基于模糊决策的车道偏离预警系统研究

Research on Lane Departure Warning System Based on Fuzzy Decision

为改善单一预警算法出现的误警率高、漏警率高、预警不及时等问题,避免车道偏离预警系统出现频繁报警、干扰驾驶员操作等情况,提高系统的抗干扰性和适应性,论文提出了一种基于多因素模糊模式识别的驾驶员意图分析方法。结合横向车辆偏离时间TLC(Time to Lane Crossing)和车辆目前位置CCP(Car Current Position)的车道偏离预警算法,同时考虑到驾驶员的驾驶特性,将驾驶员类型、车辆轮胎与车道边界线的距离、跨道时间作为影响车道偏离的因素并进行模糊化处理,设置合理的模糊规则,输出车道偏离程度的模糊评价指数。最后在Matlab中搭建控制模型,并与CARSIM做联合仿真来验证算法。仿真结果表明算法具有较好的实效性。

In order to improve the single early warning algorithm,such as high false alarm rate,high missing alarm rate and untimely early warning,avoid frequent alarm and interference of driver operation in lane departure warning system,and improve the anti-interference and adaptability of the system,this paper proposes a driver intention analysis method based on multi factor fuzzy pattern recognition.Combined with the lane departure warning algorithm based on time to lane crossing TLC(Time to Lane Crossing)and car current position CCP(Car Current Position),and considering the driving characteristics of drivers,the driver type,the distance between vehicle tires and lane boundary line,and the crossing time are taken as the influencing factors of lane departure.The reasonable fuzzy rules are set and the fuzzy evaluation index of lane departure degree is output.Finally,the control model is built in Matlab,and the algorithm is verified by joint simulation with CarSim.Simulation results show that the algorithm has good effectiveness.