To realize accurate environment perception, which is the technological key to enabling autonomous vehicles to interact with their external environments, it is primarily necessary to solve the issues of object detection and tracking in the vehicle-movement process. Multi-sensor fusion has become an essential process in efforts to overcome the shortcomings of individual sensor types and improve the efficiency and reliability of autonomous vehicles. This paper puts forward moving object detection and tracking methods based on LiDAR—camera fusion. Operating based on the calibration of the camera and LiDAR technology, this paper uses YOLO and PointPillars network models to perform object detection based on image and point cloud data. Then, a target box intersection-over-union (IoU) matching strategy, based on center-point distance probability and the improved Dempster–Shafer (D–S) theory, is used to perform class confidence fusion to obtain the final fusion detection result. In the process of moving object tracking, the DeepSORT algorithm is improved to address the issue of identity switching resulting from dynamic objects re-emerging after occlusion. An unscented Kalman filter is utilized to accurately predict the motion state of nonlinear objects, and object motion information is added to the IoU matching module to improve the matching accuracy in the data association process. Through self-collected data verification, the performances of fusion detection and tracking are judged to be significantly better than those of a single sensor. The evaluation indexes of the improved DeepSORT algorithm are 66% for MOTA and 79% for MOTP, which are, respectively, 10% and 5% higher than those of the original DeepSORT algorithm. The improved DeepSORT algorithm effectively solves the problem of tracking instability caused by the occlusion of moving objects.

The continuous increase in the penetration rate of autonomous vehicles in highway traffic flow has become an irreversible development trend; in this paper, a novel hybrid prediction model of deep sequence learning and an integrated decision tree is proposed for human–machine mixed driving heterogeneous traffic flow scenarios, so as to realize the accurate prediction of the driving intention of the target vehicle in the traffic environment by autonomous vehicles (AVs). Firstly, the hybrid model uses the attention mechanism-based double-layer gated network model (Bilayer-GRU-Att) to effectively capture the time sequence dependence of the target vehicle’s driving state, and then accurately calculate its trajectory data in different prediction time-domains (t_pred). Furthermore, the hybrid model introduces the eXtreme Gradient Boosting decision tree optimized by the Grey Wolf Optimization model (GWO-XGBoost) to identify the lane-changing intention of the target vehicle, because the prediction information of the future trajectory data of the target vehicle by the aforementioned Bilayer-GRU-Att model is properly integrated. The GWO-XGBoost model can accurately predict the lane-changing intention of the target vehicle in different prediction time-domains. Finally, the efficacy of this hybrid model was tested using the HighD dataset for training, validation, and testing purposes. The results of a benchmark analysis indicate that the hybrid model proposed in this paper has the best error evaluation index and balanced prediction time consuming index under the six prediction time-domains. Meanwhile, the hybrid model demonstrates the best classifying performance in predicting the lane-changing intentions of “turning left”, “going straight”, and “turning right” driving behaviors.

Distracted and fatigue-related driving behaviors are major causes of road traffic accidents, creating an urgent need for reliable driver monitoring systems. Vision-based detection methods have garnered widespread attention due to their low cost of deployment and practical applicability. However, existing lightweight models often suffer from limited global contextual perception and insufficient preservation of fine details. Motivated by these challenges, this study introduces an improved distracted and fatigue-related driving behavior detection model, YOLOv12-LAD, built on the YOLOv12 architecture. The proposed framework integrates a Large Separable Kernel Attention module (LSKA) to enhance global contextual perception, an Adaptive Downsampling module (ADown) to mitigate information loss during feature compression, and a Dynamic Sampling module (DySample) to enable content-adaptive feature reconstruction and improve multi-scale behavior representation. Experimental results show that YOLOv12-LAD achieved 97.5% precision, 96.3% recall, and 98.4% mAP@50 with only 2.5 million parameters, 6.2 GFLOPs, and an inference speed of 249 FPS. Ablation studies, comparisons with representative models, cross-dataset evaluation, and real-vehicle tests further verify the effectiveness and robustness of the proposed method. The proposed method demonstrates strong performance while maintaining computational efficiency, making it suitable for real-time vision-based driver monitoring applications.