Preceding vehicles have a significant impact on the safety of the vehicle, whether or not it has the same driving direction as an ego-vehicle. Reliable trajectory prediction of preceding vehicles is crucial for making safer planning. In this paper, we propose a framework for trajectory prediction of preceding target vehicles in an urban scenario using multi-sensor fusion. First, the preceding target vehicles historical trajectory is acquired using LIDAR, camera, and combined inertial navigation system fusion in the dynamic scene. Next, the Savitzky–Golay filter is taken to smooth the vehicle trajectory. Then, two transformer-based networks are built to predict preceding target vehicles’ future trajectory, which are the traditional transformer and the cluster-based transformer. In a traditional transformer, preceding target vehicles trajectories are predicted using velocities in the X-axis and Y-axis. In the cluster-based transformer, the k-means algorithm and transformer are combined to predict trajectory in a high-dimensional space based on classification. Driving data from the real-world environment in Wuhan, China, are collected to train and validate the proposed preceding target vehicles trajectory prediction algorithm in the experiments. The result of the performance analysis confirms that the proposed two transformers methods can effectively predict the trajectory using multi-sensor fusion and cluster-based transformer method can achieve better performance than the traditional transformer.

A core issue inherent to decision‐making and path‐planning tasks is managing the uncertainties in the motion of dynamic obstacles. Therefore, this article proposes a new decision‐making and path‐planning framework, based on game theory, that considers the multistate future actions of surrounding vehicles. First, multistate future actions of neighboring vehicles, whose driving styles vary, are estimated and fed into a decision‐making module for risk assessment. Then, based on the Stackelberg game theory, the ego vehicle and the rear object vehicle are modeled as two players in the game, and their optimal decisions are obtained. In addition, the path‐planning model incorporates a potential‐field model that utilizes several potential functions to explain the varied styles and physical limitations of the surrounding vehicles. Finally, the trajectory of the ego vehicle is obtained through model predictive control that is based on the outputs of the decision‐making and constructed potential‐field models. The results of simulation experiments that used designed scenarios demonstrate that the proposed method effectively manages various social interactions and generates safe and appropriate trajectories for autonomous vehicles. In addition, the simulation results demonstrate that considering multistate trajectories caused the decision‐making and path‐planning modules to be appropriate for unpredictable environmental conditions.

The deep convolutional neural network has led the trend of vision-based road detection, however, obtaining a full road area despite the occlusion from monocular vision remains challenging due to the dynamic scenes in autonomous driving. Inferring the occluded road area requires a comprehensive understanding of the geometry and the semantics of the visible scene. To this end, we create a small but effective dataset based on the KITTI dataset named KITTI-OFRS (KITTI-occlusion-free road segmentation) dataset and propose a lightweight and efficient, fully convolutional neural network called OFRSNet (occlusion-free road segmentation network) that learns to predict occluded portions of the road in the semantic domain by looking around foreground objects and visible road layout. In particular, the global context module is used to build up the down-sampling and joint context up-sampling block in our network, which promotes the performance of the network. Moreover, a spatially-weighted cross-entropy loss is designed to significantly increases the accuracy of this task. Extensive experiments on different datasets verify the effectiveness of the proposed approach, and comparisons with current excellent methods show that the proposed method outperforms the baseline models by obtaining a better trade-off between accuracy and runtime, which makes our approach is able to be applied to autonomous vehicles in real-time.

To improve the accuracy and robustness of autonomous vehicle localization in a complex environment, this paper proposes a multi-source fusion localization method that integrates GPS, laser SLAM, and an odometer model. Firstly, fuzzy rules are constructed to accurately analyze the in-vehicle localization deviation and confidence factor to improve the initial fusion localization accuracy. Then, an odometer model for obtaining the projected localization trajectory is constructed. Considering the high accuracy of the odometer’s projected trajectory within a short distance, we used the shape of the projected localization trajectory to inhibit the initial fusion localization noise and used trajectory matching to obtain an accurate localization. Finally, the Dual-LSTM network is constructed to predict the localization and build an electronic fence to guarantee the safety of the vehicle while also guaranteeing the updating of short-distance localization information of the vehicle when the above-mentioned fusion localization is unreliable. Under the limited arithmetic condition of the vehicle platform, accurate and reliable localization is realized in a complex environment. The proposed method was verified by long-time operation on the real vehicle platform, and compared with the EKF fusion localization method, the average root mean square error of localization was reduced by 66%, reaching centimeter-level localization accuracy.