The correct recognition of driver’s braking intention is the foundation to realize the brake-by-wire system, and it is also the key to improving the energy recovery efficiency of electric vehicles with the hybrid braking system. However, the recognition speed and accuracy interact with each other, which increases the recognition difficulty. A fast recognition method is proposed based on partial brake pedal signals to improve the recognition speed and accuracy simultaneously. The brake pedal signal was decomposed into intrinsic mode function components with the whale optimization algorithm optimized variational mode decomposition (WOA-VMD), and the samples were extract from the components by the sample entropy. The effective features were selected with the grey relative analysis (GRA) algorithm. The braking intention recognition model was built with the whale optimization algorithm optimized support vector machine (WOA-SVM) algorithm and validated on the dSPACE semi-physical simulation platform. The results show that the noise in the signal was reduced effectively with the WOA-VMD algorithm, and the irrelevant features were filtered out effectively with the GRA algorithm, and the recognition speed and accuracy were improved significantly with the WOA-SVM algorithm. Compared with traditional algorithms, the accuracy increases by 12.67%, and the time is shortened to 0.223 s.
Fast Recognition Method of Driver Braking Intention Based on Partial Braking Pedal Signals
Int.J Automot. Technol.
International Journal of Automotive Technology ; 26 , 4 ; 1185-1197
2025-06-01
13 pages
Article (Journal)
Electronic Resource
English
Fast Recognition Method of Driver Braking Intention Based on Partial Braking Pedal Signals
Springer Verlag | 2025
|Driver braking intention recognition method with failure protection function
European Patent Office | 2020
|Automobile composite braking control method based on braking intention recognition
European Patent Office | 2023
|Personalized pedal feeling simulator based on braking style of driver
European Patent Office | 2023
|