Application of fuzzy neural network based on T-S model for mobile robot to avoid obstacles

The problem of avoiding obstacles for mobile robot is quite difficulty, because work circumstance of the mobile robot is usually unknown. It was against this background that a study was undertaken with the specific aim of mobile robots reaching the destination without collision. A fuzzy neural network method based on Takagi-Sugeno(T-S) model was proposed to be used in the study. It not only has the advantage of fuzzy logic and neural network, but also has good self-study ability. The data collected by 8 ultrasonic sensors were classified firstly. Then the navigation algorithm based on T-S model was carried out. The test results show that the mobile robot using this fuzzy neural network can recognize the obstacles in all environment types, decide its action, and then arrive at destination after 231 seconds averagely. It is faster than the mobile robot using BP neural network which takes 239 seconds averagely.