A Fully Controllable Agent in the Path Planning using Goal-Conditioned Reinforcement Learning.
GyeongTaek LeePublished in: CoRR (2022)
Keyphrases
- path planning
- reinforcement learning
- dynamic environments
- mobile robot
- agent learns
- multi agent
- path planning algorithm
- navigation tasks
- collision avoidance
- collision free
- obstacle avoidance
- robot path planning
- motion planning
- action selection
- autonomous agents
- trajectory planning
- multiagent systems
- indoor environments
- multi agent systems
- optimal path
- mobile agents
- multi robot
- path planner
- potential field
- autonomous navigation
- dynamic and uncertain environments
- learning agent
- autonomous vehicles
- path finding
- decision making
- degrees of freedom
- reward signal
- software agents
- multiple robots
- intelligent agents
- markov decision processes
- unknown environments
- multiple agents
- state space
- dynamic programming