Reinforcement learning-based collision avoidance: impact of reward function and knowledge transfer.
Xiongqing LiuYan JinPublished in: Artif. Intell. Eng. Des. Anal. Manuf. (2020)
Keyphrases
- knowledge transfer
- collision avoidance
- reward function
- reinforcement learning
- transfer learning
- reinforcement learning algorithms
- path planning
- state space
- markov decision processes
- optimal policy
- mobile robot
- dynamic environments
- knowledge sharing
- inverse reinforcement learning
- learning tasks
- function approximation
- model free
- transition model
- temporal difference
- multi agent
- machine learning
- state variables
- neural network
- fuzzy neural network
- multiple agents
- transition probabilities
- dynamic programming
- supervised learning
- generative model
- learning algorithm
- e learning
- path finding
- data mining