Table of Content

25 June 2025, Volume 47 Issue 6 Previous Issue   

Select

Research on the DDDQ Magnetic Coupled Structure of the Wireless Charging System for Electric Vehicles Yang Yang,Zhi Zhang,Bo Xu,Yang Dong 2025, 47 (6):  1007-1021.  doi: 10.19562/j.chinasae.qcgc.2025.06.001 Abstract ( 480 )   HTML ( 51 )   PDF (9539KB) ( 252 )   Save The magnetic coupling mechanism， as the key component in the wireless charging system for electric vehicles， greatly determines the overall performance of the system. In this paper， a double D double Quadrature （DDDQ） magnetic coupled structure is proposed， which can not only improve the anti-offset capability of wireless charging system through coupling and complementing of the coils， but also be compatible with static wireless charging scenarios and dynamic wireless charging scenarios. To verify the performance of the proposed coupling structure， in this paper firstly simulation models for both static and dynamic wireless charging systems are established for testing and analysis. The simulation results show that the DDDQ magnetic coupling structure enhances coupling capability under offset conditions in static wireless charging and reduces coupling fluctuations in dynamic wireless charging. Secondly， the LCC-S compensation topology is chosen to compensate the power of system， and the output characteristics of static wireless charging system with different magnetic coupled structures are built and compared respectively， it is shown that wireless charging system based on the DDDQ magnetic coupled structure can still maintain better transmission capability in the case of misalignment. Then， the DDDQ magnetic coupled structure is applied to dynamic wireless charging scenario， and efficient resonant transmission is realized after configuring parameters. Finally， multiple test simulation points inside and outside the vehicle are selected according to relevant standards to evaluate the electromagnetic safety of the system in different scenarios. The results show that the DDDQ magnetic coupling structure meets electromagnetic safety standards. Figures and Tables | References | Related Articles | Metrics

Select

Research on Thermal Runaway Modeling and Safety Boundary of Li-ion Batteries Under Extreme Temperature Shock Xiaoyu Li,Shen Zhao,Jun Tian,Songli Zhang,Yanli Zhu 2025, 47 (6):  1022-1036.  doi: 10.19562/j.chinasae.qcgc.2025.06.002 Abstract ( 667 )   HTML ( 14 )   PDF (11305KB) ( 182 )   Save Thermal runaway is a key issue affecting the safety of lithium-ion batteries. Herein， the non-adiabatic environment external temperature impact test platform is established for studying the thermal runaway characteristics of the single battery and battery module of NCM523 batteries and LFP batteries， analyzing the combustion behaviors of different batteries and battery modules in addition to the thermal runaway propagation after forming a module. The three-dimensional conjugate heat transfer-thermal runaway coupling model is established to obtain the law of the thermal runaway triggered by extreme temperature shock under different state of charge and the distance from the heat source and explore boundary conditions for thermal runaway propagation of modules. The results show that reaction onset temperature for anode-electrolyte roughly declines as the SOC decreases， with lower heat generated by the side-reaction. When the distance from the heat source is greater than 150 cm， it is difficult to make the LFP battery thermal runaway in 400s. PMI foam is able to achieve millimetre-level thermal barrier. When its thickness is greater than 3.75mm or thermal conductivity is less than 0.03W/（m·K） ， it can effectively inhibit the propagation of thermal runaway. Figures and Tables | References | Related Articles | Metrics

