ISCAD - Design, Control and Car Integration of a 48 Volt High Performance Drive

Through its new technology the ISCAD system provides up to 300 kW with a safe-to-touch 48 Volt battery.This results in a high-performance drive that is intrinsically electrically safe.The novel machine design requires new concepts for power electronics,control systems,and integration strategies.Further degrees of freedom yield to challenges and even more important to possibilities for all kinds of applications.The innovation of the design will be explained and adjustments for power electronics and control will be worked out.The focus of this paper is the 3rd generation of Intelligent Stator Cage Drive(ISCAD)prototypes.This first integrated and compact prototype has been used to re-equip a Geely Emgrand EV.The entire system and the mentioned process are described in detail.In addition,the in-house designed 48 V battery is highlighted.Simulations and measurements give an overall demonstration of the potential and the current status of the drive.

A Driving Strategy Model of Self-Driving Cars

The self-driving cars are highly developed and about to meet the market, but the driving strategies and corresponding behaviors with others still need to be tested. In this paper, based on its characteristics and behaviors of manual-driving vehicles, we propose the driving strategies of manual-driving cars as well as self-driving cars. And we use the cellular automaton to simulate the traffic reality under different conditions, and to evaluate the efficiency of a road when self-driving cars are put into use. This research can be a reference by traffic planning and vehicles performance test, and further research can be designed in a model which can calculate the efficiency of a road when the percentage of self-driving cars are different.

AB095. Development of an assessment system of driver visual behaviours on a car simulator

Background:(I)To describe the development and components of the automobile simulator driving behavior evaluation system developed by CRIR-Institut Nazareth et Louis-Braille;(II)to present the preliminary results of the content evaluation of the driving behavior evaluation grid.Methods:The evaluation system consists of five components:(I)the VS500M Car Simulator(Virage Simulation);(II)four VS500M driving scenarios,modified to minimize the occurrence of simulator sickness and expose subjects to commonly encountered driving situations on highways and city boulevards;(III)the Tobii Pro Glasses 2 eye tracking device;(IV)a car simulator driving behavior observation grid(DBOG);(V)a software application used during the behaviour evaluation phase,where synchronized video tracking,certain data from the simulator(e.g.,speed)and the DBOG grid are presented.Initially,the expected safe driving behaviors were identified,including 235 of a visual nature,supported by literature data and consultation of the project steering committee and an expert in driving assessment.Driving behaviors were assessed in 22 subjects without visual impairment(mean age 55±20 years).Subsequently,the items were revised to determine their relevance based on their importance in terms of road safety or on the frequency with which behaviors were observed among participants.For analysis purpose,the items of the DBOG were grouped according to their content,by type of expected driving behavior(e.g.,following a stop,look to the left and right before crossing the intersection)or element to be detected(e.g.,pedestrians).Results:Some visual behaviors are difficult to observe with the eye tracker device because they are more dependent on peripheral than central vision.Others are rarely observed,possibly because they are little or not realized in daily life or the representation of reality on the simulator does not stimulate their adoption.On the other hand,the visual detection behaviors expected in a situation where safety can be compromised are mostly carried out(e.g.,detection of oncoming vehicles,side mirror verification when changing lanes).Conclusions:This first phase of the study has made possible to better target the items to be kept in the car simulator driving behavior observation grid,and those that will have to be removed as they are too difficult to observe or too rarely adopted by the participants.Content validity and inter-rater reliability will be assessed from the simplified grid.

The Effect of Self-Driving Car on Urban Traffic

Based on the idea of infinitesimal analysis, we establish the basic model of relation between speed and flow. Since putting a certain amount of self-driving car will affect the average speed of mixed traffic flow, we choose the proportion of self-driving car to be a variable, denoted by k. Based on the least square method, we find two critical values of k that are 38.63% and 68.26%. When k 38.63%, the self-driving cars have a negative influence to the traffic. When 38.63% < k < 68.26%, they have a positive influence to the traffic. When k > 68.26%, they have significant improvement to the traffic capacity of the road.

