The invention discloses a multi-mode perception fusion vehicle lane change prediction method, computer equipment and a storage medium. The prediction method comprises the following steps: S1, synchronously obtaining feature information of a target vehicle and a sequence image collected by a front-view camera of the vehicle; s2, inputting the sequence image into an image feature extraction network so as to obtain image features of the sequence image; s3, splicing the feature information of the target vehicle and the image features obtained in the S2, and inputting the spliced feature information into a feature fusion network based on an attention mechanism to obtain fused features; and S4, inputting the fused features into a pre-trained lane changing prediction model so as to obtain the lane changing intention of the target vehicle. According to the prediction method, feature fusion is carried out by using an attention mechanism method, and the accuracy of vehicle lane change prediction can be effectively improved.
本发明公开了多模感知融合的车辆变道预测方法、计算机设备及存储介质,该预测方法包括以下步骤:S1:同步获取目标车辆的特征信息和本车前视摄像头采集的序列图像;S2:将序列图像输入图像特征提取网络中,从而获得序列图像的图像特征;S3:将目标车辆的特征信息和S2得到的图像特征进行拼接,然后输入基于注意力机制的特征融合网络中,从而得到融合后特征;S4:将融合后特征输入预先训练好的变道预测模型中,从而得到目标车辆的变道意图。该预测方法使用注意力机制方法进行特征融合,能有效提高车辆变道预测的准确性。
Multi-mode sensing fusion vehicle lane change prediction method, computer equipment and storage medium
多模感知融合的车辆变道预测方法、计算机设备及存储介质
2022-11-01
Patent
Electronic Resource
Chinese
Lane change prediction method and device, electronic equipment and storage medium
European Patent Office | 2024
|European Patent Office | 2023
|Vehicle lane change prompting method, device, equipment, medium and computer program product
European Patent Office | 2023
|Vehicle lane change detection method and device, electronic equipment, vehicle and storage medium
European Patent Office | 2024
European Patent Office | 2024
|