Adaptive neuro-fuzzy predictive control for design of adaptive cruise control system

Abstract

Proliferation of the number of vehicles is one of the main causes of traffic congestion, accidents, energy waste and environmental pollution. Recently, several intelligent applications are equipped in modern vehicles such as advanced driver assistance systems (ADAS), especially an adaptive cruise control (ACC) system which was successfully implemented on some luxury cars and still remains to be an interesting topic of research. An adaptive neuro-fuzzy predictive control (ANFPC) is proposed in designing of ACC system in this paper. By controlling the ACC vehicle through the throttle force or brakes, the ACC vehicle follows its predecessor and maintains the safe distance with the minimized tracking error. In the ANFPC scheme, a Takagi-Sugeno (T-S) fuzzy model is utilized to approximate the preceding vehicle model and then the predicted state sequence of the preceding vehicle is obtained. More importantly, the predictive control law is derived by a fuzzy neural networks (FNNs) approach. Simulation results demonstrate that the proposed ANFPC can achieve better performance than other methods in terms of safety, comfort and fuel consumption, simultaneously.