What question did this study set out to answer?

To overcome low sample efficiency and data collection challenges in reinforcement learning for autonomous surface vessel path planning by implementing knowledge transfer.

July 31, 2025IEEE Transactions on Intelligent Transportation Systems

Efficient Dynamic Path Planning Algorithm for ASVs Based on Transfer Reinforcement Learning

View Full Paper

Ask AI

Bookmark

To overcome low sample efficiency and data collection challenges in reinforcement learning for autonomous surface vessel path planning by implementing knowledge transfer.
Constructed a state similarity model to identify and transfer collision avoidance experiences compliant with International Regulations for Preventing Collisions at Sea (COLREGs).
Designed a progressive exploration method based on source-to-target sample pool proportions to prevent negative transfer.
Tested the algorithm across five simulated collision avoidance scenarios and verified real-time performance through physical ship trials.
Achieved an average collision avoidance success rate of 91% across five simulated scenarios, exceeding baseline DDPG by over 30%.
Increased reward value convergence speed by 50% after 1,000 training episodes.
Attained a 100% path safety compliance rate during real-time collision avoidance in physical ship tests.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Experimental study demonstrates superior collision avoidance and faster learning in autonomous surface vessels, indicating enhanced navigation safety.

Cite This Study

Xu et al. (2025) studied this question.

synapsesocial.com/papers/6a82d093aa66c98ddba0d35f https://doi.org/10.1109/tits.2025.3590732

View Full Paper

Ask AI

Bookmark

Implication

Experimental study demonstrates superior collision avoidance and faster learning in autonomous surface vessels, indicating enhanced navigation safety.

Efficient Dynamic Path Planning Algorithm for ASVs Based on Transfer Reinforcement Learning

Key Points

Discussion

Implication

Cite This Study

Efficient Dynamic Path Planning Algorithm for ASVs Based on Transfer Reinforcement Learning

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider