The evaluation of scenario complexity is crucial for improving adaptability and flexibility of autonomous vehicles in coping with complex environments and enhancing the applicability of the algorithms. A graph-based algorithm for evaluating scenario complexity is developed in this paper， which fully considers interactive topology and categorizes traffic scenarios into three complexity levels. The reasonability and effectiveness are validated in ramp merging scenarios. To demonstrate its scalability， the evaluation algorithm is applied in the development of the trajectory prediction and decision-making algorithms of automated driving. The proposed algorithms are then tested using natural driving datasets and vehicle-in-the-loop experiments. The results indicate that scenario complexity evaluation enables early estimation of prediction uncertainty， enhances the real-time and optimality of decision-making algorithms. In data replay tests， the complexity assessment module can reduce the failure rate and collision rate during lane merging by approximately 38% and 92%， respectively， indicating promising application prospects.

Current crash protection research for pedestrians has been conducted primarily on adults， with no classification of adults and children. In this paper， a numerical simulation model of human-vehicle collision is established according to the relevant crash provisions of Euro-NCAP， and the simulation analysis shows that universal airbags cannot effectively reduce the head injuries of adults and children at the same time. Therefore， an airbag energy absorption device for adult and child classification protection is studied according to the characteristics of airbags and the physiological differences between adults and children， and an improved YOLOv5 pedestrian target detection model is proposed to realize the classified recognition of adults and children. According to the classification results， the vehicle control module dynamically adjusts the parameters of the airbag energy absorption device， so that the device can be deployed to different states for adults and children respectively， realizing the classification protection of pedestrians. The results show that the designed target detection model is able to achieve the classification and recognition of pedestrians， with an increase of 3.11% and 4.32% in terms of adult and child category detection accuracy， respectively， compared with the initial model. After installing the airbag energy absorption device， the HIC value of the adult head can be reduced by a maximum of 63.4% and the peak acceleration can be reduced by a maximum of 61.7%， while the HIC value of the child head can be reduced by 31.4% and the peak acceleration can be reduced by 53.2%. The thesis research results can provide scientific theoretical support for the design of pedestrian active and passive safety protection devices.

Effectively predicting the future risk indicators of multiple traffic participants under the driver's field of vision is the key to providing risk warnings to human drivers and avoiding potential collision risk. Most existing research on risk only considers the pairwise interaction between a single individual and the vehicle in the scene， and conducts research from the perspective of evaluation rather than prediction， while ignoring the different interaction between heterogeneous traffic participants and future risk status. This paper proposes a heterogeneous multi-objective risk prediction method Risk-STGCN based on spatiotemporal graph convolutional neural network， using graph convolution and temporal convolution to learn single-frame scene graph information and timing information respectively， combined with multi-layer timing prediction network to predict the multi-objective risk indicator TTC. Training and verification are conducted on the open source data set BLVD and the real vehicle self-collected data set， which is then compared with commonly used sequence prediction models. The experimental results show that the average TTC error of the proposed model on different data sets is less than 0.95 s， with multiple experimental indicators better than other models mentioned in this paper. The proposed model has good robustness and improves the interpretability of risk prediction in complex traffic scenarios.

For the stability control problem of distributed four-wheel-drive electric vehicles under extreme conditions， considering the influence of sensor noise of yaw rate， lateral and longitudinal acceleration， as well as the estimation error of slip angle， a phase plane stability domain division method based on extension theory and an adaptive Tube-based Model Predictive Control algorithm （ATMPC） are proposed to quickly quantify the stability level of the vehicle and ensure the vehicle driving stability while maintaining tracking accuracy. The designed vehicle yaw stability control system utilizes hierarchical design architecture. The upper layer employs the extension theory to associate the vehicle slip angle-yaw rate phase plane with extension control domain and determines the control domain based on the actual vehicle state and calculates the dependent function to realize the decision-making of the control target weights and modes of the lower layer's Tube-MPC. The lower layer utilizes Tube-MPC to track the desired vehicle slip angle and yaw rate， enabling precise decision-making regarding the yaw moment， and adopts the tire loading ratios optimization method for the allocation of the yaw moment. The control strategy is validated by Carsim/Simulink co-simulation. The results show that the proposed control framework and ATMPC strategy can significantly enhance the driving stability of vehicles in extreme conditions and improve robustness in noisy environments， outperforming traditional MPC.

Scenario-based simulation test method is an important means of automated driving vehicle safety verification； however， current test scenarios generation methods are mostly for independent scenarios. How to simulate the human real driving process to generate continuous interactive test scenario with challenges has become a problem that needs to be solved urgently in automated driving test evaluation. In this paper， an automated driving anthropomorphic continuous interactive test scenarios generation method is proposed. Firstly， the architecture for anthropomorphic continuous interactive test scenarios generation is established， and the vehicle motion behavior analysis is conducted based on the HighD dataset. On this basis， the current behavior of tested automated driving vehicle based on the trajectory similarity feature is analyzed， and the prediction of the future trajectory through the state transfer matrix is realized. Then， the type of the future behaviors of the traffic vehicles based on the trajectory interaction rules are determined， and the specific trajectory is generated by Transform network. Finally， the key performance indicators such as danger and anthropomorphism of the generated test scenarios are evaluated in simulation test environment， which proves the effectiveness of the method proposed in this paper.

