Loading...
Welcome to Journal of Automotive Safety and Energy,

Current Issue

  • 2025, Vol. 16 No. 3 Published on:30 June 2025 Previous issue    Next issue
    Review, Progress and Prospects
    Review on the emission characteristics and near-zero emission control for ammonia-hydrogen internal combustion engines
    LI Liguang, SHANG Quanbo, TANG Yongjian, DENG Jun
    2025, 16(3):  345-366.  doi:10.3969/j.issn.1674-8484.2025.03.001
    Abstract ( 352 )   HTML ( 1319)   PDF (3877KB) ( 2108 )  

    Facing to the background of global efforts to reduce carbon emissions and transition toward low- and zero-carbon energy systems in response to climate change, ammonia-hydrogen internal combustion engines have emerged as a promising and increasingly studied solution in the transportation sector due to their potential for zero carbon emissions. Ammonia offers several advantages as a fuel, including high hydrogen energy density, ease of storage and transport, and excellent anti-knock properties. However, its inherently slow combustion characteristics and nitrogen-containing nature bring significant challenges, particularly in terms of high nitrogen oxide (NOx), unburned ammonia (NH3), and nitrous oxide (N2O) emissions. Optimizing the combustion process of ammonia-hydrogen fuels and achieving near-zero emissions in internal combustion engines (ICEs) are relatively new research areas that presents formidable technical challenges.

    This paper reviews the latest research developments in the emission characteristics and near-zero emission control strategies for ammonia-hydrogen fueled ICEs. First, in terms of emission mechanisms, NOx formation during ammonia combustion is governed by complex pathways and is highly sensitive to equivalence ratio, pressure, and temperature. Earlier mechanistic studies focused primarily on low-pressure and medium-to-high temperature conditions, which differ significantly from the high-pressure, high-temperature environments of modern engines, highlighting a current gap in research. Second, in-cylinder pollutant formation and control remain key to emission reduction. In-cylinder control techniques, including optimization of fuel injection strategies, ignition timing, and intake conditions can effectively balance the relationship between emissions and thermal efficiency. Studies have shown that hydrogen enrichment can improve combustion efficiency and reduce NH3 and N2O emissions, though it may increase NOx formation. Lastly, aftertreatment technologies are critical to achieving near-zero emissions. Due to the unique characteristic of emissions from ammonia-hydrogen combustion, new dedicated aftertreatment systems are required. These include selective catalytic reduction (SCR) for NOx, ammonia slip catalysts (ASC), and strategies for addressing high global warming potential gases such as N2O. Additionally, hydrogen-selective catalytic reduction (H2-SCR) offers a novel pathway for mitigating hydrogen-related emissions in such engines. Future researches should focus on the synergistic optimization of in-cylinder combustion and specific aftertreatment systems, the development of low-temperature, high-efficiency catalysts, and the exploration of integrated aftertreatment solutions to meet increasingly stringent emission regulations and approach near-zero emissions. While ammonia-hydrogen dual-fuel ICEs hold significant promise in achieving carbon neutrality, their widespread adoption will require overcoming several technical challenges, particularly in emission control.

    Figures and Tables | References | Related Articles | Metrics
    Automotive Safety
    Optimization design for automobile ORSs based on composite deep Gaussian process regression network
    WANG Wenjie, SUN Yi, LIU Zhao, ZHU Ping
    2025, 16(3):  367-375.  doi:10.3969/j.issn.1674-8484.2025.03.002
    Abstract ( 122 )   HTML ( 9)   PDF (1509KB) ( 59 )  

    A data-driven optimization method was investigated for automobile occupant restraint systems (ORS) based on composite deep Gaussian process regression network to improve the safety performance and to develop the efficiency of the ORS. In terms of the prediction of occupant dummy injury values, an improved composite deep Gaussian process regression network was proposed as the prediction model by combining neural network architecture with Gaussian process regression. Based on the prediction results, the ORS parameter optimization was carried out by using the group-based crow search algorithm. The method’s effectiveness was verified by using engineering simulation data. The results showed that this ORS design reduces the dummy injuries by up to 30.77% with an average of 12.11% compared to the original engineering scheme. Therefore, the method can predict the injury values for multiple parts of the dummy with a high-quality ORS design.

