重庆理工大学学报2026，Vol.40Issue(9)：1-9,9.DOI:10.3969/j.issn.1674-8425(z).2026.05.001
基于卷积支持向量机的驾驶意图识别
Driving intention recognition based on convolutional support vector machine
施爱平 1周志 1徐泽琛 1丁礼君1
作者信息
- 1. 江苏大学汽车与交通工程学院,江苏镇江 212000
- 折叠
摘要
Abstract
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.关键词
驾驶意图识别/卷积支持向量机/灰狼优化算法Key words
driving intention recognition/convolutional SVM/GWO分类
信息技术与安全科学引用本文复制引用
施爱平,周志,徐泽琛,丁礼君..基于卷积支持向量机的驾驶意图识别[J].重庆理工大学学报,2026,40(9):1-9,9.
