The three-dimensional object detection algorithm based on point cloud is one of the key technologies in the autonomous driving system. Currently， the voxel-based anchor-free detection algorithm is a research hotspot in academia， but most researches focus on designing complex refinement stage， at the expense of huge algorithm latency， to bring limited performance improvement. Although the single-stage anchor-free point cloud detection algorithm has a more streamlined detection process， its detection performance cannot satisfy the needs of autonomous driving scenarios. In this regard， based on the anchor-free detection algorithm CenterPoint， a single-stage anchor-free point cloud object detection algorithm for autonomous driving scenarios is proposed in this paper. Specifically， the encoding and decoding sparse module is introduced in this paper， which greatly promotes the information interaction of the spatial non-connected areas of the three-dimensional feature extractor， ensuring that the three-dimensional feature extractor can extract features that satisfy various target detection. In addition， considering that it is challenging to adapt the existing two-dimensional feature fusion backbone to the center-based head， in this paper self-calibrated convolution and large kernel attention modules are introduced in to effectively extract point cloud features of the target area， which are then gathered into the center point area， thereby improving the algorithm's recall and accuracy of the target. The proposed algorithm in this article is trained and experimentally verified on the large-scale public dataset of nuScenes. Compared with the benchmark algorithm， mAP and NDS are increased by 5.97% and 3.62% respectively. At the same time， the actual road experiments with the proposed algorithm are conducted on a self-built vehicle platform， further proving the effectiveness of the proposed algorithm.

It is crucial to develop a lightweight real-time online temperature prediction model for electric drive assembly （EDA） to effectively monitor its future abnormal temperature state in advance and ensure vehicle safety. Based on multi-physics coupling and data-driven fusion modeling， this paper proposes an online prediction method for the transient temperature field of EDA. Firstly， a multi-physical coupling finite element model of EDA electric-magnetic-thermal-flow multi-physics coupling is established， and the accuracy of the model is verified by bench test. Secondly， several transient temperature field datasets under normal working conditions are generated via multi-physical field coupling model for subsequent proxy model verification. Then， combined with the finite element model to obtain the simplified thermal network topology and the graph convolutional neural network， a relational spatial-temporal graph convolutional neural network prediction model driven by model and data is proposed. Finally， the effectiveness and real-time performance of the proposed temperature prediction model are verified by offline simulation and online test under different ambient temperatures and working conditions. Analysis results on the measured offline dataset show that the global prediction error and average absolute error are 4.4 and 1.25 ℃， reduced by 17.3%， 28.1%， 5.3% and 29.3%， respectively， compared with the conventional temporal graph convolutional neural network and gated recurrent unit. Meanwhile， the online prediction results of the bench are also very close to the real measured values， with the global prediction error and average absolute error of 3.99 and 0.66 ℃. In conclusion， the proposed real-time on-line temperature prediction method can accurately predict the real temperature change of EDA.

A lithium battery early fault diagnosis method based on WOA-VMD and Shannon entropy is proposed in this paper to solve the problem of current battery management systems being unable to diagnose early faults. Firstly， the whale optimization algorithm is introduced to optimize the parameters of the variational mode decomposition algorithm to improve its decomposition performance and obtain intrinsic mode function components containing more fault feature information. Then， the voltage signal of the individual battery is decomposed and reconstructed to reduce the impact of measurement noise and additional excitation voltage. Furthermore， a sliding window is used to calculate the Shannon entropy range of individual voltage and the overall Shannon entropy of individual voltage dispersion to set appropriate thresholds for early fault diagnosis. After verification with actual vehicle data， this method can provide fault warning about 10 minutes in advance without generating false warnings for vehicles without faults. It has strong robustness and reliability.

