In order to fully understand the research methods and current status of deep learning-based trajectory prediction of intelligent vehicles, through the analysis and summary of existing literature, the input representation, output types, and prediction methods of deep learning-based trajectory prediction models are analyzed. The results show that trajectory prediction methods based on deep learning demonstrate outstanding performance in long-term, multi-modal motion and vehicle-road interaction scenarios.

To address the challenge of determining the weight matrix in the Linear Quadratic Regulator (LQR) algorithm for intelligent vehicle path tracking control, a LQR path-following controller based on genetic algorithm (GA) optimization is proposed. Firstly, a 2-DOF vehicle path tracking error dynamics model is established, and a LQR controller with preview is designed to calculate the optimal front wheel angle output of the vehicle; Then, for the determination of LQR controller parameters, a GA optimization strategy with vehicle lateral error, yaw error, and output front wheel angle as the objective function is designed. The optimization solution yields the optimal weight matrices Q and R; Finally, in the environment of MATLAB/Simulink and CarSim joint simulation, the tracking performance and robustness of LQR controller optimized by the genetic algorithm is verified. The results show that: under the double-lane-changing road and continuous lane-changing conditions, the optimized GA_LQR control vehicle’s lateral error peak value is reduced by 86.6% and 84.2%; the heading error peak value is reduced by 17.7% and 14.6%, respectively; the front wheel angle is smaller. The vehicle’s lateral and yaw tracking capabilities are enhanced as well. Moreover, the vehicle still exhibits good path tracking, speed tracking, and driving stability under different speeds on double-sine roads.

In recent years, with the rapid development of autonomous driving technologies, the demand of intelligent vehicles for environment perception technology is also higher and higher. Due to the high accuracy of LiDAR data that can better obtain the 3D information in the environment, it has become a research hotspot in the field of 3D target detection. In order to provide more accurate environmental information for intelligent vehicles, the main research contents in the field of 3D target detection by LiDAR are summarized. Firstly, the advantages and disadvantages of various environment sensing sensors for self-driving vehicles are analyzed; secondly, according to the different data processing methods in 3D target detection algorithms, the detection algorithms based on point cloud and the detection algorithms fused with image and point cloud are reviewed; then, the mainstream self-driving datasets and their evaluation methods for 3D target detection are sorted out; and finally, the current 3D target detection algorithms for point cloud are summarized and outlooked. The results show the importance of the 2D view method and the multimodal fusion method in the current research for the development of autonomous driving technologies.

Unstructured road recognition is a challenging problem in unmanned driving, involving the complexity of the road itself, such as unfixed type, irregular shape, uneven surface and blurred borders. In order to have a comprehensive understanding of vision-based unstructured road recognition methods and research status, through the analysis and summary of existing literature, this paper analyzes the existing three mainstream methods, which are road features-based, road model-based and machine learning-based methods, and collates the currently commonly used unstructured road open source data sets. The results show that the method based on road characteristics and road model has high computational complexity and low recognition accuracy, and the method based on machine learning can significantly improve the recognition accuracy, but the problems such as large data demand, long training time and poor interpretation are existed as well.

During the charging and discharging process, lithium-ion batteries generate a large amount of heat. If the temperature is too high, it may cause battery failure or safety problems. As one of the most important technologies of battery cooling system, lithiumion battery cold plate cooling technology is significant to ensure battery safety. Firstly, this paper systematically analyzes the relevant methods of cold plate liquid cooling design, and compares the advantages and disadvantages of different methods. The mainstream cooling method for lithium-ion battery thermal management systems is currently liquid cooling, which boasts higher heat dissipation efficiency. With the increase in battery energy density, the future development trend of electric vehicle thermal management systems may move towards hybrid cooling, systematic design, and intelligent management.

Vehicle safety performance is a primary concern for consumers when selecting a new vehicle, where the crash resistance is of utmost. Central to this is the energy absorbing box, which can dissipate impact energy through deformation and crushing mechanisms, thereby maximizing safety of passengers. To optimize the energy absorbing performance of thin-walled structure and enhance the crash protection of vehicle, the performance evaluation indicators and structural types of thin-walled energy absorbing box is analyzed，and the deformation and energy absorbing characteristic are revealed. Finally, strategies and suggestions for the future development of thin-walled energy absorbing box are proposed, aiming to provide insights for research and practice in the field of vehicle safety.

With the innovation and development of new alloy materials, pressure casting technology with high product quality, precision, and productivity is widely used in the manufacturing of various automotive components. This article summarizes the principles, characteristics, processes, development status, and some methods for controlling defects in die-casting processes. The application and development of aluminum and magnesium alloy die castings in automobiles are reviewed, and some typical automotive aluminum and magnesium alloy die castings are introduced. The viewpoint that the application of aluminum and magnesium alloy die castings in automobiles has great significance in promoting automobile lightweight is proposed. Finally, the development prospects of die castings in the automotive industry are prospected.

The model of hydrogen system of high-power fuel cell engine is designed, the core components and control system of hydrogen system are modeled and simulated in MATLAB/Simulink platform. Based on the hydrogen system model, the CLTCP and NEDC common driving conditions are simulated on dynamic. The results showed that the anode working pressure requirements could be guaranteed by the inlet pressure, and the pressure difference between cathode and anode was maintained in 0.02 MPa. The hydrogen utilization rate of the two conditions were 99.67 % and 99.76 % respectively. That means the model can be used in the development of high-power vehicle PEMFC system.