PathRL: An End-to-End Path Generation Method for Collision Avoidance via Deep Reinforcement Learning.
Wenhao YuJie PengQuecheng QiuHanyu WangLu ZhangJianmin JiPublished in: ICRA (2024)
Keyphrases
- end to end
- collision avoidance
- generation method
- reinforcement learning
- path finding
- collision free
- path planning
- mobile robot
- optimal path
- dynamic environments
- visual navigation
- congestion control
- admission control
- multipath
- formation control
- learning algorithm
- application layer
- ad hoc networks
- wireless ad hoc networks
- rate allocation
- heuristic search
- state space
- fuzzy neural network
- multi hop