Application of fuzzy neural networks in information fusion for mobile robot

摘要

An information fusion method of multi-sensor was proposed according to sensors used in mobile robots. The fuzzy logic and neural network were combined to construct fuzzy neural networks (FNN), and the calculating models of fuzzy neural networks were also put up. The mobile robot could recognize moving obstacles and environment types online, and could decide its action by using the information fusion method put forward, and thus realize mobile robot's movement without collision in dynamic environment. The training of network and test results indicated that the method which was used in avoiding moving obstacles of mobile robot is practicable and effective.