抄録
This paper describes a new steering control system for an autonomous road vehicle using the fuzzy neural network algorithm. Most of conventional fuzzy controls are represented by linguistic rules based on an expert's or engineer's control knowledge for a system. But the expert's knowledge is not so clear that it is difficult to represent the characteristics of the control system by these linguistic rules perfectly.
So, this paper proposes a fuzzy neural network control algorithm which includes both the advantages of the fuzzy logic and the neural network to overcome the disadvantages of them. The FCM (Fuzzy C-means Method) clustering algorithm is applied to create the optimal control rules with data measured in off-line control mode. These rules are learned by the error back propagation learning algorithm.
A computer simulator is developed to compare the performance of the proposed fuzzy neural network with those of the conventional fuzzy logic and the CMAC (Cerebellar Model Articulation Controller) neural network for steering control of the autonomous road vehicle. The proposed FNNC is found to be superior in all aspects by computer simulations.
