A dual-motor non-synchronizer multi-gear mechanical transmission system can effectively improve the energy economy and power performance of heavy-duty commercial vehicles. In order to analyze the optimal performance of the system under the optimal parameters，and to consider the effect of vehicle weight and differences between no load and full load，this study optimizes both the drive motor parameters and mechanical transmission speed ratio parameters，and compares the energy economy and power performance of the single/dual-motor multi-gear mechanical transmission system under the optimal parameters. The results show that the dual-motor drive system reduces the sensitivity of the vehicle energy economy to the difference in vehicle weight and no/full load，with the maximum speed of the dual-motor drive system increased by about 8%，and the acceleration time saved by about 28%.

The wet clutch is the core component of the vehicle transmission system. It is prone to rub-impact between the friction plate and steel plate during high-speed separation，resulting in a sharp increase in drag torque，and affecting its transmission efficiency and reliability. Therefore，in this paper，to reduce the rub-impact drag torque in high-speed wet clutch，the micro-texture on the surface of the friction plate is optimally designed. Firstly，a parameterized modeling method of arbitrary micro-texture shape lines on the surface of friction plate is proposed. Then the number，depth，circumferential proportion，radial proportion and shape line parameters of the micro-texture are selected to construct the design variables，constraint conditions and optimal objective function for micro-texture optimization. By combining the experiment design method，approximation modeling simulation and global search optimization method，an optimal design model of micro-texture on the surface of friction plate is established. Finally，a comparison test of drag torque before and after micro-texture optimization is carried out. The results show that the optimized micro-texture can significantly reduce the rub-impact drag torque at high circumferential speed，and greatly delay the critical speed at which the rub-impact phenomenon of friction pair occurs.

The promotion of intelligent cockpit and virtual testing protocols bring new challenge to assess the occupant injury，with the injury mechanism and injury risk assessment parameters more diversified. Based on the TUST IBMs 6YO-O and the BP neural network algorithm，a predictive model for the correlation between occupant sitting angle and head injury indicators in frontal 100% overlapping rigid barrier condition is constructed in this paper，and the correlation and difference between evaluation indicators with the different seating postures are explored. The results show that the constructed correlation injury prediction model has high reliabilities （R ² > 0.90），which can be used for injury prediction and analysis. Existing head injury evaluation indicators have good consistency in the small angle range （95°~108°），but for the occupants with larger seating postures，there are significant differences to assess the head injury risks using different injury evaluation indicators. Therefore，there is certain limitation of the head injury assessment parameters implemented currently. In the future virtual testing，the kinematic and biomechanical parameters should be integrated to assess more comprehensively for the head injury risks. The research results can provide data and theoretical support for the improvement of child restraint systems，virtual testing，and selection of head injury evaluation parameters for occupants with larger seating postures.

The vibration and noise performance of the heat pump system plays an important role in the NVH evaluation of new energy vehicles. Based on the structure and working characteristics of the heat pump system and the principles of vehicle vibration and noise control，a research on the NVH control methods of the heat pump system is carried out in this paper from the vibration and noise excitation source，structural mode distribution，transfer path，evaluation conditions and other dimensions. Through the analysis and solution of NVH problems in the product development of a domestic hybrid electric car，the results indicate that the NVH control of heat pump system is a system engineering，and the compressor，air conditioning pipeline，HVAC shell，acoustic package and compressor control strategy are key factors of NVH，which provides a clear technical reference for the NVH performance control of new energy vehicle heat pump systems.

For the existing acquisition method of vehicle sideslip angle （VSA） based on fusion strategy，optimization and improvement are required for different scenarios，structures or signals. As a result，the calculation complexity and system cost rise up. A soft-sensing method for the VSA based on switch strategy is proposed in this paper. In the pretreatment part for sensor measurement signals，the Bessel filter is employed to realize the delay processing and noise filtering of lateral acceleration signal，and a reliability test module is designed to effectively eliminate the mutations in the signals of yaw rate，et al. On this basis，the switch strategy is determined based on the advantages of the existing kinematic and dynamic schemes. For the presented strategy，the continuous operating intervals of the kinematic scheme are shortened in a great deal of effort to restrain the error accumulation，and the dynamic one is restricted in linear region to avoid performance degradation. Simulations and hardware-in-loop experiments are implemented in multiple conditions to verify the effect of the proposed method. The experimental results show that the proposed method in this paper has obvious advantages in accuracy and execution time compared to the one utilizing the classical fusion strategy. Moreover，they are both robust to the change of road conditions.