Select

Research on Multi-level Safety Management Method of Power Battery Under Vehicle-Cloud Collaboration Linsui Cheng,Jibin Yang,Pengyi Deng,Xiaohua Wu,Xiaohui Xu,Ke Deng 2025, 47 (6):  1037-1047.  doi: 10.19562/j.chinasae.qcgc.2025.06.003 Abstract ( 291 )   HTML ( 8 )   PDF (4147KB) ( 167 )   Save In order to solve the problem of thermal runaway safety of power batteries in new energy vehicles， in this paper a multi time scale safety management method for batteries based on vehicle-cloud collaboration. On the short time scale， the value-rate model and two-dimensional fault feature are employed for the real-time diagnosis of outliers and inconsistencies pertaining to battery packs on the vehicle-end battery management system. On the long-timescale， the cloud utilizes a substantial quantity of historical data and information about vehicle-end warnings， combined with improved Shannon entropy， multi-scale fuzzy entropy， and anomaly coefficient methods， to predict the trend of the thermal runaway based on incoming data. Finally， validation based on historical data from actual thermal runaway vehicles shows that the proposed method can issue early warnings on the cloud at least five days in advance， achieving multi-level management of the long and short time domain collaboration between the vehicle and cloud ends of power batteries. Figures and Tables | References | Related Articles | Metrics

Select

Rapid Sorting of Retired Lithium-Ion Batteries by Integrating Clustering and Classification Hanqing Tang,Li Wang,Mingqiang Lin,Ji Wu 2025, 47 (6):  1048-1059.  doi: 10.19562/j.chinasae.qcgc.2025.06.004 Abstract ( 226 )   HTML ( 4 )   PDF (4868KB) ( 153 )   Save In order to improve the sorting efficiency of retired batteries and reduce data collection cost， to achieve a sorting effect close to complete data by using incomplete data， in this paper， a fast sorting method for retired batteries that integrates clustering and classification is proposed. Firstly， cluster analysis is conducted on complete data to preliminarily group the batteries， and the grouping results are used as battery labels. Secondly， the most valuable segments are selected from incomplete datasets， and a two-layer classification model is trained for classification fitting， thereby achieving rapid sorting of retired lithium-ion batteries. Finally， a validation experiment is conducted on 235 lithium-ion batteries， and the results show that the proposed method has a classification detection accuracy of 95.1% and significantly reduces the time required for sorting. Figures and Tables | References | Related Articles | Metrics

Select

A TCN-LSTM Model-Based Approach for Real Vehicle Battery Health State Evaluation Jie Hu,Haojie Wang,Min Wei,Zhihong Wang,Lin Chen,Wentao Huang,Hanrui Kang 2025, 47 (6):  1060-1071.  doi: 10.19562/j.chinasae.qcgc.2025.06.005 Abstract ( 296 )   HTML ( 5 )   PDF (5682KB) ( 288 )   Save For the problem of insufficient accuracy in battery state of health （SOH） evaluation caused by the poor quality of real-world vehicle data， in this study a battery health state evaluation method based on the TCN-LSTM model is proposed. Firstly， the random search algorithm is employed to extract constant-current charging voltage segments. Subsequently， a weighted fusion approach combining locally weighted regression and third-order polynomial regression is used to fit the global degradation trend and the local degradation trend of battery capacity. Then， features related to battery aging， including the capacity retention-corrected cumulative charge capacity， fully charged voltage， and battery consistency， are constructed and extracted. Finally， a TCN-LSTM-based evaluation model for battery health state is constructed to explore the relationship between extracted features and battery aging from multiple dimensions. The results show that the TCN-LSTM model effectively evaluates the complex capacity degradation relationship of power batteries under real-world vehicle data， achieving an RMSRE of only 0.002 1. Figures and Tables | References | Related Articles | Metrics

Select

Analysis of Hygrothermal Characteristics and Condensation Evolution in the Micro-environment of Power Battery Packs Jiqing Chen,Pu Huang,Xinping Jiang,Jihong Chen,Fengchong Lan 2025, 47 (6):  1072-1085.  doi: 10.19562/j.chinasae.qcgc.2025.06.006 Abstract ( 189 )   HTML ( 4 )   PDF (9965KB) ( 89 )   Save Under the coupled effect of external environment and its own operating conditions， condensation may occur inside the power battery pack. The accumulation of condensation can turn into water accumulation， posing a serious threat to the safe operation of the battery pack. By conducting temperature and humidity characteristic tests on the power battery pack and utilizing CFD numerical simulation analysis methods， the evolution process of temperature and humidity characteristics inside the battery pack and the mechanism of condensation generation are studied. The inflow of coolant instantaneously increases the likelihood of condensation while high-power operation of the module can inhibit the formation of condensation. High-temperature and high-humidity environment is detrimental to the normal operation of the battery pack， which not only may lead to localized high temperatures inside the battery pack but also increase the risk of condensation. This research provides a reference for the design and optimization of environmental humidity and heat management inside power battery packs. Figures and Tables | References | Related Articles | Metrics

