Open AccessResearch on vehicle adaptive cruise control based on BP neural network working condition recognition
Author(s) -
Wang Yongwei,
Guo Zhijun,
Wu Jingbo,
Fu Shenzhen
Publication year - 2022
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/tje2.12094
Subject(s) - artificial neural network , cruise control , computer science , control theory (sociology) , mode (computer interface) , matlab , fuzzy logic , division (mathematics) , cruise , algorithm , control (management) , artificial intelligence , engineering , mathematics , arithmetic , aerospace engineering , operating system
To solve the problems of the vehicle following model mismatch under different driving conditions and poor vehicle following performance, this paper proposes an adaptive cruise control system based on BP neural network to identify different driving conditions. Firstly, eight typical driving cycles are selected as samples, which are redivided into low‐speed mode and medium high‐speed mode. According to the selected characteristic parameters and re‐division of driving mode, a new BP neural network structure is designed. Secondly, the characteristic parameter values are calculated by using the characteristic parameters, and the characteristic parameters are normalized. The working condition recognition model is obtained by training the normalized data values. Finally, according to the classification of working conditions, the fuzzy control upper control algorithm and the MPC upper control algorithm for medium‐high speed conditions are designed respectively. Based on the above control algorithms, the joint simulation is carried out with Car‐Sim and MATLAB/Simulink. The simulation results show that the low‐speed fuzzy logic algorithm and the medium high‐speed MPC algorithm proposed in this paper can make the vehicle have a better following performance and comfort.