北华大学学报（自然科学版）2025，Vol.26Issue(1)：99-106,8.DOI:10.11713/j.issn.1009-4822.2025.01.017
基于双目视觉传感器的前向碰撞预警算法改进
Improvement of Forward Collision Warning Algorithm Based on Binocular Vision Sensor
摘要
Abstract
The existing forward collision warning systems primarily utilize multiple millimeter-wave radars in combination or integrate millimeter-wave radar with visual sensors.However,these approaches often face challenges such as high costs and algorithmic limitations.After comparing the performance and application advantages and disadvantages of various sensors,a binocular vision sensor has been selected as the sensor for the forward collision warning system.This study integrates an improved Time to Collision(TTC)algorithm with Kalman filtering and utilizes the binocular vision sensor to compare the TTC values against an adaptive threshold to assess risk levels,thereby ensuring driving safety and reducing accident rates.Simulations of two different driving scenarios were conducted in a Matlab environment based on the improved algorithm.The results show that the TTC algorithm incorporating Kalman filtering significantly improves the collision time warning response timeliness and reliability compared to the traditional TTC algorithm.关键词
前向碰撞预警/双目视觉传感器/风险评估/改进TTC算法/Matlab仿真Key words
forward collision warning/binocular vision sensor/risk assessment/improved TTC algorithm/Matlab simulation分类
交通运输引用本文复制引用
孙建伟,乔彦超,金雅旋,郭淑清..基于双目视觉传感器的前向碰撞预警算法改进[J].北华大学学报（自然科学版）,2025,26(1):99-106,8.基金项目
吉林省教育厅科学技术研究项目(JJKH20230061KJ). （JJKH20230061KJ）
