Crowd Evacuation Guidance Based on Combined Action Reinforcement Learning.
Yiran XueRui WuJiafeng LiuXianglong TangPublished in: Algorithms (2021)
Keyphrases
- reinforcement learning
- action selection
- multi agent
- pedestrian dynamics
- partially observable domains
- state space
- cellular automata
- action space
- temporal difference
- reward shaping
- temporal difference learning
- reinforcement learning algorithms
- function approximation
- combining multiple
- function approximators
- simulation model
- transfer learning
- emergency evacuation
- learning process
- machine learning
- state action
- transition model
- markov decision processes