Select

Innovative Liquid Cooling Structure and Thermal Optimization for High-Energy Lithium Battery Modules Xiang Chen,Liping He,Yongkun Xiao,Duanmao Chen,Yaodong Li 2025, 47 (6):  1086-1094.  doi: 10.19562/j.chinasae.qcgc.2025.06.007 Abstract ( 235 )   HTML ( 1 )   PDF (2670KB) ( 104 )   Save Thermal runaway in power battery is a significant cause of safety incidents in new energy electric vehicles. To enhance the thermal safety and reliability of high-energy power battery systems in electric vehicles， a novel cooling structure is designed for the high-energy density 21700-battery module. Based on Computational Fluid Dynamics （CFD） theory， the Fluent software is employed to conduct numerical simulation of the temperature field and cooling effect of the 21700 battery cell and the cooling structure of the module. The orthogonal experiment is used in conjunction with range analysis and variance analysis to study the effect of the novel parameters in cooling structure， contact angle α， and cooling parameters in process， coolant velocity v and flow direction mode р， on cooling performance. The weight ranking of these parameters' influence and their optimal combination are determined. The results show that the parameters influencing cooling performance， ranked by their impact from greatest to least， are contact angle α， coolant flow rate v， and coolant flow direction р. The optimal parameters combination of contact angle α， flow direction mode р， and coolant velocity v is 73°， mode 3， and 0.010 m/s， respectively. Under the optimal parameters， the maximum temperature and the maximum temperature difference of the battery module is 28.68 and 2.65 ℃， respectively， which meet the requirements of the maximum temperature and maximum temperature difference of the power battery module. The novel liquid cooling structure for high-energy 21700 lithium-ion battery module proposed in this paper is expected to provide a new structure， new technology and new method for cooling high-energy and high-power battery modules. Figures and Tables | References | Related Articles | Metrics

Select

Virtual Simulation Research on Crash Safety of Electric Vehicles Under Side Pole Impact Jun Wang,Chenghao Ma,Zongxuan Shen,Bobin Xing,Yong Xia 2025, 47 (6):  1095-1102.  doi: 10.19562/j.chinasae.qcgc.2025.06.008 Abstract ( 288 )   HTML ( 5 )   PDF (3844KB) ( 197 )   Save The existing statistical results of side pole collision accidents reveal that the accident forms of electric vehicles （EVs） and internal-combustion engine vehicles are similar. However， considering the risk of battery pack collision and compression failure， the research on side pole collision of EVs should fully consider complex collision conditions and include safety evaluation factors such as obstacle type， geometric size， impact angle， impact location and speed. The finite element simulation is used in this study to conduct a virtual evaluation of the safety of a certain electric vehicle’s side pole collision. An orthogonal design of experiment is carried out to extract the mechanical response and body posture of the vehicle at different collision positions. The high correlation between collision speed and battery pack intrusion and cell deformation is identified. Impact location and impact angle collectively determine rotational behavior of EVs and resultant intrusion， which is further verified by simulation. As the main lateral force transmission path， the lateral beam of the battery pack enhances the lateral structural stiffness of the battery pack and significantly reduces intrusion. Further investigation of crash safety of EVs should take structural stiffness and rotational behaviors into consideration. Figures and Tables | References | Related Articles | Metrics