    Figures and Tables | References | Related Articles | Metrics
    Comparative analysis of driver response between Chinese and western drivers under various collision conditions
    LIU Yu, ZHANG Huida, WU Xiaofan, JIANG Han, LI Guibing
    2025, 16(3):  376-385.  doi:10.3969/j.issn.1674-8484.2025.03.003
    Abstract ( 129 )   HTML ( 10)   PDF (3044KB) ( 184 )  

    A finite element (FE) occupant model of Chinese 50th% male was developed to fully understand the differences in response to impact loading between Chinese and Western, based on the existing occupant model in Western 50th% size and the geometric scaling of the body surface of a Chinese 50th% male volunteer. Simulations were conducted under 4 collision conditions: the SOB (small overlap barrier), the MPDB (mobile progressive deformable barrier), the FRB (full width rigid barrier) and the FarSide by using the subsystem model of occupant compartment and the FE human body models (HBM). Differences in kinematic, dynamic, and biomechanical responses of the Chinese and the Western 50th% male drivers under typical collision conditions were compared and analyzed. Compared with the response of Western 50th% HBM, the HIC (head injury criteria) values of Chinese 50th% HBM under SOB and FRB conditions are 15.48% and 26.2% lower, while the HIC value in MPDB case is 50% higher; in FRB case, the Chinese 50th% HBM has a chest compression rate 31.44% lower, but exhibits higher rib strain; The maximum lateral displacement of the head under FarSide conditions is similar, but the Chinese 50th% HBM has a significantly lower brain strain, and the lateral shear/axial force on the neck and chest compression rate of the Chinese 50th% HBM are lower more than 20%. Therefore, the vehicle restraint systems designed for westerner might be difficult to achieve optimal protection for Chinese people.

    Figures and Tables | References | Related Articles | Metrics
    Evaluation on the complexity of scenarios for VRU on urban roads based on self-organizing K-means
    CHENG Rui, LU Chuncheng, YUAN Quan, CUI Tao, To. Jeremy, WANG Tao
    2025, 16(3):  386-395.  doi:10.3969/j.issn.1674-8484.2025.03.004
    Abstract ( 121 )   HTML ( 3)   PDF (1538KB) ( 165 )  

    In order to address the requirements of high-risk testing environments for validating intelligent vehicle collision avoidance systems, while simultaneously to enrich the content and methods for evaluating autonomous driving scenarios involving vulnerable road users (VRU). This study collected and systematically analyzed traffic accident cases in Guilin City, Guangxi Province, from 2016 to 2020. A total of 1 429 vehicle-VRU collision accident data were screened. Based on accident investigation experience, 13 risk factors were identified, and 10 typical vehicle-VRU collision scenarios applicable to urban traffic conditions in China were constructed using self-organizing K-means clustering analysis. An evaluation model for the complexity of VRU scenarios was established utilizing information entropy theory. The state of variables and the weight of each dimension were determined through a combination of logistic regression models and back propagation (BP) neural networks, and the complexity of various scenarios was calculated. Additionally, the Gaussian mixture model was employed to cluster the complexity levels, resulting in four distinct scene complexity categories. The results show that on roads with a speed limit of 30 km/h, the nighttime side collision between a straight-moving car and an electric bicycle crossing the road outside a pedestrian crossing area is the most complex scenario. The findings in this study provide an experimental scenario reflective of urban road characteristics in China for intelligent vehicle safety testing and offer a basis for the formulation of external VRU collision avoidance strategies and decision-making.

    Figures and Tables | References | Related Articles | Metrics
    Scenarios of electric two-wheeler and pedestrian collision accidents based on video information
    YANG Yao, WANG Bingyu, ZHANG Xiang, ZHANG Yue
    2025, 16(3):  396-404.  doi:10.3969/j.issn.1674-8484.2025.03.005
    Abstract ( 135 )   HTML ( 5)   PDF (1701KB) ( 110 )  

    In order to obtain the characteristic variables and extract typical danger scenarios in electric two-wheeler and pedestrian collision accidents, 273 cases of electric two-wheeler and pedestrian collision cases with video information were collected from the Internet. Furthermore, 13 accidents characterization variables, such as collision speed, collision angle, pedestrian's first collision point, and pedestrian's first touchdown site were analyzed via using descriptive statistical methods. K-modes method was then used to obtain the typical scenarios by conducting cluster analysis for the 13 characteristic variables. The results show that the driving speed of electric two-wheeler is generally lower than 30 km/h and the collision position mostly concentrates in the front area of electric two-wheeler. Electric two-wheeler and pedestrian collision accident scenarios can be divided into five typical scenarios: vertical collision scenario, nighttime same-direction collision scenario, same-direction side-impact scenarios, nighttime collision scenario of riders with helmets and opposite-direction collision scenario. This study can provide a reference for the study of pedestrian safety countermeasures and safety test scenarios in electric two-wheeler-pedestrian collisions.

