A GA-based learning algorithm for the learning of fuzzy behaviour of a mobile robot reactive control system
Jiancheng Qiu · 1997
This paper introduces a learning algorithm to automatically learn fuzzy membership functions of fuzzy behaviours for a mobile robot reactive control system. Genetic algorithms are used to implement the learning processes. The learning methodology emphasises the functionalities of individual behaviours, various types of learning environments and a simple-to-complex multistage learning course. The genetic algorithms are designed to support the learning processes through the effective exploration of population and the exploitation of learning environments. The simulation results are discussed to show the effectiveness of the learning algorithm.