Specialized Deep Residual Policy Safe Reinforcement Learning-Based Controller for Complex and Continuous State-Action Spaces.
Ammar N. AbbasGeorgios C. ChasparisJohn D. KelleherPublished in: CoRR (2023)
Keyphrases
- continuous state
- action space
- reinforcement learning
- policy search
- state space
- control policies
- markov decision processes
- state and action spaces
- continuous state spaces
- real valued
- action selection
- state action
- state dependent
- partially observable markov decision processes
- optimal policy
- robot navigation
- finite state
- stochastic processes
- markov decision process
- control policy
- continuous action
- control system
- planning problems
- markov decision problems
- function approximation
- motion control
- initial state
- single agent
- function approximators
- reinforcement learning algorithms
- autonomous robots
- optimal control
- dynamic environments
- markov chain
- dynamic programming
- multi agent