Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship
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....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-01-01
|
| Series: | International Journal of Naval Architecture and Ocean Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2092678223000687 |
| _version_ | 1797350617154322432 |
|---|---|
| author | Do-Hyun Chun Myung-Il Roh Hye-Won Lee Donghun Yu |
| author_facet | Do-Hyun Chun Myung-Il Roh Hye-Won Lee Donghun Yu |
| author_sort | Do-Hyun Chun |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-08T12:47:28Z |
| format | Article |
| id | doaj.art-92c7b85c4d8345b1b6096bb5abb6ccfb |
| institution | Directory Open Access Journal |
| issn | 2092-6782 |
| language | English |
| last_indexed | 2024-03-08T12:47:28Z |
| publishDate | 2024-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Naval Architecture and Ocean Engineering |
| spelling | doaj.art-92c7b85c4d8345b1b6096bb5abb6ccfb2024-01-21T05:06:47ZengElsevierInternational Journal of Naval Architecture and Ocean Engineering2092-67822024-01-0116100579Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous shipDo-Hyun Chun0Myung-Il Roh1Hye-Won Lee2Donghun Yu3Department of Naval Architecture and Ocean Engineering, Seoul National University, Seoul, Republic of KoreaDepartment of Naval Architecture and Ocean Engineering, and Research Institute of Marine Systems Engineering, Seoul National University, Seoul, Republic of Korea; Corresponding author.Division of Naval Architecture and Ocean Systems Engineering, Korea Maritime & Ocean University, Busan, Republic of KoreaDepartment of Naval Architecture and Ocean Engineering, Seoul National University, Seoul, Republic of KoreaIn 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.http://www.sciencedirect.com/science/article/pii/S2092678223000687Collision avoidanceCollision riskCOLREGsDeep reinforcement learningAutonomous ship |
| spellingShingle | Do-Hyun Chun Myung-Il Roh Hye-Won Lee Donghun Yu Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship International Journal of Naval Architecture and Ocean Engineering Collision avoidance Collision risk COLREGs Deep reinforcement learning Autonomous ship |
| title | Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship |
| title_full | Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship |
| title_fullStr | Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship |
| title_full_unstemmed | Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship |
| title_short | Method for collision avoidance based on deep reinforcement learning with path-speed control for an autonomous ship |
| title_sort | method for collision avoidance based on deep reinforcement learning with path speed control for an autonomous ship |
| topic | Collision avoidance Collision risk COLREGs Deep reinforcement learning Autonomous ship |
| url | http://www.sciencedirect.com/science/article/pii/S2092678223000687 |
| work_keys_str_mv | AT dohyunchun methodforcollisionavoidancebasedondeepreinforcementlearningwithpathspeedcontrolforanautonomousship AT myungilroh methodforcollisionavoidancebasedondeepreinforcementlearningwithpathspeedcontrolforanautonomousship AT hyewonlee methodforcollisionavoidancebasedondeepreinforcementlearningwithpathspeedcontrolforanautonomousship AT donghunyu methodforcollisionavoidancebasedondeepreinforcementlearningwithpathspeedcontrolforanautonomousship |