MEERL: Maximum Experience Entropy Reinforcement Learning Method for Navigation and Control of Automated Vehicles.
Xin BiCaien WengPanpan TongBo ZhangZhichao LiPublished in: ITSC (2023)
Keyphrases
- real time
- reinforcement learning
- fully automatic
- similarity measure
- preprocessing
- mutual information
- significant improvement
- detection method
- pairwise
- objective function
- cost function
- high accuracy
- information entropy
- computational complexity
- mathematical model
- high precision
- fitted q iteration
- classification method
- edge detection
- neural network