Characterising the Robustness of Reinforcement Learning for Continuous Control using Disturbance Injection.
Catherine R. GlossopJacopo PaneratiAmrit KrishnanZhaocong YuanAngela P. SchoelligPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- control problems
- optimal control
- diesel engine
- control system
- adaptive control
- multi agent
- robot control
- control method
- control policy
- function approximation
- data acquisition
- state space
- control strategy
- markov decision processes
- control strategies
- data sets
- action selection
- reinforcement learning algorithms
- continuous domains
- optimal policy
- learning algorithm
- traffic signal
- continuous state spaces
- strong robustness
- neural network