Optimizing Trajectories for Highway Driving with Offline Reinforcement Learning.
Branka MirchevskaMoritz WerlingJoschka BoedeckerPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- traffic safety
- traffic accidents
- function approximation
- motion patterns
- learning algorithm
- state space
- model free
- machine learning
- real time
- moving objects
- temporal difference
- learning process
- reinforcement learning algorithms
- moving object trajectories
- reinforcement learning methods
- markov decision processes
- optimal policy
- dynamic programming
- learning tasks
- optimal control
- dynamical systems
- space time
- reward function
- multiple targets
- mobile robot
- phase space
- stochastic approximation
- control system
- trajectories of moving objects
- genetic algorithm