Subgoal-Driven Navigation in Dynamic Environments Using Attention-Based Deep Reinforcement Learning.
Jorge de HeuvelWeixian ShiXiangyu ZengMaren BennewitzPublished in: CoRR (2023)
Keyphrases
- dynamic environments
- reinforcement learning
- potential field
- autonomous navigation
- reinforcement learning agents
- reinforcement learning algorithms
- mobile robot
- path planning
- autonomous agents
- collision free
- changing environment
- single agent
- obstacle avoidance
- autonomous systems
- function approximation
- agent systems
- state space
- highly dynamic environments
- adaptive control
- real environment
- markov decision processes
- temporal difference
- reinforcement learning methods
- collision avoidance
- indoor environments
- partial order
- multi agent
- goal oriented
- robot navigation
- model free
- belief space
- augmented reality
- agent based systems
- optimal policy