Automatic Ship Collision Avoidance Decision-making Method via Improved PPO Algorithm
Shuhui Hao, Wei Guan, Zhewen Cui, Sheng Qu, Tingting Hu
Dalian Maritime University
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
In the future development of intelligent shipping, ensuring the safe navigation of unmanned surface vehicles (USVs) in mixed traffic scenarios has emerged as a critical challenge. Existing autonomous collision avoidance algorithms often exhibit insufficient International Regulations for Preventing Collisions at Sea (COLREGs) compliance and low decision-making efficiency in complex waterways. To address these limitations, this paper proposes a COLREGs-compliant USV collision avoidance decision-making method based on improved proximal policy optimization (IPPO) algorithm. By synergistically integrating Beta action distribution modeling, advantage function normalization and policy entropy, this method enhances the policy optimization capability and accelerates the convergence speed of PPO algorithm. Additionally, the ship domain and COLREGs are incorporated into the reward function design. Dynamic environment simulations validate the method’s effectiveness for USV navigation in complex, COLREGs-constrained maritime scenarios, while comparative analysis of the average reward curves reveals the theoretical advancements in PPO algorithm optimization.
逐年被引趋势
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
工程Maritime Navigation and Safety
Reinforcement Learning in Robotics · Robotic Path Planning Algorithms
参考文献 21
此处列出前 3 条
引用本文 3
按被引量排序,此处列出前 3 条