Safe reinforcement learning control for continuous-time nonlinear systems without a backup controller.
Soutrik BandyopadhyayShubhendu BhasinPublished in: CoRR (2022)
Keyphrases
- nonlinear systems
- adaptive control
- tracking control
- optimal control
- reinforcement learning
- controller design
- control law
- inverted pendulum
- adaptive neural
- lyapunov function
- control problems
- adaptive fuzzy
- control method
- stability analysis
- control system
- fuzzy control
- fuzzy controller
- feedback control
- dead zone
- control strategy
- tracking error
- fuzzy model
- takagi sugeno
- fuzzy systems
- state space
- variable structure
- control strategies
- fuzzy logic controller
- adaptive neural control
- control algorithm
- dynamic programming
- sliding mode control
- learning rate
- neural network
- control policy
- real time
- pid controller
- lyapunov stability
- optimal policy
- dynamical systems
- closed loop
- control scheme
- input output
- dynamic environments
- genetic programming
- pi controller
- markov decision processes
- sliding mode
- learning algorithm
- chaotic systems