Name
Title | ||
|---|---|---|
Dynamic Path Planning Algorithm for Unmanned Ship Based on Deep Reinforcement Learning |
Abstract | ||
|---|---|---|
In order to enable the unmanned ship to have the ability of autonomous path planning in the complex marine environment, which can avoid obstacles and reach the destination accurately in the unknown environment, this paper combines the ability of deep learning to obtain information with the decision-making ability of reinforcement learning, and proposes an algorithm based on deep reinforcement learning. The simulation results show that after 3000 rounds of training in a given environment, the success rate of sailing to the end point in this environment is 100%. The path planning algorithm based on deep reinforcement learning performs well in the simulation experiment, achieves a very high accuracy after a large number of training times, and meets the needs of the operation and use of the unmanned ship on the water. This algorithm can be carried in the actual environment to realize the autonomous navigation of the unmanned ship on the water. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1007/978-981-19-1253-5_28 | Bio-Inspired Computing: Theories and Applications |
Keywords | DocType | ISSN |
Unmanned ship, Path planning, Deep learning, Reinforcement learning | Conference | 1865-0929 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
6 | ||