Loading...
Welcome to Journal of Automotive Safety and Energy,
汽车安全与节能学报

Current Issue

  • 2020, Vol. 11 No. 1 Published on:31 March 2020 Previous issue    Next issue
    Progress & Prospects
    Progress and key technologies of flying cars
    ZHANG Yangjun, QIAN Yuping, ZHUGE Weilin, ZHANG Lei, PENG Jie, XU Bin, WANG Zexing
    2020, 11(1):  1-16.  doi:10.3969/j.issn.1674-8484.2020.01.001
    Abstract ( 775 )   PDF (4701KB) ( 3217 )  
     Flying cars as a new type of vehicle for urban air transportation and future travel, are getting more and more attention in the automotive and aeronautical fields, it has become an important development trend for the transborder integration of automotive and aeronautical technology and industry. This article introduces the connotation of flying cars and the rise of the urban air mobility (UAM) concept. It describes more than one hundred years of a practical exploration process of flying cars and then represents the current research status for UAM of airbuses and air-taxis along with the future development trend of intelligent transportation of air-land amphibious vehicles. Moreover, it discusses the major bottlenecks and impediments confronting the development of flying cars. In addition to the high power density electric propulsion, high lift-drag ratio and lightweight body structure, and low-attitude flight intelligent driving, as the main approaches and key technologies break through the bottleneck performance of flying cars. Furthermore, it proposes the three-stage goals and vision for the development of flying cars in China, and suggests to build a flying vehicle technology innovation system that integrates automotive and aviation transboundary, and to promote the demonstration application of flying cars in urban air transportation and emergency rescue as the key breakthrough for the near future development of the flying car.
    Related Articles | Metrics
    Shell’s view on future mobility fuels: A patchwork, or “mosaic” approach will be needed to address societies energy needs
    JIA Tian, ZHENG Bin, Warnecke W, Kolbeck A, Aradi A, Kofod M, Clark R, Wilbrand K
    2020, 11(1):  17-35.  doi:10.3969/j.issn.1674-8484.2020.01.002
    Abstract ( 449 )   PDF (19411KB) ( 641 )  
    Facing the dual challenge of increased energy demand and emission reduction, the transition in transport and mobility is inevitable. This paper reviews and evaluates multiple advanced powertrain systems and alternative fuel options, in combination with future powertrain developing trend. These include conventional liquid fossil fuels (i.e. gasoline and diesel) and enabling advanced internal combustion engine technologies, clean alternative fuels [e.g. biofuels, gas-to-liquid (GTL) fuels, power-to-liquid (PTL) fuels, liquified nature gas (LNG) etc], and electrified powertrains (incl. hybrid, battery electric and fuel cells). Fuel options in liquid state is focused. Shell believes there’s no single option able to tackle the complex energy challenge, and thus a mosaic of new fuels and powertrains, applied in different applications respectively, will be needed. We will also introduce 3 Shell scenarios published recently based on analysis of different political, economic and social trajectories, namely Shell Mountain, Ocean and Sky Scenarios, followed by Shell’s view on transition and decarbonization in transport sector.
    Related Articles | Metrics
    Automotive Safety
    Driver reliance characteristics on forward collision warning systems in adjacent vehicle cut-in situations
    LI Guofa，LAI Weijian，LIAO Yuan，WANG Wenjun，CHENG Bo
    2020, 11(1):  36-43.  doi:10.3969/j.issn.1674-8484.2020.01.003
    Abstract ( 361 )   PDF (1285KB) ( 315 )  
     An evaluation method was investigated to assess driver reliance characteristics on forward collision warning systems based on a driving simulator to improve driving safety in adjacent vehicle cut-in situations. Using alarm timing (time to collision, TTC) as the control variable, driving behavior data from 12 participants were collected in adjacent vehicle cut-in situations. Two objective indexes (brake reliance index and secondary task index) and two subjective indexes (risk level index and trust level index) were proposed to establish the evaluation system model to realize the quantitative evaluation of driver reliance level on the systems. An L9(34) orthogonal experiment was designed and conducted. Regression models of driver reliance indexes were established. The results show that alarm timing is the most significant factor affecting driver reliance. A late alarm (TTC = 2.4 s) degrades the effectiveness of the systems, while an early alarm (TTC = 1.2 s) causes drivers’ over-reliance on the systems. Therefore, appropriately delaying the alarm timing (TTC = 1.8 s) can improve driver reliance for safety considerations.
