Amortized Q-learning with Model-based Action Proposals for Autonomous Driving on Highways.
Branka MirchevskaMaria HügleGabriel KalweitMoritz WerlingJoschka BoedeckerPublished in: CoRR (2020)
Keyphrases
- autonomous driving
- action selection
- model free
- grand challenge
- reinforcement learning
- stereo vision
- multi agent
- cooperative
- state action
- function approximation
- urban traffic
- vision algorithms
- worst case
- learning algorithm
- traffic flow
- state space
- action space
- search tree
- computer vision systems
- computer vision algorithms
- optimal policy
- dynamic programming
- image sequences
- computer vision