This paper proposed a trajectory tracking control strategy that combined Model Predictive Control (MPC), Radial Basis Function (RBF) neural network, and Sliding Mode Control (SMC) to address the low accuracy of vehicle trajectory tracking caused by model mismatch and external environmental interference during the driving process of autonomous vehicles. By establishing a vehicle kinematic model predictive control, the expected yaw rate of the vehicle in the current state was calculated, and the deviation value from the actual yaw rate was input to the RBF-SMC controller. By utilizing RBF’s ability to quickly approach nonlinear models, combined with sliding mode control to output front wheel angles, the lateral trajectory tracking control of the vehicle was achieved. The simulation experimental results show that this method significantly improves trajectory tracking accuracy compared with traditional controllers, and exhibits good robustness under different driving conditions.

Nonlinear Model Predictive Control (NMPC) method-based motion control has attracted considerable attention in the field of autonomous driving. However, the steady-state error problem has not been comprehensively investigated, especially for nonlinear MPC. This paper seeks to solve the steady-state error problem based on Offset-Free NMPC (OF-NMPC) with a lateral-longitudinal coupling structure. The proposed OF-NMPC uses an Unscented Kalman Filter (UKF) to observe the states and disturbances and incorporates them into the prediction model and reference calculation to eliminate the steady-state error. One of the challenges of OF-NMPC is the need to use optimization methods to obtain reference values, which will obviously increase the considerable computational burden. Based on the appropriate simplification, we get the reference analytical solution without solving nonlinear optimization problems online in real-time. Simulation and real vehicle experiments show that the proposed OF-NMPC can effectively eliminate the steady-state error and improve the system’s dynamic performance.

To improve calculation efficiency of lane-keep control algorithm, a Model Predictive Control (MPC) algorithm for lane-keeping was constructed. A single-rail vehicle model was derived based on rigid body dynamics. The standard vehicle model was based on rigid body dynamics and considered lateral and longitudinal tire force characteristic. Based on that, a simplified vehicle model was derived by assuming zero slip angle and slip ratio. The simplified model formulated yaw rate using linear equations and eliminated the tire model, thereby reducing the complexity of the constraint equations in the MPC. Considering the tracking error, control input, and the cost item of its change rate as the objective function, the control effect of standard vehicle model and simplified vehicle model were compared and analyzed. The results show that the simplified vehicle model achieves similar control performances to the standard model in lane-keeping MPC and avoids the problem of being unable to solve the tire model when the vehicle speed is close to zero. Additionally, the simplified model significantly reduces the computational time required for MPC optimization.

In order to reduce the weight and cost, the topology optimization method based on variable density method was used to carry out multi-objective topology optimization of the battery bracket. The optimization results show that the mass of the optimized battery bracket has been reduced by 16.7%, and the structural performance meets the design requirements, which has been verified through vehicle road tests. The optimization process shows that topology optimization provides a fast and effective solution for lightweight design of structures.

In order to eliminate fault of gearbox gear switch in use, this paper, by correlating with the faulty parts in the use of gear switch, analyzed oil intake problem caused by poor sealing of gear switch, switch on-and-off failure under actuating pressure as well as contact ablation, to further improve the performance of the gear switch. The verification results show than the proposed scheme can effectively solve the problems existing in the use of the gear switch.

To address the shortcomings of the traditional method of only considering dimensional chain tolerance calculation in the appearance matching of closure, this paper introduced the deformation of decklid caused by boundary condition changes into the tolerance calculation process based on the comparison of finite element analysis and bench measurement analysis. Based on the analysis of the influencing parameters of the appearance deformation of decklid and the corresponding tolerance calculation results, a combination of CAE analysis and tolerance calculation was used to improve the appearance quality matching evaluation method of decklid. This method supports for improving the appearance matching performance of vehicle closures during the development stage and online appearance matching adjustment.

In order to investigate the effect of frontal collision deceleration waveform parameters on dummy chest injury, the method of equating the deceleration waveform to a Bi-Slope Approximation Wave (BSAW) was used to complete the frontal collision simulation using MADYMO. The results show that the dummy damage corresponding to the bilinear deceleration waveform curve with a larger slope in the first stage and a higher peak at the end is smaller under the condition that the vehicle speed and deformation are guaranteed to be constant; the influence of BSAW parameters on the chest deceleration peak along the X-direction is from the first stage peak, the second stage peak, and the first stage peak moment in decreasing order under the condition that speed change remains unchanged.

To predict injury of the occupant in vehicle collision more rapidly and accurately, a training database for deep learning models was established based on frontal 100% overlap rigid barrier real-world collision data, and data preprocessing and features extraction were conducted. Deep learning models were constructed separately based on Long Short-Term Memory (LSTM), Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) neural network, and Temporal Convolutional Networks (TCN) for injury prediction training. The validation results show that the model prediction accuracy reaches 0.8579, 0.8209 and 0.9674, respectively, demonstrating feasibility of the proposed method.

In order to study the seat safety of passenger cars in China, this paper analyzed the casualties of each seat in three collision modes based on China In-Depth Accident Study (CIDAS) (2011~2022) statistical data, calculated the fatal risk of passengers in the front and rear seats using the risk model, and used risk weighting model and geometric average model of the risk to calculate risk of each seat. The results show that the fatal risk coefficient of the front row is 1.18 compared with the back row. Taking the risk of 100% of the driver’s seat position as the reference standard, the risk of the passenger seat, the left rear seat, the right rear seat and the middle rear seat was 79.57%, 105.23%, 93.28% and 191.69%, respectively.

In order to enhance the safety of the battery system in electric vehicles under side pole impact, a finite element model incorporating the battery system was established. The accuracy of this finite element model was validated through vehicle side pole impact tests. Using this model, the deformation status and acceleration response of the battery pack in side rigid pole impact scenarios were analyzed with structural deformation and acceleration impact as the primary evaluation criteria. The results indicate that under side pole impact, both the external framework and internal support structure of the battery pack experience significant deformation, and all positions within the battery system withstand substantial acceleration impacts. Based on this result, the paper proposed safety-specific design for this impact condition when designing power battery pack and vehicle body sill beam structure.