    Related Articles | Metrics
    Analysis of pedestrian-vehicle collision accident characteristics based on the video information from NAIS database
    2020, 11(1):  44-52.  doi:10.3969/j.issn.1674-8484.2020.01.004
    Abstract ( 344 )   PDF (2728KB) ( 792 )  
    Accident characteristics were analyzed for 51 cases of pedestrian-vehicle collision with video information selected from the National Automobile Accident In-Depth Investigation System (NAIS) database of China. The analysis contents included the pedestrian-vehicle collision critical scenarios, the pre-crash relative position of pedestrian-vehicle, the pedestrian collision dynamics response, the pedestrian-vehicle collision wrap around distance (WAD) distribution, the head impact point distribution and so on. The results show that 10 kinds of scenarios extracted basically covered the pedestrian-vehicle collision scenarios. The field of view (FoV) angle is more important for pedestrian protection than that for detection distance. The sedans are easy to cause positive rotation of pedestrians, while the single-compartment vehicles are easy to cause negative rotation of pedestrians. The WAD of pedestrian-vehicle collisions are mainly distributed on both sides of vehicles. The head impact points causing fatal injuries are mainly distributed in the lower half of front windshield, the middle part of the left and right sides, and the vicinity of pillar A. Therefore, using collision video information can improve the analysis accuracy of pedestrian-vehicle collision characteristics.
    Related Articles | Metrics
    Multi-objective optimization design of auto body considering aerodynamic resistance and crosswind stability
    QI Chang, HAN Yuanji, YANG Shu, LU Zhenhua
    2020, 11(1):  53-60.  doi:10.3969/j.issn.1674-8484.2020.01.005
    Abstract ( 309 )   PDF (1629KB) ( 434 )  
    A multi-objective automatic optimization-design of the body shape parameters was carried out considering aerodynamic drag characteristics and cross-wind stability. An automayic optimization process was set up on an optimization platform modeFRONTIER, comprehensively using parametric modeling technology, computational fluid dynamics (CFD), design of experiment (DOE), response surface model (RSM) and intelligent optimization algorithm, integrating Pro / Engineer, ICEM, and Fluent. Using this process, an MIRA fast-back auto body geometry was modified and designed to obtain an optimal trade-off design-solution-set considering the aerodynamic drag characteristics and cross-wind stability based on the genetic algorithm (GA). The result  reduces the aerodynamic drag coefficient by 5.2% and reduces the lateral force coefficient by 5.8%. Therefore,  a multi-objective optimization of aerodynamic resistance and crosswind stability of the vehicle body has been achieved.
    Related Articles | Metrics
    Active anti-roll control method for a semi-trailer based on MPC theory
    YANG Echuan, XIE Chuanren, WANG Jiang, CHEN Ruinan, OU Jian
    2020, 11(1):  61-70.  doi:10.3969/j.issn.1674-8484.2020.01.006
    Abstract ( 307 )   PDF (2423KB) ( 906 )  
    An active anti-roll control method was proposed for a roll stability of a semi-trailer. A seven-degree of freedom (7-DOF) vehicle dynamic model and a three-degree of freedom reference model were established. An Unscented Kalman Filter (UKF) was adopted to estimate the lateral load transfer rate (LTR). A model predictive control (MPC) theory was used to solve optimal solution of an active anti-roll torque for each axle. Comparative analysis under typical condition was performed in a Simulink / Trucksim co-simulation environment. The results show that the state quantities of the semi-trailer are converged and the lateral load transfer rate is kept within 0.7 with both MPC controller and PID controller. The anti-rolling torque required for the MPC control is smaller and more uniform than those for the PID controller, and the state variables change more stably. Therefore, the MPC controller has a better robustness while improving a roll stability of a semi-trailer.
