Work

Construction of risk spectrum for vehicle stability state based on data-driven

Year: 2025

Type: article

Abstract: This paper constructs a data-driven vehicle stability risk spectrum to quantify safety risks and precisely identify stability states. Firstly, a dataset of multi-dimensional state attributes of vehicles is obtained through driving simulators. Then, based on the density-based spatial clustering of application with noise (DBSCAN) algorithm, vehicle states are clustered into stable, tend towards stable, tend towards unstable and unstable. Using clustered data to train a one-dimensional convolutional neural network (1D-CNN), a vehicle stability risk spectrum is constructed based on classification probability. Finally, using the Lyapunov index method, nonlinear system phase plane method, Carsim/Simulink simulation platform, and hardware-in-the-loop testing, typical stable states are tested and verified. The results show that the vehicle stability risk spectrum constructed can effectively predict the state of vehicles and calculate the safety and stability boundary. This paper aims to timely and accurately identify dangerous states of vehicles, provide a reference for active safety control system design.

Source: Measurement

DOI: 10.1016/j.measurement.2025.117013