Autonomous emergency braking based on radial basis function neural network variable structure control for collision avoidance

• Published in: ITNEC : 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference : 15-17 December 2017 pp. 378 - 383
• Main Authors: He, Xiangkun, Ji, Xuewu, Yang, Kaiming, Wu, Jian
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2017
• Subjects: AEB; collision avoidance; Frequency modulation; Geophysical measurement techniques; Ground penetrating radar; neural network; Noise measurement; variable structure control
• DOI: 10.1109/ITNEC.2017.8284756

Autonomous emergency braking (AEB) control is one of important vehicle intelligent safety technologies to avoid collision. This paper presents an emergency rear-end collision avoidance control strategy using hierarchical control framework which consists of threat assessment layer, tire slip ratio control layer. The threat assessment layer continuously calculates threat metrics associated with collision avoidance by braking control. As for the tire slip ratio control layer, a radial basis function neural network (RBFNN) variable structure control (VSC) algorithm is designed to track optimal slip ratio, which enabled the controlled vehicle to generate the highest possible deceleration. Finally, simulation test is conducted via MATLAB/Simulink platform on dry and wet asphalt pavement at high speed. The results show that the proposed AEB control scheme effectively performs collision avoidance maneuvers.