    Related Articles | Metrics
    Self-calibration driver fatigue detection for advanced RISC machine (ARM) platform
    FANG Bin, XU Shuo, FENG Xiaofeng
    2020, 11(1):  71-78.  doi:10.3969/j.issn.1674-8484.2020.01.007
    Abstract ( 278 )   PDF (1869KB) ( 393 )  
     A self-calibration driver fatigue detection method was proposed for Advanced RISC Machine (ARM) platform. Because of the difference of driver's height, body shape and the different installation position of camera, a driver's initial state self-calibration method was used to improve the system robustness. An improved multi task convolution neural network (MTCNN) based on deep learning was used to recognize the face and extract the feature points, to obtain the state information of head posture, eyes, mouth movement. A deep convolution neural network based on operator sequence was used to judge the driver's fatigue state level. The results show that the driver self-calibration method increases the recognition accuracy by 15%. Using MTCNN method and ARM neon acceleration technology increases the running speed of “Orange Pi Zero H5”, “Raspberry Pi”, and Android mobile phone by about 50%, respectively 200, 150, and 140 ms. Therefore, this method not only improves the robustness of the system, but also meets the real-time requirements.
    Related Articles | Metrics
    Judgement of the worst condition of the crashworthiness in a front-to-side collision of two SUVs
    WU Yubao，NIU Weizhong，LEI Yuntao
    2020, 11(1):  79-85.  doi:10.3969/j.issn.1674-8484.2020.01.008
    Abstract ( 274 )   PDF (2588KB) ( 381 )  
     A finite element front-to-side collision model of two Ford SUVs was developed to determine the worst condition of the crashworthiness of an SUV body near driving seat when the car encounters a front-toside collision from another SUV of same model at a certain speed by referring to the requirements of the 2018 China New Car Assessment Program (C-NCAP) and by using HyperWorks and LS-DYNA software, and the intrusion data as well as their distributions of the side components near driving seat of rest car were calculated in simulation when the side of the car was collided at 6 positions by striking car of 50 km/h speed in 4 directions. The results show that when a front-to-side collision of two Ford SUVs happen and the colliding speed is certain, the maximum intrusions of the side components near driving seat of struck car occur in the condition that the colliding angle is near 120° and the collided position is within the range from regulation-stated site to 200 mm ahead of the site, namely the crashworthiness of struck SUV body near driving seat appears the worst in such condition.
    Related Articles | Metrics
    Longitudinal vehicle speed control algorithm with active disturbance for intelligent driving based on disturbance observation
    HUANG Jianing, CHEN Tao，XIE Hui, ZHANG Guohui，RUAN Diwang, YAN Long
    2020, 11(1):  86-93.  doi:10.3969/j.issn.1674-8484.2020.01.009
    Abstract ( 281 )   PDF (1318KB) ( 705 )  
     A longitudinal vehicle speed control algorithm was proposed based on disturbance observation and then verified by real vehicle to reduce the time lag of longitudinal speed control for intelligent driving vehicles and to improve its active immunity. The model used a feed-forward control module with control amounts being output in advance to improve the responsiveness of the vehicle speed following; using an active disturbance rejection control (ADRC) module as feed-back link, using an extended state observer (ESO) to estimate the internal and the external disturbances online. And performed compensation at the control end to achieve accurate closed-loop control for vehicle speed. The real vehicle tests were under the conditions of curved roads and roundabouts. The results show that the algorithm controls the vehicle speed to quickly track from idle speed to the target speed within 5 s. The overall average error is 0.17 km/h. Therefore, the algorithm has a better responsiveness, a better accuracy, and a better disturbance rejection than those by a traditional PID (proportion integration differentiation) method.