CAR模型在油田含水率预测中的应用

为了预测油田动态产量,研究油田含水率随油田的开发时间变化的模型。通过分析油田含水上升变化规律,建立带控制项的自回归(CAR)预测模型。该模型可以全程预测油田开发过程的含水率,并且计算方法简单、预测精度较高、适用于不同油区,可为制定油田开发方案提供可靠的科学依据。该模型应用于某油区实际问题的预测,可获得较为满意的预测结果。

复杂光照场景下的自动驾驶汽车行人检测

针对在低光照或光照条件变化剧烈的场景下行人检测易出现漏检、定位不准问题,提出基于图像增强(EnlightenGAN)与YOLOv5联合学习的行人检测方法,该方法中的特征提取网络可以学习到经过对抗神经网络增强模块重构的图像结构细节和颜色特征。采用平滑处理后的相对熵损失函数作为YOLOv5的置信度损失,以提升网络的泛化性能。实验结果表明,基于图像增强与YOLOv5联合学习的方法在自制的复杂光照场景行人数据集上有效降低不同光照对检测行人的敏感性,精确率达到87.58%;在Caltech数据集上改进YOLOv5损失后收敛加快且检测精确率提升1.73百分点。

基于Adams/Car汽车转向系统力矩波动优化分析

详细阐述转向系统力矩波动的基本原理,并基于Adams/Car悬架系统模型进行DOE分析,找到对转向传动比波动影响最关键的因素;然后结合实际车辆,优化得到转向传动比波动最小时的相位角最佳值.在实车上对优化方案进行验证.

Performance Development Target Setting of Passenger Car Diesel Engine

Setting engine emission targets to meet diesel car requirements is particularly important in engine performance development phase. Many researches are focused on associating vehicle performance with engine targets, but most work is done by testing, which is time and cost consuming, furthermore, the relationship of vehicle and engine will change when either engine or vehicle changes. A GT-Drive model to simulate New European Driving Cycle (NEDC) for passenger car is developed and calibrated by testing data, model precision is controlled within 5%. Time distribution of engine operating conditions when car running NEDC cycle has been analyzed, 10 critical major engine operating points are summarized according to running time proportion. Emission of NOx and smoke control regions containing these 10 points for target engine are set. Vehicle emissions are simulated and evaluated during engine development after engine performance test data are got, and engine combustion system layout and calibration are adjusted until vehicle targets are met. Vehicle is tested in chassis dynamometer finally, the testing results show a good agreement with the simulated results with an error of less than 5%, which proves that the emission value exchange of vehicle and engine is reliable. Performance target decomposition method for passenger car diesel presented can greatly shorten the development cycle and save costs.

基于Adams Car的驱动轴运动仿真

运用多体动力学软件Adams中的Car专业模块对麦弗逊悬架车型的驱动轴进行运动仿真,根据仿真结果验证在极限工况时驱动轴的角度和滑移量是否满足设计要求。

Neuropsychological Aspects of Aging and Driving for Inclusive Automotive Interior Design

As people grow older, their cognitive functions undergo changes which may result in uncomfortable driving situations and even increase the risk of accidents. This research aims to understand the neuropsychological aspects of healthy aging and their possible relationship in changes in motor performance or ability. The research methodology is descriptive and includes a collection of basic studies in the scope of neuroscience, driving tasks, and older driver behavior. The final analysis points to certain changes in the functioning of the cerebral cortex and its connections as being responsible for poor performance in some basic driving tasks, but which can be compensated for by means of adapted mechanisms in motor vehicles. This study may contribute as a methodological tool, to the automobile design process, for the selection of Advanced Driver Assistance Systems (ADAS) or other emerging technologies which can compensate cognitive changes, improve safety, comfort and inclusion of older drivers in the future of automobile interior design. We conclude that in normal aging, people may present some cognitive deficits especially linked to the frontal and parietal lobe, which interfere with the necessary driving skills albeit not with the dangerous ones and can be compensated by technological solutions. The main innovation of this article is related to the bibliographical compilation and critical analysis in terms of identifying the neuropsychological aspects of normal human aging and car driving and its relationship with possible contributions from vehicular technology, as well its consider future trends in automotive interior design that should be analysed with caution in order to users inclusion.

Study on Assessment Model of Potential Risk in Car Following Scenario

Car following is the most common driving scenario,and rear-end collisions in car following scenario are the most common accidents.TTC(time to collision)is widely employed as the risk indicator in existing rear-end collision warning systems,however,assessment model using only TTC may ignore some high-risk scenarios.Therefore,the concept of potential risk in car following scenario is proposed,and the driver’s maximum brake response time to avoid collision is defined as TM(time margin),which is the basis of the assessment model in this study.One hundred and thirty-nine(139)dangerous car following cases from China-FOT(Field Operation Test)naturalistic driving study database are extracted in reference from detection criteria on dangerous cases,of which TM at the braking moment of preceding vehicle is calculated,thus dangerous domains of two risk levels are determined by the distribution of TM.A potential risk warning system is also designed based on the assessment model,in order to further reduce rear-end collisions in combination with rear-end collision warning system.

