Multi-objective deep reinforcement learning for crowd-aware robot navigation with dynamic human preference.
Guangran ChengYuanda WangLu DongWenzhe CaiChangyin SunPublished in: Neural Comput. Appl. (2023)
Keyphrases
- robot navigation
- multi objective
- reinforcement learning
- continuous state
- autonomous mobile robot
- autonomous robots
- landmark recognition
- evolutionary algorithm
- multi objective optimization
- scene understanding
- multiple objectives
- optimization algorithm
- dynamic environments
- real time stereo
- map building
- real time
- objective function
- markov decision processes
- multi criteria
- object recognition
- topological map
- three dimensional
- learning algorithm