    Related Articles | Metrics
    Pedestrian detection based on depthwise separable convolution and multi-level feature pyramid network
    JIANG Yicheng, LI Fan
    2020, 11(1):  94-101.  doi:10.3969/j.issn.1674-8484.2020.01.010
    Abstract ( 216 )   PDF (1597KB) ( 403 )  
     A pedestrian detection method was proposed based on convolutional neural network to improve the accuracy of pedestrian detection. The method took a YOLOv3-tiny algorithm as a base. In the backbone network part, in order to deepen the network depth, an original convolutional network structure was replaced by a depthwise separable convolution. In the detection part, an improved multi-level feature pyramid network was proposed. The network consisted of eight feature pyramids with the same structure. The feature pyramid was also composed of depthwise separable convolutions. The feature pyramid was connected in series, features of the same size obtained by different pyramids were merged. Then the fused feature pyramid was used for detection. Tests on a Caltech Pedestrian dataset were done. The results show that the miss rate of this method is 57.83%, which is 32.53% lower than that of the histogram of oriented gradient (HOG) method, and 4.67%, 3.21% lower than that of the deep learning method SA Fast-RCNN and MS-CNN, with a running speed of 34 ms/frame. Therefore, this method meets the real-time requirement.
    Related Articles | Metrics
    Large scale distributed transportation vehicle routing for smart mobility systems
    LI Wei,GUO Jifu, XIAN Kai,SHANG Pan, YANG Shaofeng
    2020, 11(1):  102-110.  doi:10.3969/j.issn.1674-8484.2020.01.011
    Abstract ( 262 )   PDF (3222KB) ( 321 )  
    A new application of distributed computing for large-scale traffic vehicular assignment and routing problems was proposed for smart mobility and proactive traffic management applications. A range of research needs and realization challenges for parallel computing implementation on multi-central processing units （CPU）through multi process interface (MPI) were discussed. A space-time event-based vehicle routing model was applied in a large scale urban network simulation setting. The primal vehicle routing model was decomposed into a set of computationally efficient sub-problems, which could significantly reduce the simulation time cost and communication overhead. The sub-problems were then assigned to independent distributed CPUs that can execute their tasks simultaneously and maintain excellent load balancing. The proposed method was applied to simulate a pilot study in Beijing metropolitan area, specifically in large scale routing and scheduling cases, the computational efficiency was examined under different number of CPU cores. The results show that the proposed parallel computing method can significantly reduce the computing time and reach a speedup of more than 200 on 512 computation nodes.  
    Related Articles | Metrics
    Deep convolution neural network for real-time object detection of intelligent-driving
    SHEN Enen, HU Yumei, CHEN Guang, LUO Pan, ZHU Hao
    2020, 11(1):  111-116.  doi:10.3969/j.issn.1674-8484.2020.01.012
    Abstract ( 263 )   PDF (1651KB) ( 217 )  
     A real-time deep neural network for object detection was proposed based on a real object detection model of YOLO and an algorithm of Faster R-CNN to improve the running speed of deep learning neural network and to meet the real-time requirements of the algorithm for intelligent-driving. The neural network retained the secondary detection mode of the R-CNN series and region proposal network (RPN), removed the priori box, and used the YOLO to predict the location directly. The position prediction error caused by ROI-pooling in the Faster R-CNN was decreased, combined with a ROI-Align method in a Mask R-CNN. The improved network was tested on KITTI dataset. The results show that the improved neural network detection takes only 38 ms at once detection, the detection average accuracy of improved networks is higher than YOLO and Faster RCNN with a good generalization ability for objects with different sizes at a faster speed with a higher detection precision.
