Implementing reinforcement learning for on-demand vehicle rebalancing in MATSim.
Tarek ChouakiSebastian HörlJakob PuchingerPublished in: ANT/EDI40 (2022)
Keyphrases
- reinforcement learning
- function approximation
- vehicle detection
- pedestrian detection
- search tree
- reinforcement learning algorithms
- vehicle speed
- temporal difference learning
- cloud computing
- temporal difference
- binary search trees
- model free
- real time
- action selection
- autonomous vehicles
- transfer learning
- state space
- traffic monitoring
- multi agent reinforcement learning
- traffic accidents
- robotic control
- machine learning
- markov decision processes
- multi agent
- learning algorithm
- markov decision process
- license plate
- function approximators
- traffic congestion
- optimal policy
- traffic management
- outdoor environments
- learning process
- video sequences
- computer vision
- genetic algorithm
- transport systems