Select

An Experimental Study on Suppressing Thermal Runaway of Large-Capacity Lithium Iron Phosphate Battery Pack by Micro-Positive Pressure Nitrogen Shi Li,Zhuangzhuang Jia,Guangjie Shen,Qingsong Wang,Jinhua Sun 2025, 47 (6):  1103-1111.  doi: 10.19562/j.chinasae.qcgc.2025.06.009 Abstract ( 218 )   HTML ( 3 )   PDF (10261KB) ( 59 )   Save Lithium iron phosphate （LFP） battery features a long lifetime and high safety， which is widely used in electric commercial vehicles. However， thermal runaway accidents may still occur during its use. To inhibit the expansion of fire in LFP battery packs， an efficient， easy-to-operate and low-cost internal thermal runaway suppression technology is proposed， that is， injecting micro-positive pressure nitrogen into battery packs. The effect and feasibility of this technology are systematically analyzed by module thermal runaway tests of packs with/without micro-positive pressure nitrogen and the condensation test in the pack. The test results show that： （1） Under the condition of no suppression measures， the thermal runaway of all modules occurs within 140 s， and the battery burns continuously for 413 s， with a maximum surface temperature of 627.6 °C. The insulation resistance of the battery pack is 43.7 MΩ， reduced by 95%， and the internal relative humidity increasing by about 200%. （2） Under the condition of micro-positive pressure nitrogen suppression， only one battery explosion-proof valve of the battery module is opened， with the opening time 526 s later than that without nitrogen， and the maximum temperature on the battery surface is 148 ℃， so it is comprehensively judged that there is no thermal runaway in the battery module. Under the condition of micro-positive pressure nitrogen， there is no condensation in the pack， which effectively inhibits the respiration effect and reduces the risk of secondary disasters such as an insulation short circuit in the pack. This study provides new research ideas for the design and safety prevention and control technology of high-capacity lithium iron phosphate battery packs. Figures and Tables | References | Related Articles | Metrics

Select

Research on Internal Temperature Estimation and ISC Diagnosis Strategy of Lithium-ion Batteries Based on EIS Junqiu Li,Ziming Liu,Zhengnan Liu,Zhixiong Chai 2025, 47 (6):  1112-1121.  doi: 10.19562/j.chinasae.qcgc.2025.06.010 Abstract ( 331 )   HTML ( 9 )   PDF (5441KB) ( 439 )   Save Internal short circuit （ISC） fault of lithim-ion batteries in new energy vehicles， as a critical stage in the evolution of battery thermal runaway， pose significant threats to battery safety. There is currently a lack of relevant technologies for real-time diagnosis of ISC in lithium-ion batteries，but electrochemical impedance spectroscopy （EIS） technology has shown great potential for ISC diagnosis. In this paper， a research on online ISC diagnosis strategies is conducted for lithium-ion batteries based on EIS measurement chips. A DNB chip based EIS online measurement scheme for lithium-ion batteries is constructed， and EIS measurement experiments for ISC batteries are completed. The EIS response and internal average temperature change laws under ISC are obtained， and temperature sensitive impedance characteristic frequencies are extracted. The experiments show that the measurement scheme has a relative error of impedance modulus less than 5% in the frequency range of 1 kHz-0.1 Hz， and an impedance phase angle measurement error less than 2° below 100 Hz. A model for estimating the average internal temperature of batteries based on impedance phase angle is established， and a real-time diagnosis strategy for ISC based on EIS temperature monitoring is developed. Four series connected LFP batteries experiments show that compared with traditional surface temperature based diagnosis strategies， this strategy shortened the diagnosis time by 1 400 s， with the accuracy of ISC resistance estimation increased by 30%. Figures and Tables | References | Related Articles | Metrics