    Figures and Tables | References | Related Articles | Metrics
    Intelligent vehicle grading warning and braking system based on projection warning
    ZOU Tiefang, FU Xijun, LI Yanchun
    2025, 16(3):  405-413.  doi:10.3969/j.issn.1674-8484.2025.03.006
    Abstract ( 132 )   HTML ( 5)   PDF (1407KB) ( 54 )  

    In order to improve the active safety performance of vehicles and reduce the number of human and vehicle collisions in the scene of blind area occlusion, an autonomous vehicle hierarchical early-warning braking system was proposed based on intelligent projection, focusing on the risk warning of smart car and pedestrian accidents.The hierarchical control strategy was used to model the braking system. The upper controller was a safety decision-making method for graded warning and braking systems, which explored the safe speed in the blind area through the change of the moving position of the vehicle and the pedestrian. The first stage braking was carried out according to the comparison between the safe speed in the blind area and the current speed, and the second stage braking was carried out when there was a collision risk with the pedestrian. The lower actuator established the inverse longitudinal dynamics model of the vehicle, and the throttle opening model and brake pressure control model of the vehicle were established according to the expected deceleration speed output of the upper controller, so as to maintain or decelerate the current speed. At the same time, the system used the road projection technology of smart cars to project warning information on the ground ahead, and carried out information interaction with pedestrians. The safety and feasibility of this system were verified by taking the traditional classic MAZDA model as the reference model. After 1 260 simulations, it was found that the hierarchical early-warning braking system did not avoid 14 accidents. For the unavoidable cases, 336 deep pedestrian interaction simulations were further conducted to simulate the real reactions of pedestrians after seeing the projected warnings. The results show that within the normal reaction capacity of pedestrians, the hierarchical early-warning braking system can avoid all accident cases. The research results can provide support for the research of automatic emergency braking system to protect pedestrian safety and promote its practical application.

    Figures and Tables | References | Related Articles | Metrics
    Vehicle state UKF estimation considering noise and initial state uncertainties
    ZHANG Zhiyong, DU Chenzhou, YI Sheng, YU Hui
    2025, 16(3):  414-424.  doi:10.3969/j.issn.1674-8484.2025.03.007
    Abstract ( 97 )   HTML ( 6)   PDF (2035KB) ( 50 )  

    An improved unscented Kalman filter (UKF) vehicle state estimation method was proposed to improve the estimation accuracy of vehicle states in the presence of noise covariance matrix and initial state uncertainties. This method introduced a windowing process based on the maximum a posteriori (MAP) estimation strategy to achieve dynamic estimation of the noise covariance matrix, while also integrating a static particle filter (SPF) algorithm to estimate the initial vehicle states. The improved UKF's estimation accuracy was verified using a co-simulation platform with CarSim and MATLAB/Simulink. The results show that, when measurement noise deviates from the true value, the windowed MAP dynamic estimation method for the noise covariance matrix improves the estimation accuracy of longitudinal and lateral speeds by 90% and 80%, respectively, compared to the standard UKF. In comparison to the UKF with adaptive noise covariance matrix adjustment, the estimation accuracy increases by 75% and 56%, respectively. Under initial state uncertainty, the SPF method improves the estimation accuracy of longitudinal and lateral vehicle speeds by 94% and 90%, respectively. Therefore, the proposed improved UKF estimation method significantly enhances estimation accuracy and robustness in the presence of noise covariance matrix and initial state uncertainties.

