HJB-RL: Initializing Reinforcement Learning with Optimal Control Policies Applied to Autonomous Drone Racing.
Keiko NagamiMac SchwagerPublished in: Robotics: Science and Systems (2021)
Keyphrases
- control policies
- reinforcement learning
- optimal control
- control policy
- optimal policy
- action space
- continuous state
- function approximation
- dynamic programming
- autonomous learning
- state space
- control problems
- reinforcement learning algorithms
- markov decision processes
- multi agent
- control system
- finite horizon
- rl algorithms
- motion control
- approximate dynamic programming
- model free
- state dependent
- markov decision process
- action selection
- decision making
- neural network