Select

MSF-Diffuser: A Multi-sensor Adaptive Fusion Autonomous Driving Method Based on Diffusion Model Under BEV Mingchen Wang,Hai Wang,Yingfeng Cai,Long Chen,Yicheng Li 2025, 47 (6):  1122-1132.  doi: 10.19562/j.chinasae.qcgc.2025.06.011 Abstract ( 346 )   HTML ( 11 )   PDF (3315KB) ( 322 )   Save Autonomous driving algorithms are a major research focus in the field of intelligent vehicles. Currently， to achieve panoramic autonomous driving， most domestic approaches use multi-sensor fusion. However， existing solutions face problems such as low sensor utilization and unreasonable fusion strategies. For these problems， in this paper， an autonomous driving framework based on multi-sensor fusion （camera+LiDAR+Radar） under a bird's-eye view （BEV） is proposed. In this framework， dual encoding based on point and velocity is used， coupled with feature interaction to extract millimeter-wave radar point cloud features， thereby enhancing the utilization of millimeter-wave radar information and facilitating subsequent fusion. In the fusion module， LSTM is used to store the features from multiple modalities as well as the fused BEV features， which allows for the calculation of feature consistency loss between different modalities and continuity loss for the fused BEV features and historical frames， leading to smoother and more precise feature fusion. Finally， the diffusion model is introduced and the Multi-modal U-Net is proposed for denoising， which improves the robustness of trajectory planning. Extensive experiments are conducted using the CARLA simulator on the authoritative Longest-06 benchmark and Town-05 Long benchmark， getting a DS （Driving Score） of 73.80\pm1.01 and 73.7\pm1.3 respectively. The results show that the proposed approach achieves better panoramic autonomous driving with superior performance and flexibility compared to existing methods. Figures and Tables | References | Related Articles | Metrics

Select

Autonomous Vehicle Object Detection by LiDAR Point Cloud Feature Completion in Snowfall Scenarios Lingyun Zhu,Haiyang Wang 2025, 47 (6):  1133-1143.  doi: 10.19562/j.chinasae.qcgc.2025.06.012 Abstract ( 265 )   HTML ( 6 )   PDF (5917KB) ( 134 )   Save Under snowy conditions， interference from snowflakes on LiDAR leads to point cloud feature loss， significantly degrading the accuracy of 3D object detection models. In this paper， a Transformer-based feature completion algorithm for snow-affected point clouds is proposed. Firstly， a point cloud loss completion module is designed to jointly extract missing features from raw point clouds using multi-head attention mechanisms and mixed density networks. Subsequently， an encoder-decoder architecture is constructed for feature generation， coupled with a fusion redefinition module that aligns features via channel attention mechanisms. Finally， the bounding box prediction strategy is optimized to enhance detection reliability. Experimental results demonstrate that the proposed method achieves improvement of 2.06% and 2.73% in car and pedestrian detection accuracy on the CADC dataset， and a 1.51% average precision gain across three object categories on the KITTI dataset. Quantitative analysis of snowfall intensity and point cloud generation patterns further validates the robustness and engineering applicability of the proposed method. Figures and Tables | References | Related Articles | Metrics

Select

Research on Environmental Perception Information Unified Fusion Method of Intelligent Vehicle Based on Interactive Multiple Models Xin Jia,Songlin Li,Yuansheng She,Feng Hong 2025, 47 (6):  1144-1154.  doi: 10.19562/j.chinasae.qcgc.2025.06.013 Abstract ( 261 )   HTML ( 6 )   PDF (3818KB) ( 88 )   Save For the problem that current multi-sensor information fusion in intelligent vehicle environmental perception systems often involves phased fusion of different sensors， making it difficult to balance the accuracy advantages of individual sensors and the redundancy advantages of multi-source information， an object-level parallel-structured unified multi-sensor information fusion method based on the interacting multiple model （IMM） is proposed in the paper. Object-level fusion has excellent modularity and encapsulation. The parallel structure can fully utilize information redundancy advantages， and the interacting multiple models enable unified and efficient fusion of multi-source data， compensating for the limitation of individual sensors. After spatiotemporal alignment of multi-source sensor data， the nearest neighbor method and DS evidence theory are used to achieve multi-sensor information association， and then dynamic unified fusion based on the interacting multiple models is conducted. Real-vehicle experiments are conducted using millimeter-wave radar and a vision system for environment perception. The results show that the proposed method effectively improves the reliability and stability of target vehicle perception and tracking， enhancing the adaptability of the system. Figures and Tables | References | Related Articles | Metrics

