A CONTROL SYSTEM FOR MOVEMENT AND DIVERGENCE OF UNMANNED SHIP ACCORDING TO COLREG

A variant of unmanned ship control system with obstacle avoidance in accordance with International Regulations for Preventing Collisions at Sea (COLREG) is proposed in the paper. The system consists of three subsystems: trajectory control subsystem, obstacle avoidance subsystem, course and speed control subsystem. The trajectory control subsystem generates a course that allows you to move along a predetermined trajectory. A gradient vector of the auxiliary function algorithm is used for this. In this algorithm, the alfa weight coefficient, which determines the degree of convergence of the unmanned ship to a given trajectory, is set. To ensure more efficient control convergence of the unmanned ship to a given trajectory, it is proposed to use the variable value of alfa, which depends on the condition of the coincidence of the unmanned ship course with the direction of the route and the value of alfa change for one iteration. The obstacle avoidance subsystem is based on the method of Velocity Obstacles, when all variants of the unmanned ship velocity vectors, at which a collision with an obstacle in the future is possible, are determined provided that the obstacle does not maneuver. The subsystem is activated when a dangerous situation occurs. The distance of the closest approach to the obstacle and the time of following to the point of the closest approach are calculated for identify a dangerous situation. Also, the subsystem implements a mechanism for determining the type of dangerous situation and compliance with COLREG. In total, four basic avoidance scenarios have been identified: head-on, crossing (obstacle moves to the left or right), and overtaking. To perform collision avoidance, the cost function of all variants of the course and speed is calculated, and the best variant is selected. The selected course and speed are transmitted to the course and speed control subsystem. The system is implemented in the Matlab / Simulink modeling environment. The simulation results confirmed the operability of the system, which ensures the movement of the unmanned ship along a given route and avoidance obstacles in case of dangerous situations in accordance with COLREG.

unmanned ship, COLREG, control system, algorithm, vector field, collision avoidance, trajectory, route, velocity obstacle, modeling

1. Fossen T. I. Handbook of marine craft hydrodynamics and motion control / T. I. Fossen. - John Wiley & Sons, 2011. - 582 p.

2. Дыда А. А. Подход к управлению судном по траектории на основе градиента вспомогательной функции / А. А. Дыда, К. Н. Чумакова, И. И. Пушкарев // Научные проблемы водного транспорта. - 2020. - № 65. - С. 27-36. DOI: 10.37890/jwt.vi65.125.

3. Дыда А. А. Алгоритм обхода статических препятствий для безэкипажного судна / А. А. Дыда, И. И. Пушкарев, К. Н. Чумакова // Вестник Государственного университета морского и речного флота имени адмирала С. О. Макарова. - 2021. - Т. 13. - № 3. - С. 307-315. DOI: 10.21821/2309-5180-2021-13-3-307-315.

4. Kuwata Y. Safe maritime navigation with COLREGS using velocity obstacles / Y. Kuwata, M. T. Wolf, D. Zarzhitsky, T. L. Huntsberger // 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems. - IEEE, 2011. - Pp. 4728-4734. DOI: 10.1109/IROS.2011.6094677.

5. Sawada R. Automatic ship collision avoidance using deep reinforcement learning with LSTM in continuous action spaces / R. Sawada, K. Sato, T. Majima // Journal of Marine Science and Technology. - 2021. - Vol. 26. - Is. 2. - Pp. 509-524. DOI: 10.1007/s00773-020-00755-0.

6. Shi J. Deep Learning in Unmanned Surface Vehicles Collision-Avoidance Pattern Based on AIS Big Data with Double GRU-RNN /j. Shi, Z. Liu // Journal of Marine Science and Engineering. - 2020. - Vol. 8. - Is. 9. - Pp. 682. DOI: 10.3390/jmse8090682.

7. Триполец О. Ю. Обзор существующих методов расхождения безэкипажных судов / О. Ю. Триполец // Вестник Государственного университета морского и речного флота имени адмирала С. О. Макарова. - 2021. - Т. 13. - № 4. - С. 480-495. DOI: 10.21821/2309-5180-2021-13-4-480-495.

8. Шарлай Г. Н. МППСС-72 с комментариями: учебное пособие / Г. Н. Шарлай. - М.: Моркнига, 2017. - 136 с.

9. Colito J. Autonomous mission planning and execution for unmanned surface vehicles in compliance with the marine rules of the road: thesis /j. Colito. - Washington: University of Washington, 2007. - 58 p.

10. Benjamin M. R. A method for protocol-based collision avoidance between autonomous marine surface craft / M. R. Benjamin, J. J. Leonard, J. A. Curcio, P. M. Newman // Journal of Field Robotics. - 2006. - Vol. 23. - Is. 5. - Pp. 333-346. DOI: 10.1002/rob.20121.

11. Fiorini P. Motion Planning in Dynamic Environments Using Velocity Obstacles / P. Fiorini, Z. Shiller // The International Journal of Robotics Research. - 1998. - Vol. 17. - Is. 7. - Pp. 760-772. DOI: 10.1177/ 027836499801700706.

12. Дыда А. А. Построение модели динамики безэкипажного судна по курсу на основе экспериментальных данных / А. А. Дыда, К. Н. Пляшешник, И. И. Пушкарев // Вестник Государственного университета морского и речного флота имени адмирала С. О. Макарова. - 2020. - Т. 12. - № 4. - С. 716-725. DOI: 10.21821/2309-5180-2020-12-4-716-725.