基于卷积支持向量机的驾驶意图识别OA

To address the feature extraction difficulties and low recognition accuracy in driving intention recognition,this paper proposes a driving intention recognition method based on CNN-SVM.It integrates the adaptive feature extraction function of convolutional neural networks and the super strong generalization classification performance of support vector machines(SVM),thus achieving higher accuracy in recognizing driving intentions.The feature parameters(vehicle speed,acceleration,pedal opening,and pedal opening change rate)are selected.CNN extracts data features,and then SVM classifies them.Grey Wolf Optimizer(GWO)algorithm is introduced to optimize the model parameters and improve the accuracy of driving intention recognition.To verify its effectiveness,the model is compared with three other models(CNN-SVM,CNN-LSTM,and GWO-LSTM).Results demonstrate it outperforms all of them.