Select

Lane-Level LiDAR-Visual Fusion SLAM in Autonomous Driving Environment Qinglu Ma,Qiuwei Jian,Meiqiang Li,Zheng Zou 2025, 47 (6):  1155-1168.  doi: 10.19562/j.chinasae.qcgc.2025.06.014 Abstract ( 271 )   HTML ( 1 )   PDF (10587KB) ( 76 )   Save To enhance the road environment perception capability of autonomous vehicles during multi-lane driving and operations， a LLV-SLAM （lane-level LiDAR-visual fusion SLAM） for autonomous driving environment is proposed and a real-time localization and mapping algorithm （simultaneous localization and mapping， SLAM） suitable for LiDAR-Visual fusion is developed. Firstly， histogram equalization is introduced based on visual feature point extraction， and in-depth information of feature points is obtained using LiDAR. Visual feature tracking is employed to improve the robustness of the SLAM system. Secondly， visual keyframe information is used to correct the motion distortion of LiDAR point clouds， and LeGO-LOAM （lightweight and ground-optimized lidar odometry and mapping on variable terrain） is integrated with ORB-SLAM2 （oriented FAST and rotated BRIEF SLAM2） to enhance loop closure detection and correction performance， thereby reducing cumulative errors in the system. Finally， the pose obtained from the visual images is transformed into the coordinate system as the initial pose for the LiDAR odometry， assisting the LiDAR SLAM in 3D scene reconstruction. The experimental results show that the fused LLV-SLAM method outperforms traditional SLAM algorithms in several key aspects. It reduces the average localization latency by 41.61%. Furthermore， the average localization errors in the x， y， and z directions are reduced by 34.63%， 38.16%， and 24.09%， respectively. Rotational errors in the roll， pitch， and yaw angles are also reduced by 40.8%， 37.52%， and 39.5%， respectively. Additionally， the LLV-SLAM method effectively mitigates scale drift in the LeGO-LOAM algorithm， improves real-time performance and robustness， meeting perception requirements in multi-lane road environment. Figures and Tables | References | Related Articles | Metrics

Select

Three-Dimensional Multi-object Tracking Algorithm Based on Angle Intersection over Union and Adaptive Lifecycle Xinyu Sun,Lisheng Jin,Zhen Huo,Huanhuan Wang,Yang He,Dong Liu 2025, 47 (6):  1169-1176.  doi: 10.19562/j.chinasae.qcgc.2025.06.015 Abstract ( 143 )   HTML ( 3 )   PDF (4426KB) ( 120 )   Save For the problems of trajectory error association and premature deletion that intelligent vehicles face when tracking surrounding objects in real traffic environment， a three-dimensional multi-object tracking algorithm based on angle intersection over union and adaptive lifecycle is proposed in this paper. Firstly， low-confidence and overlapping detection boxes are removed using score filtering and non-maximum suppression， respectively， while inter-frame displacement of trajectories is eliminated by combining the constant velocity model and Kalman filter. Secondly， the angle intersection over union is designed by considering the factors of position and angle， and the optimal matching of the bipartite graph is determined through application of the Hungarian algorithm. Finally， an adaptive lifecycle strategy is formulated based on the first principle of the relationship between trajectory interruption and object distance to dynamically manage the trajectory status. The experiments show that the improved method attains 0.07%， 0.27%， and 0.29% Mismatch for Vehicle， Cyclist， and Pedestrian， respectively， on the Waymo dataset， a reduction of 0.03%， 0.18%， and 0.21% compared to the baseline algorithm. On the nuScenes dataset， the proposed method obtains an IDS of 312， a reduction of 49.9% compared to the baseline algorithm. The proposed angle intersection over union and adaptive lifecycle is able to reduce the number of identity switches， while the improved three-dimensional multi-object tracking algorithm can achieve accurate and stable temporal trajectories. Figures and Tables | References | Related Articles | Metrics

