On hazards of using fundamental anti-windup technique for H2 state-space controllers with an explicit observer.
Claes OlssonPublished in: ECC (2003)
Keyphrases
- state space
- reinforcement learning
- heuristic search
- control system
- dynamic programming
- state variables
- markov decision processes
- optimal policy
- reinforcement learning algorithms
- dynamical systems
- neural network
- particle filter
- markov chain
- risk assessment
- search space
- machine learning
- real time
- optical flow
- control strategy
- information systems
- learning algorithm
- action space
- evolutionary robotics