Adaptive Neuro-Fuzzy Predictor-Based Control for Cooperative Adaptive Cruise Control System

In this paper, an adaptive neuro-fuzzy predictor-based control (ANFPC) approach with integrated automotive radar and vehicle-to-vehicle (V2V) communication for the design of the cooperative adaptive cruise control (CACC) system is presented, which concerns not only the safety and riding comfort of a vehicle but also enhances its fuel efficiency. This paper consists of two main parts: preceding vehicle state estimation and following vehicle controller. First, the prospective information of the preceding vehicle, such as position, velocity, and acceleration, can be derived through radar sensor, and the control force of the preceding vehicle can be transmitted to the following vehicles through V2V communication. A Takagi-Sugeno fuzzy model is then utilized to estimate the preceding vehicle model, and the predicted state sequence of the preceding vehicle can be obtained. Second, based on these predicted data, the following vehicle is controlled by the proposed ANFPC scheme to maintain each vehicle within the desired distance headway and thus achieve string stability of vehicle platooning and fuel efficiency. The experimental results on the CarSim environment show that the proposed control strategy for the CACC system can significantly reduce the fuel consumption while ensuring driving comfort and safety.