Self-learning fuzzy logic controllers for pursuit–evasion differential games
- Resource Type
- Authors
- Howard M. Schwartz; Sameh F. Desouky
- Source
- Robotics and Autonomous Systems. 59:22-33
- Subject
- Adaptive neuro fuzzy inference system
Fuzzy classification
Training set
Artificial neural network
Neuro-fuzzy
Computer science
business.industry
General Mathematics
Fuzzy control system
Machine learning
computer.software_genre
Fuzzy logic
Defuzzification
Computer Science Applications
Control and Systems Engineering
Genetic algorithm
Fuzzy mathematics
Reinforcement learning
Fuzzy set operations
Fuzzy number
Fuzzy associative matrix
Artificial intelligence
business
computer
Software
- Language
- ISSN
- 0921-8890
This paper addresses the problem of tuning the input and the output parameters of a fuzzy logic controller. The system learns autonomously without supervision or a priori training data. Two novel techniques are proposed. The first technique combines Q(@l)-learning with function approximation (fuzzy inference system) to tune the parameters of a fuzzy logic controller operating in continuous state and action spaces. The second technique combines Q(@l)-learning with genetic algorithms to tune the parameters of a fuzzy logic controller in the discrete state and action spaces. The proposed techniques are applied to different pursuit-evasion differential games. The proposed techniques are compared with the classical control strategy, Q(@l)-learning only, reward-based genetic algorithms learning, and with the technique proposed by Dai et al. (2005) [19] in which a neural network is used as a function approximation for Q-learning. Computer simulations show the usefulness of the proposed techniques.