As an important stage of the automotive design process， conceptual design requires rapid conceptual design and evaluation. The current methods generally use a combination of parametric design and CAE to achieve analysis based conceptual design of car body structures. With the development and maturity of machine learning and deep learning algorithms， intelligent design methods will become the main innovative technology for body structure design. In this article， a combination of data-driven and optimization design method is used to independently develop the vehicle structure intelligent design software tool （S-iVCD）. Firstly， based on residual networks and thermal map regression algorithms， feature points of the vehicle body structure are extracted to achieve automated modeling of the conceptual model of the vehicle body structure. Secondly， based on Gaussian process sampling， a body structure dataset is collected and a fully connected neural network model is used to construct the body structure network model. The parameters of various components of the vehicle body can be input into the trained network model to obtain the overall performance results of the vehicle body. Finally， by combining data-driven computing with the moving asymptote algorithm， a multi-objective optimization design of the vehicle body structure that considers mass， bending stiffness， and torsional stiffness is quickly achieved. By comparing with finite element examples， the error of the calculation results is within the allowable range， with the optimization calculation time greatly shortened， and the lightweight rate reaching 7.4%. This indicates that the data-driven body structure optimization design method is effective in improving efficiency in the conceptual design stage of automobiles.

There may be inconsistencies in temperature， charge state， aging state （capacity and internal resistance） between individual cells in a battery module. Due to the existence of the "short board effect"， the inconsistencies will affect the overall performance of the battery module， so timely and accurate inconsistencies diagnosis is very necessary. Considering that the above-mentioned inconsistencies will affect the electrode process characteristics， which will be reflected in the Electrochemical Impedance Spectroscopy （EIS） and Distribution of Relaxation Time （DRT）， in this paper， after clarifying the effect of several kinds of inconsistencies on EIS and DRT by combining the equivalent circuits， an inconsistencies diagnosis method for battery modules based on EIS and DRT is innovatively proposed. The performance of unsupervised clustering algorithms such as K-means， AP （Affinity Propagation） and DBSCAN （Density Based Spatial Clustering of Applications with Noise） is comparatively analyzed by mixing the abnormal batteries into a group of batteries with good consistency. The results show that the DBSCAN diagnostic accuracy is 99.2%， which can realize the accurate diagnosis of the inconsistency difference of single cells within the battery module.

A deep-learning LSTM-based POD model （LSTM-POD） based on long short-term memory （LSTM） and proper orthogonal decomposition （POD） is developed for the turbulent wake of the square-back Ahmed automotive general model. A high time-resolution reconstruction is achieved by establishing the mapping relationship between the POD modal coefficients of the non-time-resolved planar velocity field and the time-resolved velocity signals at a number of discrete points， and the effect of different time-step configurations， i.e.， the single time step （LSTM-Sin） and multiple time steps （LSTM-Mul） on the reconstruction results is compared. The results show that the LSTM-POD model has strong learning and generalization ability in time series reconstruction， In addition， LSTM-Mul considers temporal continuity and correlation， the reconstructed mode coefficients （lower order） and velocity fields of which are more consistent with the POD reconstructed results compared with that of LSTM-Sin. The deep learning model proposed in this study can alleviate the problems of high resource consumption and low computational efficiency in obtaining high time resolution flow field data through experiments and high-precision numerical simulation.

For the trade-off between prediction model accuracy and computational cost for path tracking control of autonomous vehicles， a learning-based model predictive control （LB-MPC） path tracking control strategy is proposed in this paper. A two-degree-of-freedom single-track vehicle dynamic model is established， and an in-depth analysis is conducted on its step response error with respect to variation in vehicle speed， pedal position， and front wheel steering angle compared to the IPG TruckMaker model. Methods for constructing error datasets and receding horizon updates are designed， and the Gaussian process regression （GPR） is employed to establish an error-fitting model for real-time error compensation and correction of the nominal single-track model. The error correction model is utilized as the prediction model， and a path tracking cost function is designed to formulate a quadratic programming optimization problem， proposing a learning-based model predictive path tracking control architecture. Through joint simulation using the IPG TruckMaker & Simulink platform and real vehicle experiments， the real-time performance and effectiveness of the proposed GPR error correction model and LB-MPC path tracking control strategy are verified. The results show that compared to the traditional model predictive control （MPC） path tracking control strategy， the proposed LB-MPC strategy reduces the average path tracking error by 23.64%.