    Related Articles | Metrics
    Automotive Energy Efficiency & Environment Protection
    Characteristics of combustion and emission of natural gas engine using gas-jet ignition
    ZHAO Ziqing, WANG Zhi, LI Fubai, WANG Jianxin
    2020, 11(1):  117-126.  doi:10.3969/j.issn.1674-8484.2020.01.013
    Abstract ( 215 )   PDF (889KB) ( 874 )  
     Jet ignition is an efficient way to achieve lean burn and improve thermal efficiency of the engine. Gas-jet ignition (GJI) combustion was studied with the designed jet igniter， and the characteristics of combustion and emission under passive jet ignition and active jet ignition strategies were investigated. The results show that active ignition extends the excess air ratio to 2 and thermal efficiency is improved 1.5% compared with passive ignition, and based on the active jet ignition, thermal efficiency is further improved to 44.5% when exhaust gas recirculation (EGR) is introduced. The NOx emission at the most efficient point decreases 66% with active jet ignition, whereas the combustion efficiency decreases 3% due to the increase of THC and CO emission. NOx emission further decreases 79% and combustion efficiency keeps stable at 96% with the introduction of EGR.
    Related Articles | Metrics
    Impacts of pulse frequency and amplitude on the unsteady characteristics of a two-stage turbine
    ZHAO Rongchao, ZHUGE Weilin，MA Xuelong, ZHANG Yangjun
    2020, 11(1):  127-134.  doi:10.3969/j.issn.1674-8484.2020.01.014
    Abstract ( 192 )   PDF (2210KB) ( 358 )  
    The pulsating exhaust in turbocompound technology significantly exerts effects on the performance of the two-stage turbine. The effects of exhaust pulse frequency and amplitude were investigated to reveal the unsteady characteristic of the two-stage turbine under pulsating flow conditions on the base of a threedimensional unsteady flow simulation model. In the model, the flows in the high-pressure turbine were modeled using full stage passage, which could consider the asymmetric flows caused by the volute. The flows in the lowpressure turbine were modelled using single passage to improve the calculation efficiency. The results show that as the frequency increases from 40 to 120 Hz, the peak instantaneous power of high-pressure stage turbine increases significantly by 15.4%, while the peak instantaneous power of low-pressure stage turbine changes within 1%. With the increase of pulse amplitude, the efficiency of low-pressure stage turbine decreases more greatly than that of high-pressure stage turbine. When the coefficient of the pulse amplitude is 1.6, the rotor efficiencies of the high- and low-pressure stage drop down 3.66% and 8.09%, respectively. The incidence angle of low-pressure turbine rotor from the middle to the tip of the blade varies significantly during the pulse period, resulting in significant flow loss at the leading edge of the blade.
    Related Articles | Metrics
    Influential factors of the real driving emission for light-duty gasoline vehicles
    YU Wenlin, YE Wenlong, LONG Huiyou, YAN Jie
    2020, 11(1):  135-142.  doi:10.3969/j.issn.1674-8484.2020.01.015
    Abstract ( 327 )   PDF (1755KB) ( 399 )  
    The real driving emission test （RDE） was carried out with four gasoline vehicles to investigated the influence of cold start, the dynamic parameters and two different calculation methods to real driving emission， and the pollutant emission was calculated with the CO2 moving average window method and the new Euro 6 method. The results show that the deviation of the influence of cold start on CO and NOx emission is within 10%, and cold start has a greater influence on urban PN emission for vehicles without gasoline particulate filter (GPF), with a maximum difference is up to 32.25%, so it is necessary to pay more attention on the calibration of new vehicles. Relative position acceleration (RPA) is positively correlated with particle number (PN) emission, but has no obviously correlation with CO and NOx. The v*apos,95 (the 95th percentile of the value of velocity times positive  acceleration) is positively correlated with CO and PN emissions, as opposed to NOx emission, and the positive correlation coefficient is greater than negative correlation coefficient. For the same valid test, the calculation result with newly Euro 6 is larger than that with the CO2 moving average window method, the newly method can more truly reflect the vehicle pollutant emission in the RDE test.
    Related Articles | Metrics