Select

V2X Vehicle-Parking Cooperative Perception for the Whole-Area of Underground Parking Lots Zhaozheng Hu,Juan Tan,Jianan Zhang,Changjun Yang,Na Cui,Jie Meng 2025, 47 (6):  1177-1187.  doi: 10.19562/j.chinasae.qcgc.2025.06.016 Abstract ( 221 )   HTML ( 4 )   PDF (8934KB) ( 68 )   Save Accurate environmental perception is fundamental to the realization of Automated Valet Parking （AVP） functions. Traditional AVP systems primarily rely on single-vehicle perception； however， with the continuous development of intelligent technologies at parking facilities， collaborative interaction between vehicles and facilities has become an inevitable trend for the implementation of AVP. A V2X （Vehicle-to-Everything） collaborative whole-areas perception method for underground parking lots is proposed， transforming the global perception challenge into a large-scale graph model problem. By integrating sensor data from facility-side lidar and cameras， along with perception data from connected vehicles， the method establishes various edge constraints based on vehicle poses. To enhance perception accuracy， a large-scale graph model method that incorporates lane-level map information is proposed in this paper， which takes parked vehicles as semi-static constraints while integrates lane-level map data for lateral constraints. A sliding window is introduced during the solving process to reduce the scale of the graph model， with final perception results output in map form for vehicle use. Through simulation experiments and field experiments in underground parking lot scenarios with an area of over 2 500 square meters， the results show that this method significantly improves the perception ability in complex parking lot environment and achieves whole-area perception of underground parking lots. Figures and Tables | References | Related Articles | Metrics

Select

Vehicle Target Detection Algorithm Based on Lightweight RT-DETR-tiny Aiqi Long,Zhiguo Feng,Zhenbo Zhang,Xingqiang Tian,Wei Xiang 2025, 47 (6):  1188-1197.  doi: 10.19562/j.chinasae.qcgc.2025.06.017 Abstract ( 417 )   HTML ( 10 )   PDF (3812KB) ( 228 )   Save For the hardware limitation in autonomous driving scenarios and the challenges faced by lightweight algorithms in detecting small target vehicles， a novel lightweight vehicle detection algorithm， RT-DETR-tiny， is proposed. Firstly， a new Redundant Graph Rapid Generation Module （ReduFast block） is proposed， which uses a cascade feature extraction structure to avoid the loss of small target feature information caused by redundant information， and reduce the computational redundancy. The lightweight network ReduFastNet， designed based on this module， serves as the feature extraction network， achieving faster inference speed compared to other lightweight networks. Then， during the feature fusion stage the DGSTM module is incorporated to further streamline the model， while the EAAIFI module is designed to ensure real-time performance during feature fusion. Finally， for the problem of boundary boxes being susceptible to noise in small target vehicle detection， the DIOU is introduced to optimize the original lossoriginal loss function， enhancing the accuracy of the target center positions and mitigating excessive penalties of the model caused by on aspect ratio fluctuations of the prediction box. Experimental results demonstrate that， on the BDD100K-Urban nighttime dataset， the proposed algorithm achieves a detection accuracy of 75.3%， with only a 0.1% loss， while parameters and computational load decreases by 37.1% and 33.5%， respectively， achieving a detection frame rate of 45.1 frames per second and enhancing detection speed by 5 percentage points. In comparison to other mainstream lightweight object detection models on the UA-DETRAC-Small Car dataset， RT-DETR-tiny balances high detection accuracy and minimal parameter count and computational load， outperforming similar object detection algorithm， which facilitates accuracy and edge deployment for real-time vehicle detection in autonomous driving contexts. Figures and Tables | References | Related Articles | Metrics

