Obtaining Robust Control and Navigation Policies for Multi-Robot Navigation via Deep Reinforcement Learning.
Christian JestelHartmut SurmannJonas StenzelOliver UrbannMarius BrehlerPublished in: CoRR (2022)
Keyphrases
- robot navigation
- autonomous robots
- reinforcement learning
- continuous state
- robot control
- control policy
- autonomous mobile robot
- optimal policy
- control policies
- policy search
- function approximation
- scene understanding
- initially unknown
- control system
- robotic systems
- optimal control
- mobile robot
- landmark recognition
- topological map
- unknown environments
- map building
- state space
- robot localization
- real time
- autonomous vehicles
- markov decision process
- partially observable
- reinforcement learning algorithms
- temporal difference
- markov decision processes