A New Approach for Tactical Decision Making in Lane Changing: Sample Efficient Deep Q Learning with a Safety Feedback Reward.
M. Ugur YavasN. Kemal UreTufan KumbasarPublished in: CoRR (2020)
Keyphrases
- decision making
- reinforcement learning
- data mining
- decision makers
- decision support
- learning algorithm
- cooperative
- real time
- state space
- relevance feedback
- user feedback
- lane detection
- multi agent reinforcement learning
- model free
- traffic flow
- function approximation
- sample size
- detection algorithm
- decision support system
- supply chain
- fuzzy logic
- neural network