Select

PolarSparse4D: Polar Parametrization for Vision-Based Surround-View Temporal Sparse 3D Object Detection Chao Wei,Shuxin Sui,Luxing Li 2025, 47 (6):  1198-1206.  doi: 10.19562/j.chinasae.qcgc.2025.06.018 Abstract ( 235 )   HTML ( 3 )   PDF (4655KB) ( 38 )   Save To address the trade-off between accuracy and real-time performance in vision-based surround-view 3D object detection for autonomous vehicles， PolarSparse4D， a sparse query-based method using polar parametrization， is proposed. The model consists of an image encoder， a 3D anchor decoder and an auxiliary quality assessment branch for training. Firstly， to avoid the detection distance limitation caused by parameter normalization， a feature encoding method that decouples the center distance and azimuth angle of the 3D anchor boxes is designed. Secondly， by designing an anchor spatial information interaction self-attention module and a temporal information fusion module， the spatiotemporal information fusion process of anchors is completed efficiently and accurately. Finally， an anchor box parameter quality assessment branch is established to improve the detection accuracy and model convergence speed significantly. The experimental results on the nuScenes validation set show that the proposed model achieves 41.3% and 52.5% on mAP and NDS， respectively， with a speed of 19.2 FPS， demonstrating high accuracy and real-time capability. Figures and Tables | References | Related Articles | Metrics

Select

Intelligent Tire Wear Detection Method Based on an Embedded Sensor Array Within the Tire Lisheng Jin,Xin Zhao,Xianyi Xie,Hao Yang,Bo Lu,Mingliang Song,Baicang Guo,Yaoguang Cao 2025, 47 (6):  1207-1218.  doi: 10.19562/j.chinasae.qcgc.2025.06.019 Abstract ( 314 )   HTML ( 11 )   PDF (6409KB) ( 146 )   Save In order to fully utilize the advantages of embedded integrated sensor arrays in smart tires for collecting multi-modal information of the tire-ground contact state， and to enhance the accuracy of tire wear detection， a smart tire wear detection method based on the embedded sensor array is proposed in this paper. Firstly， a sensor array composed of accelerometers and PVDF piezoelectric film sensors is constructed， and an embedded data acquisition system for the smart tire is designed to collect sensor array data from tires with various degrees of wear under different operating conditions. Next， the waveform data from the sensor array is processed using Butterworth filtering， and multidimensional time-domain features are extracted. The variations in the time-domain characteristics of the sensor array data are analyzed under changing vehicle operating conditions， revealing significant differences in the time-domain feature variations （length features and area features） of the accelerometer and PVDF piezoelectric film sensors with the change of vehicle speed and vertical load. Finally， a joint feature set that integrates the time-domain feature information from both types of sensors is established， and a machine learning model for tire wear detection is constructed. The test results show that the average absolute error of tire wear detection based on the sensor array is 0.13 mm， a reduction of 67.67% and 56.81%， respectively， compared to using only the accelerometer or only the PVDF piezoelectric film sensor. The detection accuracy of tire wear within a 0.3 mm error reaches 88.81%， while the accuracy within a 0.5 mm error reaches 96.42， which proves the effectiveness and accuracy of the machine learning model for tire wear detection based on the sensor array. Figures and Tables | References | Related Articles | Metrics

Select

Research and Application of Parallel Upgrade Method Based on Circle Ethernet Network Shuai Tan 2025, 47 (6):  1219-1230.  doi: 10.19562/j.chinasae.qcgc.2025.06.020 Abstract ( 201 )   HTML ( 7 )   PDF (4148KB) ( 97 )   Save In this article， firstly the shortcomings of traditional electronic appliance architectures and the advantages of Circle Ethernet Network architecture as well as the drawbacks of existing serial upgrade strategies are elaborated. Subsequently， the parallel upgrade method based on Circle Ethernet Network architecture is studied and the multi-level queue-scheduling algorithm is proposed. Finally， based on the aforementioned research， the method is implemented and applied in experimental tests on a certain new energy vehicle brand. The experimental results show that the parallel upgrade method proposed in this article can accelerate the upgrade speed， improve the efficiency of new energy vehicle development， production， and after-sales maintenance. Figures and Tables | References | Related Articles | Metrics