Frequent lane-change of vehicles will have a serious impact on traffic safety and traffic congestion. In the era of the Internet of Vehicles, it is helpful to regulate the driver's driving behavior by detecting frequent lane-change behaviors and promptly alerting the driver. In this paper, the accelerometer and gyroscope data collected by the smartphone inertial sensor are used to analyze the lane-change behavior of the vehicle. The data is trained by AdaBoost and combined with the sliding time window for frequency identification. In order to verify the effectiveness of the proposed algorithm, compared with SVM algorithm and BP neural network, the results show that AdaBoost algorithm is superior to SVM algorithm and BP neural network. Finally, the same experiment was conducted with the data collected by the high-precision inertial at the same time. Compared with the results of Adaboost algorithm, the practicability of the proposed method is further verified.

CICTP