The invention discloses a method for detecting the abnormal driving behavior of a vehicle. The method comprises S1, calculating the distance, the time interval, the speed, the acceleration, the direction and the turning angle of two adjacent position points; S2, performing feature extraction on the abnormal driving behavior of the vehicle; S3, when it is detected that the target exhibits an abnormal driving behavior, measuring the distance between driving behavior features at any two position points in a GPS trajectory ; S4, according to the information extracted in the step S1 and the distance between the data points obtained in the step S3, performing clustering analysis on each driving behavior in the step S2, and marking the result of a clustering cluster; S5: comparing the clusteringanalysis result obtained in the step S4 with a point in the GPS trajectory to be tested, determining whether the point is an abnormal point in driving behaviors including overspeed, sharp acceleration, sharp deceleration, sharp braking or frequent lane change, if so, obtaining the abnormal driving behavior of the vehicle according to the information of the abnormal point, otherwise, determining that the vehicle is normally driving. .