Yao, ZhiMu (Dalian Minzu University), Tao, Zhang (Dalian Minzu University), Ge, Pingshu (Dalian Minzu University), Wang, Yang (Dalian Minzu University), Liu, Junjie (Dalian Minzu University)

Research on Active Collision Avoidance Control Strategy for Intelligent Vehicle

Scheduled for presentation during the Invited session "New Advances in Automotive Control Technology" (ThB2), Thursday, October 31, 2024, 14:30−14:50, Room T2

7th IFAC Conference on Engine and Powertrain Control, Simulation and Modeling, Oct 30 - Nov 1, 2024, Dalian, China

This information is tentative and subject to change. Compiled on May 16, 2025

Abstract

To address the limitations of existing vehicle collision avoidance strategies that rely only on braking or lane-changing, a new system using both lateral and longitudinal methods is proposed. Using the reciprocal of the time-to-collision (TTC-1) as an indicator to evaluate collision risk, an improved safety distance model and a braking risk coefficient are introduced to enhance the control strategy.For lanechanging, the k-means clustering algorithm is used to analyze vehicle state parameters which can enhance precision and stability. The model predictive control (MPC) is improved with a variable prediction horizon and online lateral stability recognition. By comparing the current vehicle state to cluster centroids, the system determines real-time lateral risk levels. A fuzzy algorithm is proposed to adjust the prediction horizon which can achieve adaptive control. Finally, the Carsim/Simulink co-simulation model is built to conduct collision avoidance simulation tests, and the effectiveness of the proposed strategy is verified.