    Figures and Tables | References | Related Articles | Metrics
    Automotive Energy Efficiency and Environment Protection
    Characterization of ammonia emissions from urban motor vehicles in Taiyuan based on MOVES model
    YANG Yang, WANG Yichun, ZHANG Yi, YU Qian, ZHANG Tao
    2025, 16(3):  425-433.  doi:10.3969/j.issn.1674-8484.2025.03.008
    Abstract ( 88 )   HTML ( 4)   PDF (1113KB) ( 182 )  

    In order to investigate the ammonia emission characteristics of road motor vehicles, the Motor Vehicle Emission Simulator (MOVES) model was used to localize the model parameters with Taiyuan City as the research object, and the calculation method of the emission standard percentage of different models was proposed while improving the accuracy of parameter localization, and finally the ammonia emission inventory of motor vehicles was obtained and the ammonia emission characteristics of motor vehicles in Taiyuan City were analyzed from four perspectives of models, time, speed and space, respectively. The results show that the ammonia emissions of motorcycles, private cars, large buses, minivans, and large trucks in Taiyuan in 2021 were 4.2, 441.8, 5.05, 130.4 tons, and 49.4 tons, respectively; the order of the average ammonia emissions per unit mileage of different vehicle types was large trucks > large buses > minivans > private cars > motorcycles; the lower ammonia emissions can be obtained when the driving speeds of on-road motor vehicles is in the range of 25 to 65 km/h. The research results can provide a theoretical reference for regulating ammonia emissions from road motor vehicles.

    Figures and Tables | References | Related Articles | Metrics
    High-rigidity lightweight design of pure electric highway bus body structure
    ZHANG Xiao, HE Feng, LIU Dingyi, GONG Chengping
    2025, 16(3):  434-442.  doi:10.3969/j.issn.1674-8484.2025.03.009
    Abstract ( 107 )   HTML ( 3)   PDF (3603KB) ( 134 )  

    A multi-objective structural design was optimized for a pure electric coach body frame to enhance the stiffness and reduce the mass. A finite element model of the body frame was established, with the specific pure electric coach as the research subject, to determine weight coefficients for static and dynamic working conditions. The modified frame structure underwent variable grouping and multi-condition multi-objective dimensional optimization. Static analysis and modal analysis were subsequently conducted. The results show that the optimized body frame achieve a minimum 4% enhancement in comprehensive average stiffness with a 7% improvement in average dynamic frequency of the first four orders, and with a 1% mass reduction. Therefore, it makes vehicle frame performance being effectively upgraded while maintaining original mass or achieving controlled lightweight effects.

    Figures and Tables | References | Related Articles | Metrics
    Design optimization of a novel branched-channel fin liquid cooling plate for lithium-ion batteries
    XIE Yuanpeng, ZHANG Furen, TAN Haikun, XIAO Kang, QIU Shuaishuai, SUN Shizheng
    2025, 16(3):  443-451.  doi:10.3969/j.issn.1674-8484.2025.03.010
    Abstract ( 132 )   HTML ( 6)   PDF (3200KB) ( 123 )  

    A new type of liquid cooling plate structure based on finned type was proposed to improve the comprehensive performance of the battery thermal management system. Firstly, a rhombic finned plate liquid cooler was selected as the experimental object, and the influence of different fin structures and different fin arrangements on the heat dissipation performance of the liquid cooler was investigated. Secondly, a new fin structure was proposed based on the optimized finned plate. Finally, the NSGA-II algorithm was used for multi-objective optimization, and it was analyzed that the influence of structural parameters (fin opening wide, fin outer contour thickness, fin inner cavity wide) on the average temperature and pressure drop. The results show that the elliptical finned plate liquid cooler with a 3×8 arrangement has a 9% improvement in comprehensive performance compared to the rhombic finned plate liquid cooler, with an average temperature drop of about 0.1℃ and a pressure drop reduction of 0.43 Pa. After adding branch channels inside the elliptical fins, the average temperature of the new fin structure after multi-objective optimization drops by 0.257℃ (0.7%), and the pressure drop drops by 0.784 Pa (16.31%), indicating that the cooling performance of the liquid cold plate has been effectively improved, thereby providing a new approach and theoretical basis for the design and analysis of heat sinks.