The operating parameters of the PEM fuel cell stack have an impact on the performance of the stack output as well as the parasitic power of auxiliary devices such as air compressor， recirculation pump and cooling fan. System maximum net power output goals can be achieved by optimizing the operating parameters of the fuel cell stack. Forthe actual system， constrained by the performance of the air compressor and the regulating capacity of the backpressure valve， the adjustment range of cathode operating parameters is limited. In this paper， the 62 kW fuel cell system model is established based on MATLAB/Simulink. Through simulation analysis， the achievable ranges of parameters optimization under various load currents are determined. Genetic algorithm is employed to optimize the fuel cell stack temperature， cathode pressure， and oxygen excess ratio. The results show that increasing the temperature of the fuel cell stack at various load currents is advantageous for enhancing the system's net power， with the optimal operating temperature being 80 ℃. However， the optimization direction for the oxygen excess ratio and cathode pressure varies at different load currents. At low load current （50， 100 A）， increasing the oxygen excess ratio and cathode pressure results in a smaller growth in stack output power compared to the parasitic power. Providing lower oxygen excess ratio and cathode pressure is advantageous for enhancing the net power of the system. At high load current （300 A）， low oxygen excess ratio and cathode pressure will limit the output power of the stack， with the lowest net power of only 35.530 kW. After the oxygen excess ratio and cathode pressure are reasonably increased， the optimal net power is 53.271 kW， and the net power can be increased by 49.9% through the optimization of operating parameters.

Multi-sensor fusion is an effective way to improve intelligent vehicle perception. For the data-matching problem of the three types of sensors of LiDAR， millimeter-wave radar， and camera， traditional methods such as bipartite graph matching can’t achieve high precision， with poor matching robustness. Therefore， a multi-sensor data fusion algorithm for intelligent vehicles based on tripartite graph matching is proposed in this paper. The problem of data matching of the three sensors is abstracted as a weighted tripartite graph-matching problem. By using Lagrange relaxation， the original problem space is decomposed into subspaces， the weights of vertices and edge inside which are determined then by the cost matrix model. Furthermore， combining the perceptual error model and likelihood estimation， the posterior distribution of perceptual errors is determined. Ultimately the Lagrange Multiplier （LM） model is used for data matching. Finally， the effectiveness of the proposed matching algorithm is validated by the nuScenes training dataset and real-world vehicle tests. On the dataset， the proposed algorithm improves F1 scores by 7.2% compared to common algorithms. In various real-world vehicle scenarios， the proposed algorithm shows excellent perceptual accuracy and robustness across.

To ensure the reliability of aluminum-plastic film encapsulation for lithium-ion batteries， strict control of the aluminum layer thickness after forming is necessary. However， obtaining the thickness relies heavily on physical experiments， resulting in high cost for both early design optimization and later production process quality monitoring. In this paper， a combination of physical experiments and simulation modeling is adopted to establish a constitutive equation that can well characterize the mechanical properties of the pouch during forming. Additionally， a prediction method for the aluminum layer thickness based on the overall aluminum-plastic film thickness is proposed， enabling precise prediction of the aluminum-plastic film and aluminum layer thickness after forming. Furthermore， based on simulation Design of Experiments （DOE）， key influencing factors are screened to construct a response surface model， facilitating rapid prediction and optimal parameter matching design for different products， which also provides a solution for online monitoring of forming quality during production. The results show that the multi-layer composite aluminum-plastic film exhibits obvious anisotropy during the plastic stage. The 3-parameters Barlat-Lian constitutive model effectively represents the anisotropic properties of the film， and outperforms the single-directional elastic-plastic model， achieving accurate prediction of the aluminum-plastic film performance after forming. The constructed response surface model can replace the refined finite element model， and have excellent prediction accuracy for the thickness of the composite aluminum-plastic film and the aluminum layer， with an error less than 5%. By optimizing the process parameters， the formed thickness of the aluminum layer can be increased by 10%~20%. The integrated development application APP can meet the requirements for quick design evaluation， parameters optimization， and online monitoring of the forming quality.