Socially aware robot navigation in crowds via deep reinforcement learning with resilient reward functions.
Xiaojun LuHanwool WooAngela FaragassoAtsushi YamashitaHajime AsamaPublished in: Adv. Robotics (2022)
Keyphrases
- robot navigation
- reward function
- reinforcement learning
- policy search
- continuous state
- reinforcement learning algorithms
- markov decision processes
- initially unknown
- state space
- multi agent
- autonomous mobile robot
- optimal policy
- partially observable
- autonomous robots
- inverse reinforcement learning
- scene understanding
- transition model
- function approximation
- multiple agents
- topological map
- markov decision process
- transition probabilities
- learning agent
- landmark recognition
- action space
- state variables
- generative model
- temporal difference
- markov decision problems
- state action
- machine learning
- dynamic programming
- robotic systems
- three dimensional