    Figures and Tables | References | Related Articles | Metrics
    Intelligent Driving and Intelligent Transportation
    Night lane detection method based on deep generation network
    LIU Guosheng, SU Xiner, WANG Jianfeng, LIU Zhenwei
    2025, 16(3):  452-462.  doi:10.3969/j.issn.1674-8484.2025.03.011
    Abstract ( 183 )   HTML ( 68)   PDF (8484KB) ( 245 )  

    In order to ensure the safe driving of vehicles at night, the night lane lines were accurately recognized and lane departure warnings were made, a deep generative network EnhanceGAN for nighttime image enhancement and an end-to-end lane line detection network AttentiveLSTR based on Transformer were proposed for nighttime lane line detection, and experiments with real vehicles were conducted. The deep generative network EnhanceGAN used the improved UNet as the generator of the network, adopted a two-layer nested U-shape structure to expand the sensory field, and added a Markov local discriminator and a combined loss function to enhance the detailed information of lane line edges and textures. The lane line detection network AttentiveLSTR used ResNeXt as a feature extraction network to ensure the network depth and reduced the number of model parameters, and introduced feature pyramid networks (FPN) to extract lane line edge and shape information. The results show that compared with the mainstream methodsCycleGAN and Gamma Correction, the pro[osed method is more effective in nighttime image enhancement on the BDD100k dataset, with a high contrast between lane lines and surrounding environment, structural similarity (SSIM) of 0.883 4, natural and realistic images as a whole, peak signal-to-noise ratio (PSNR) of 40.265 4, and natural image quality evaluation index (NIQE) of 3.423 3; the detection accuracy (Acc) on the CULane dataset is 90.12%, and the processing speed is fast, with 82 frames per second (FPS). The research results can provide a reference for nighttime lane line deviation scenarios.

    Figures and Tables | References | Related Articles | Metrics
    Highway traffic flow prediction approach based on multi-dimensional attention mechanism
    YU Anjun, LI Yingdi, YANG Zheyi, FU Chongyu, TONG Weiping, YU Jia, LIU Yunhai, LIU Zhiyuan
    2025, 16(3):  463-469.  doi:10.3969/j.issn.1674-8484.2025.03.012
    Abstract ( 127 )   HTML ( 4)   PDF (1378KB) ( 88 )  

    A traffic flow prediction model based on a multi-dimensional attention mechanism was proposed to achieve precise traffic flow prediction and enhance the intelligent management level of expressways. Comparative experiments were conducted on real traffic datasets from the Zhangji Expressway to verify the accuracy and predictive accuracy of the model. The model extracted spatial and temporal features of traffic flows using graph neural networks (GNN) and temporal convolutional networks (TCN), respectively. It integrated a multi-dimensional attention mechanism to mine key information within spatiotemporal data. Additionally, a multi-task learning architecture was introduced, employing a loss function based on homoscedastic uncertainty to balance the joint learning of different tasks, thereby enhancing the generalization ability and robustness of the model. The results show that the root mean square error (RMSE) and mean absolute error (MAE) of the model on the test set are 7.467 and 5.133, respectively, demonstrating superior predictive accuracy compared to baseline models. The proposed prediction method can effectively uncover the spatiotemporal characteristics of traffic flows, describe the actual state of traffic operations, and make accurate predictions of the traffic flow on expressways.

    Figures and Tables | References | Related Articles | Metrics
    Blind spot traffic strategy for intelligent connected vehicles based on deep reinforcement learning
    LI Ziyuan, LIU Qiang, LI Dingli, LI Zilong
    2025, 16(3):  470-477.  doi:10.3969/j.issn.1674-8484.2025.03.013
    Abstract ( 95 )   HTML ( 3)   PDF (1130KB) ( 39 )  

    A blind spot passing strategy method was proposed by using the deep reinforcement learning for intelligent connected vehicles (ICV) to prevent traffic accidents between vehicles and pedestrians when passing through visual blind spots. A mathematical description model was established for typical blind spot scenarios considering three indicators of safety, efficiency and comfort; while a deep reinforcement learning model was designed based on the Double DQN (double deep Q-network) with the TTC (time to collision) indicator to establish a set of physically interpretable reward functions, with the output being the vehicle's accelerator and the brake pedal depth. Simulation experiments were conducted under three scenarios to assess the algorithm efficacy. The results show that the simulation experiments verify the effectiveness of the algorithm. The comfort is increased by more than 50% on average of this method, compared with the traditional DQN method. The method improves decision-making accuracy. Therefore, the longitudinal decision-making method achieves the safety, the efficient and the comfortable.

