A method for judging abnormal driving behaviors in diversion areas based on group intelligence perception of internet of vehicles

With the gradual increase of car ownership, the road traffic environment in diversion areas has become more complicated, resulting in the increase of abnormal driving behaviors and the high frequency of traffic accidents. Therefore, it is necessary to accurately determine driving behaviors to reduce the frequency of traffic accidents. In order to accurately determine the driving behaviors, a method for determining the abnormal driving behaviors in diversion areas based on group intelligence perception of network of vehicles is proposed. Firstly, the group intelligence perception data collection method is used to obtain driving behavior data, and a elliptic filter is used to eliminate noises in the driving behavior data. Secondly, a sliding window is introduced to extract driving behavior characteristics and typical driving behavior characteristics, and an abnormal driving behavior judgment model is established based on support vector mechanism. Finally, the model is trained based on sample data to determine the optimal parameters and complete the judgment of abnormal driving behaviors. Experimental results show that the minimum error frequency of driving behavior judgment obtained by the proposed method is 2.8%, which fully proves that the proposed method can accurately judge driving behaviors.