智能车辆车道保持系统中避障路径规划  

Collision-free path planning for lane keeping system of intelligent vehicle

作　　者：汪明磊[1] 陈无畏[1] 王家恩[1] 

机构地区：[1]合肥工业大学机械与汽车工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2014年第2期129-133,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(51075112)

摘　　要：文章针对车道保持系统在运行中存在障碍物的情况,基于多传感器对智能车辆的局部避障路径规划问题进行了研究。首先利用单目视觉传感器检测可通行道路的边界线,根据道路信息得到车辆行驶的目标路径并生成虚拟障碍物;然后利用激光测距雷达与视觉传感器相结合的信息探测道路中可能存在的障碍物信息;最后将传感器感知的实际障碍物与虚拟障碍物相结合,采用改进的向量场直方图方法进行局部避障路径规划。仿真实验结果表明该避障路径规划算法是有效的。Considering the possibility of obstacles existing on the road w hen the lane keeping system is utilized for the driver assistance ,the multisensory fusion based local collision-free path planning of the intelligent vehicle is studied .Firstly ,the traversable road boundaries are detected by using a monocu-lar vision sensor and a target path and virtual obstacles are generated by combining the detected boundaries and the road information consequently .Secondly ,the possible real obstacles existing on the road are detected by using the fusion information from laser ranging radar and vision sensors .Fi-nally ,the real obstacles detected by sensors are combined with virtual obstacles ,and a local collision-free path is planned by using improved vector field histogram （VFH） method .The results of simula-tion experiment indicate the validity of the proposed collision-free path planning method .

关 键 词：智能车辆 辅助驾驶 路径规划 信息融合 

分 类 号：U270.11[机械工程—车辆工程]