多感知融合的后装汽车防碰撞系统研究
Research on Rear-Mounted Vehicle Collision Prevention System Based on Multi-Sensing Fusion
DOI: 10.12677/csa.2025.155123, PDF,    科研立项经费支持
作者: 崔文申, 周忠圆, 崔行航, 韩 兴:东北林业大学土木与交通学院,黑龙江 哈尔滨;李 博:东北林业大学机电工程学院,黑龙江 哈尔滨
关键词: 后装防碰撞系统多传感器融合复合型安全模型分级制动Rear Installation Anti-Collision System Multi-Sensor Fusion Compound Security Model Stage Braking
摘要: 设计了一种多感知融合的外加装汽车防碰撞装置。针对当前存量车辆主动安全配置的普遍缺失,导致复杂交通场景下碰撞风险居高不下。本文提出一种基于多传感器融合的后装防碰撞系统,通过有限度的车辆改装实现主动安全功能升级。系统采用毫米波雷达与视觉传感器的异构数据互补策略,在Haar-like特征检测框架中引入卡尔曼滤波算法,并构建碰撞时间和安全车距的复合型安全模型。硬件层面设计非侵入式改装方案:毫米波雷达嵌入前保险杠非承力区域,视觉模块利用后视镜基座实现无损安装,控制单元通过车载诊断接口(OBD-II)获取车辆状态信息,并经由CAN总线协议触发分级制动响应。研究成果为汽车后市场安全升级提供了兼具工程可行性与经济性的技术路径。
Abstract: An external anti-collision device with multi-sensing fusion was designed. In view of the general lack of active safety configuration of current vehicles, the collision risk remains high in complex traffic scenarios. A rear-mounted anti-collision system based on multi-sensor fusion is proposed in this paper. Active safety functions can be upgraded through limited vehicle modification. The system adopts the heterogeneous data complementary strategy of millimeter wave radar and vision sensor, introduces Kalman filter algorithm into the Haar-like feature detection framework, and constructs the composite safety model of collision time and safety distance. Hardware level design of non-invasive modification scheme: millimeter wave radar embedded in the front bumper non-bearing area, vision module using the rearview mirror base to achieve non-destructive installation, control unit through the on-board diagnostic interface (OBD-II) to obtain vehicle status information, and through the CAN bus protocol to trigger a staged braking response. The research results provide a technical path of engineering feasibility and economy for the safety upgrade of automotive aftermarket.
文章引用:崔文申, 周忠圆, 崔行航, 韩兴, 李博. 多感知融合的后装汽车防碰撞系统研究[J]. 计算机科学与应用, 2025, 15(5): 514-523. https://doi.org/10.12677/csa.2025.155123

参考文献

[1] 胡远志, 吕章洁, 刘西. 基于PreScan的AEB系统纵向避撞算法及仿真验证[J]. 汽车安全与节能学报, 2017, 8(2): 136-142.
[2] 张庆. 基于多传感器数据融合的乘用车AEB控制策略研究[D]: [硕士学位论文]. 长春: 吉林大学, 2019.
[3] Xu, G., Xiong, Y., Niu, H., Yu, G. and Zhou, B. (2021) Collision-Warning System Integrated with Merging Behaviour Prediction Model Based on Multi-Sensor Fusion. International Journal of Vehicle Design, 86, 143-161. [Google Scholar] [CrossRef
[4] 辛佳庚, 杨复钰, 张宝迪, 等. 基于融合算法的电动汽车AEB控制策略[J]. 北京交通大学学报, 2021, 45(6): 77-86+93.
[5] 王梦缘, 惠飞, 薛硕硕, 等. 基于AEB的车辆防碰撞控制及仿真测试[C]//中国科学技术协会, 交通运输部, 中国工程院, 湖北省人民政府. 2023世界交通运输大会(WTC2023)论文集(下册). 2023: 232-239.
[6] 胡永坤, 殷晨波, 乔文华, 等. 基于激光雷达和多传感器数据融合的塔机防碰撞系统[J]. 机械设计与制造工程, 2024, 53(9): 117-121.
[7] Vinoth, K. and Sasikumar, P. (2024) Multi-Sensor Fusion and Segmentation for Autonomous Vehicle Multi-Object Tracking Using Deep Q Networks. Scientific Reports, 14, Article No. 31130. [Google Scholar] [CrossRef] [PubMed]
[8] 唐启涛, 戴小鹏, 罗莉霞. 卡尔曼滤波下多源传感器数据互补-加权迭代融合算法[J]. 传感技术学报, 2024, 37(8): 1460-1465.