基于自适应障碍物识别的汽车主动防撞系统  

Study of Vehicle Initiative Anti-Collision System Based on Adaptive Obstacle Reorganization

作　　者：解云[1] 徐彬[1] XIE Yun;XU Bin(Hefei Technology College,Anhui Hefei 238000,China)

机构地区：[1]合肥职业技术学院,安徽合肥238000

出　　处：《机械设计与制造》2018年第4期165-167,171,共4页Machinery Design & Manufacture

基　　金：汽车检测与维修技术省级教学团队(2015jxtd113)

摘　　要：为了保障驾驶安全,设计了基于自适应障碍物识别和目标跟踪的汽车防撞系统。根据激光雷达工作原理,提出了自适应阈值的最近邻聚类算法用于障碍物识别;根据城市交通实际状况,提出了基于当前统计模型的自适应Kalman滤波跟踪方法;为了提高驾驶舒适度,提出了融入驾驶员习惯的预瞄安全距离模型;对车体进行改装后实验,结果表明的算法能够快速跟踪移动目标,并且具有很高的跟踪精度;在防碰撞试验中,设计的主动防撞系统能够在安全距离及时制动车辆,说明了防撞系统的安全可靠性。To improve the driving safety,automobile collision avoidance system based on adaptive obstacle recognition and target tracking is designed.Depend on working principle of laser radar,the nearest neighbor clustering algorithm with adaptive threshold used to recognize obstacle.Rely on fact of city traffic,adaptive Kalman filter based on current statistical model is put forward to track the goal.To improve comfort level of driver,driver previewer safety distance model integrated with driving habit is raised.A car is refitted to carry out the trial,the result shows that tracking algorithm can convergence quickly with high precision,which can satisfy the use requirement.In the collision requirement,the collision system can stop the vehicle at safe distance in time,which clarifies safe reliability of the collision system.

关 键 词：汽车 主动防撞系统 自适应阈值最近邻聚类算法 当前统计模型 预瞄安全距离模型 

分 类 号：TH16[机械工程—机械制造及自动化] U463[机械工程—车辆工程]