    Figures and Tables | References | Related Articles | Metrics
    Lane detection algorithm based on adaptive segmentation network
    WANG Yifei, LI Yonghang, ZHANG Yali, WANG Chang, WANG Taiqi, YUAN Huazhi
    2025, 16(3):  478-486.  doi:10.3969/j.issn.1674-8484.2025.03.014
    Abstract ( 109 )   HTML ( 4)   PDF (3525KB) ( 100 )  

    Aiming to address the issues of reliance on prior knowledge and limited adaptability in traditional image segmentation approaches within tunnel scenarios, a lane line detection method was proposed based on an adaptive segmentation network. Firstly, a sub-region planning method based on illumination characteristics was designed, which adaptively determined the necessity of multi-region segmentation by extracting illumination feature signals and provided the corresponding sub-region configuration scheme in real time. Secondly, a lane line area segmentation method was proposed based on improved Otsu. Each sub-region can independently adjust the segmentation threshold according to the lighting characteristics to achieve precise segmentation of the lane line area. Finally, a dynamic region of interest update method was designed to update the region of interest (ROI) of the current frame based on the detection results of the previous frame. The results show that the detection accuracy of the proposed algorithm reaches 96.73%, and the average processing time per frame is 24.77 ms in typical tunnel scenarios such as complex lighting, low illumination, and discontinuous lane lines, indicating that the proposed method has the advantages in detection accuracy, detection efficiency and robustness, and can meet the needs of real-time performance.

    Figures and Tables | References | Related Articles | Metrics
    Strategy decision of man-machine co-driving switching in high-speed operation area based on combination game
    ZHOU Qianxi, WU Chengcheng, HU Zheng, SHI Liangliang
    2025, 16(3):  487-495.  doi:10.3969/j.issn.1674-8484.2025.03.015
    Abstract ( 90 )   HTML ( 2)   PDF (1077KB) ( 73 )  

    To effectively mitigate the bottleneck effect induced by mixed traffic flow and enhance road safety, a dynamic combination game model based on Stackelberg game and non-cooperative Nash game was presented for the allocation decision of control rights of man-machine vehicles in the road section of the highway operation area. VISSIM software was employed to establish a micro-hybrid traffic flow simulation scenario for the sections in the operation area, validate the effectiveness of the decision model, and analyze the influence of the model on traffic flow safety and traffic efficiency. The results show that the effectiveness of the proposed decision model exceeds 66%, the average conflict incidence is decreased by 34.15%, and the highest conflict incidence is reduced by 70.77% under low traffic flow. The model significantly enhances the safety of the road section in the operation area and effectively improves the traffic efficiency in 91.67% of driving scenarios. In the future, the model should be further applied and optimized to achieve more efficient traffic safety.

    Figures and Tables | References | Related Articles | Metrics
    Energy and stability aware path planning for autonomous vehicles in off road environments
    CHEN Xiaofeng, WANG Lanwen, MA Guo, ZHANG Lei, BAO Jiading, JING Hui
    2025, 16(3):  496-503.  doi:10.3969/j.issn.1674-8484.2025.03.016
    Abstract ( 134 )   HTML ( 9)   PDF (2368KB) ( 66 )  

    The traditional A* algorithm was modified in terms of environment modeling and cost function to reduce energy consumption and improve lateral driving stability in off-road path planning for autonomous vehicles. The Digital Elevation Model (DEM) data were processed using the Kriging interpolation and the slope calculation. A vehicle travers ability map was generated based on vehicle performances. Pitch and roll influence factors were added into the cost function of the A* algorithm. The proposed algorithm, the traditional A* algorithm, and the HA* algorithm (Height-Aware A* algorithm) were simulated and compared on the real environment maps. The results show that the proposed algorithm increases path length by up to 7.3% but reduces the energy consumption by up to 10.1% compared with the traditional A* and the HA* algorithms in off-road environments. The driving pitch angle is less than 40% for the general-performance vehicles, with being lower than 60% for the high-performance vehicles; And the roll angle of both types of vehicles is lower than 36%. Therefore, the proposed algorithm reduces energy consumption and improves lateral driving stability.

    Figures and Tables | References | Related Articles | Metrics