With the increase of new energy vehicles in the market and battery energy density， thermal runaway events gradually increase. Battery safety issues become particularly important， whereas leakage is one of the key factors inducing battery thermal runaway. In this paper， the influence of leakage on electrical performance and safety is studied by simulating leakage at the cells and modules. At the same time， based on experimental data and remote vehicle data， the characteristics of the leaked battery are extracted and the warning logic is established to achieve online monitoring of the leakage warning. For cells test， a comparative analysis is conducted on the test data of leaking and normal battery cells under cyclic and static states. It is found that compared with normal cells， leaking cells show mass reduction， thickness increase， capacity fade， DC internal resistance increase and dismantling characterization abnormity， which proves that leakage has certain impact on the electrical performance and safety. For module test， characteristics of the thickness and DC internal resistance changes from the parallel units with different leakage degrees are studied. It proves that the thickness and DC internal resistance increase with the increase of leakage degrees， which also augments the potential safety risk of the battery. For vehicle level big data， the pressure difference characteristics of the leaked battery during the starting and ending stages of charge are identified to establish warning and identification logic and conduct online monitoring.

With the rapid development of automated driving technology， ride comfort has become a key factor affecting user acceptance and overall experience with automated vehicles. In this paper， a comprehensive review of the current state of research concerning the evaluation of riding comfort in automated vehicles is presented. Firstly， the concept of comfort is thoroughly articulated， followed by an analysis of key factors influencing ride comfort. Subsequently， the quantitative indicators and evaluation models pertinent to automated vehicles are classified and elaborated in detail. The quantitative indicators are classified into four categories： subjective indicators， indicators derived from vehicle parameters， indicators based on physiological signals， and indicators related to driver behaviour. The evaluation models encompass psychophysical models， biomechanical models， statistical models， and learning-based evaluation models. Finally， prospective trends in the research of comfort in automated vehicles is brought forward， thereby offering a technical framework for further studies on the system design and user experience in this domain.

To address the problem of speed trajectory deviation of connected vehicles （CVs） caused by human driver error， a real-time eco-driving strategy for connected mixed platoons considering human driver error is proposed in this paper. Firstly， real vehicle tests are conducted to collect human driver error data of different drivers to establish the human driver error model based on Markov chain so as to predict the human driver error for a period of time in the future. Then， with the optimization objective of minimizing the fuel consumption of the entire platoon， the platoon speed trajectory optimization problem is formulated as an optimal control problem. Fast stochastic model predictive control （FSMPC） is employed to calculate the optimal speed trajectories of the connected vehicle in the mixed platoon. Both the simulation and intelligent and connected micro-car test results indicate that， compared to the traditional eco-driving strategy based on fast model predictive control （FMPC）， the proposed eco-driving strategy can effectively reduce the speed trajectory deviation and fuel consumption of the whole platoon as well as meet the real-time requirements.

As global emission regulations and energy-saving policies become increasingly stringent， gasoline engines are facing significant challenges. The urgent technical challenge is to achieve high efficiency and ultra-low emission of gasoline engines. The pre-chamber turbulent jet ignition is one of the most promising technologies for improving the thermal efficiency of gasoline engines and reducing pollutant emission. In this paper， the influence of lean combustion limit expansion and ignition timing on the optimization of thermal efficiency is investigated systematically through three-dimensional flow simulation analysis coupled with a detailed chemical reaction mechanism. The results show that the passive pre-chamber can effectively expand the lean combustion limit， improve the thermal efficiency and reduce the pollutant emission of the engine in comparison with the spark ignition. At an excess air factor of 1.5， the maximum indicated thermal efficiency is 47.24%， which is 11.89% higher than that of the original engine， with the NO _x and Soot reduced by 29.27% and 98.76%， respectively.

For the problems of path planning on curved roads， a path planning fusion algorithm based on global oriented artificial potential field method is proposed in this paper. Considering the curved road conditions， a grid map based on deformed grid is constructed. Considering the driving risk in the road environment， the heuristic function of A* algorithm is optimized based on the driving risk field theory. To improve the limitation and inherent defects of the traditional artificial potential field method， in view of the outline shapes of the subject vehicle， environment vehicles and obstacles， the artificial potential field method is improved as the local path planning method by introducing in the globally guided path. Taking the path planned by the improved A* algorithm as the global optimal guided path， the path planning fusion algorithm is designed based on the improved artificial potential field method. The simulation results show that the proposed fusion algorithm can generate effective and reasonable driving path， which is close to the real vehicle path extracted from the dataset. Moreover， the path planned in the environment with obstacles is safe and efficient， meeting the driving requirements of the vehicle.