To improve the fuel economy of vehicles，simulation and experiments are combined to improve the aerodynamic drag coefficient during driving，taking a certain SUV model as the research object. Firstly，wind tunnel tests are used to determine the areas or components that have significant impact on the overall aerodynamic drag of the vehicle. Secondly，optimizations are made to the components or areas with high contribution values to the air resistance coefficient. The results show that the front wheel deflectors，taillights and spoilers contribute greatly to the overall air resistance coefficient of the vehicle. The restyling of the front wheel deflectors effectively reduces the frontal pressure area and interference drag from the wheels. Optimizations on taillights and spoilers improve the rear negative pressure zone and shorten the reattachment distance of separated flows on the upper part of the rear window. Based on the intrinsic orthogonal decomposition method for extracting and analyzing local flow field information，it can be concluded that the first and second order modals mainly constitute the key flow states in the wake. Compared to the initial scheme，a drag reduction rate of 7.5% can be achieved by the optimized combination design，which is verified by tests and simulations. Theoretical basis and technical support are provided in this paper for restyling and model change of the next generation of SUV.

For the difficulties in feature extraction and low recognition rate in defect types and grades of rivet on aluminum alloy plates for car body，the diagnosis model and detection method for rivet failure defects are proposed based on the Gaussian convolutional deep belief network and long short-term memory network. Firstly，the specimens are designed for five types of fracture defects and an automatic detection system is constructed. The planned path and pose of the probe are set to lower lift-off effect on signals. Secondly，the dual network fusion diagnostic model is designed to extract and learn the multi-dimensional defect feature information，solving the problem of extracting defect information represented by temporal variation characteristics and spatial distribution state in detection curves. The experiments results show that the optimized model has an average recognition rate of 99.85%，with an increase of 14.54% compared with that of the traditional convolutional network and single deep belief network. The model has better compatibility and robustness，which can realize online diagnosis of internal defects of rivets.

Nondriving behavior identification is one of the important ways to improve the safety of driving. The current recognition method based on skeleton sequence and image fusion has the problems of large model calculation and the difficulty of feature fusion. To address the above problems, the skeletonimage based behavior recognition network (SIBBRNet) is proposed in this paper, which is based on the multiscale skeleton graph and the local visual context. SIBBRNet fully extracts motion and appearance features through a graph convolution network based on multiscale skeleton graphs and a convolutional neural network based on local vision and attention mechanisms, and better balances the relationship between model representation capabilities and model calculation. The feature bidirectional guided learning strategy based on hand motion, an adaptive feature fusion module and an auxiliary loss on the static feature space can guide mutual guidance and updating between motion and appearance features to achieve adaptive fusion. SIBBRNet is finally tested on the Drive & Act dataset, and the average accuracy is 61.78% for dynamic labels and 80.42% for static labels. The Floatingpoint Operations per Second (FLOPS) of SIBBRNet is 25.92G, which is 76.96% lower than that of the optimal method.

To effectively evaluate the takeover risks of L3 autonomous vehicles under different cognitive secondary tasks, a study on the risk assessment model for driving takeover is carried out. The urban expressway emergency takeover scenario is designed and driving simulation experiments under different cognitive secondary tasks are carried out. The takeover risk assessment model considering trajectory field, potential field and behavior field is established. The validity of the proposed model is verified by adopting the takeover risk index method. Combined with the measured data, the influence of different cognitive secondary tasks and avoidance operation types on the strength of takeover risk field is quantized. The results show that the MW test and KS test for the distribution of the takeover risk index between 1 and 9 s after the takeover operation by the participants are both with the result of p<0.05, indicating that the model can effectively assess the takeover risk of the vehicle during the takeover process. In addition, the root mean square error of the takeover risk index (0.062) is smaller than the root mean square error of the inverse timetocollision (0.098), indicating that the model is better than the inverse timetocollision in accurately describing the risk. The research results can provide reference for vehicle operation risk assessment and collision avoidance design in takeover process.

For the problems of single mode of car navigation interaction, inability to dynamically change interaction strategies according to contextual differences, and inadequate user experience, a context engine model that supports adaptive car navigation interaction strategies based on different contextual situations is proposed in this paper. Combining the usercentered design (UCD) and context awareness theory, automotive humancomputer interaction contexts are defined, and key contextual elements are extracted through followup surveys of typical drivers. The adaptive interaction strategies and principles of the context engine model in navigation contexts are elaborated, and users' different information needs are mapped to the detailed level of interaction performance at the presentation layer. The navigation HMI information frameworks and components that fit the different levels of driver needs are summarized and designed for interface design and analysis of two typical navigation application scenarios. The results show that the context engine can adaptively interact with driver' needs in various driving situations, providing a reference for optimizing and developing intelligent cockpit driving interaction experiences.