계에서 새로운 알고리즘을 적용한 모델을 적용하여 그 유용성을 설명하였다. 두 번째로는 충돌위험도 계산의 통합화 모델 제시이다. 선박의 경우 유지선, 피항선, 추월선, 피추월선 등 상대 선박과의 다양한 조우 상황에 따라, 항해사의 느끼는 위험도 그리고 피항 의무 등에 조금씩 차이가 있다. 이러한 현상을 반영하기 위하여, 충돌회피 동작에 사용하는 위험도 계산을 수행할 때, 조우상황에 따라 적절히 차별을 주는 모델을 제시하였다. 마지막으로는 이렇게 제시된 모델을 이용하여 다양한 상황에 따라 시뮬레이션을 통하여 검증해 보고, 그 유효성을 살펴보았다.