基于CarSim的FSAE赛车传动系统主减速比选择策略

车辆动力学虚拟仿真在汽车设计中具有重要的意义,文章从传统汽车动力学理论出发,结合CarSim汽车动力学仿真软件,提出了一种大学生方程式汽车大赛(FSAE)传动系统主减速比的选择策略。综合大学生方程式大赛规则中各动态赛项目成绩的计算方法,制定了目标分数评价指标,并根据西安交通大学毅行方程式车队2019、2020赛季的实车实验,验证了该传动系统主减速比选择策略的优势。研究结果对FSAE参赛车队传动系统主减速比的设计提供了一定的指导作用。

改进行为克隆与DDPG的无人驾驶决策模型

无人驾驶技术的关键是决策层根据感知环节输入信息做出准确指令。强化学习和模仿学习比传统规则更适用于复杂场景。但以行为克隆为代表的模仿学习存在复合误差问题,使用优先经验回放算法对行为克隆进行改进,提升模型对演示数据集的拟合能力;原DDPG(deep deterministic policy gradient)算法存在探索效率低下问题,使用经验池分离以及随机网络蒸馏技术(random network distillation,RND)对DDPG算法进行改进,提升DDPG算法训练效率。使用改进后的算法进行联合训练,减少DDPG训练前期的无用探索。通过TORCS(the open racing car simulator)仿真平台验证,实验结果表明该方法在相同的训练次数内,能够探索出更稳定的道路保持、速度保持和避障能力。

面向智能驾驶测试的可变跟驰特性交通车建模方法

提出一种面向智能驾驶测试的可变跟驰特性交通车建模方法。首先,通过对自然驾驶数据聚类分析,建立高真实交互个性化的跟驰模型,并利用模型输出耦合赋予多元权值,构建可用于智能驾驶测试的可变跟驰特性交通车模型;然后,通过建立交通车轨迹评价方法验证模型输出轨迹的合理性、多样性及真实性;最后,搭建联合仿真平台进行了所构建交通车模型对自动紧急制动(autonomous emergency braking, AEB)算法的应用测试。结果表明,本文构建的交通车模型可以输出不同跟驰特性下合理、多样且真实的轨迹,当轨迹数量达到60 000条时与真实自然驾驶速度轨迹匹配的平均均方根误差为0.427 m/s,且在不同交通车轨迹特性下待测系统行为响应不尽相同,通过权值系数的变化可以揭示待测系统响应的演化规律,并可实现待测系统性能的针对性测试。

汽车横向摆动前馈调节LQR控制设计及仿真分析

对自动驾驶的横向位置参数与速度进行有效控制,需要通过设置智能控制技术来实现。为了提高汽车无人驾驶时横向摆动控制精度,设计了一种能够自主调节的前馈调节线性最优控制LQR控制器。利用非预瞄控制模式建立了对路径误差跟踪模型,建立了包含了路径跟踪偏差和车-路参数的自适应矩阵,并开展仿真测试分析。双移线测试表明,采用前馈LQR优化控制器进行处理时得到的距离偏差不超过0.3m,获得了0.1rad以内的航向偏差。LQR控制器中加入前馈控制程序后应对控制参数进行适当调节便可以达到目的。连续换道测试表明,在LQR控制器中加入前馈控制后达到更小偏差,路径跟踪精度获得了明显提升,表现出更优控制效果。该研究对提高无人驾驶汽车运行稳定性具有很好的理论支撑价值,易于推广应用。

基于微观车流的拟人化换道决策规划

为提高无人驾驶车辆换道过程中的安全舒适性,在传统换道决策基础上引入换道对局部交通流的影响,并加入换道惯性因子。针对解耦式横纵向轨迹规划不足,提出联合横纵向约束规划;基于动态规划及二次规划算法,结合上一帧纵向规划约束当前时刻横向轨迹曲率;纵向规划过程中根据当前时刻横向规划结果筛选关键障碍物,并添加曲率限速约束,结合横纵向规划最终生成换道轨迹。实车验证结果表明,所述方法生成的横向、纵向换道轨迹能有效提升换道轨迹的合理性、稳定性和舒适性。

考虑风险时空分布特征的跟驰工况行车风险预测

行车风险预测对提升智能汽车行车安全性至关重要,为此本文提出了一种跟驰工况下的行车风险预测模型(car-following risk prediction model, CRPM)。跟驰中车辆的减速能够反映驾驶人的认知风险,故根据车辆纵向加速度标注跟驰风险等级,并构建基于各向异性行车风险场的风险时空分布特征以用作CRPM的输入。CRPM通过卷积神经网络提取风险的空间分布特性,利用双向长短期记忆网络和注意力机制处理风险的时序依赖关系,最终输出跟驰风险等级。CRPM在航拍数据集AD4CHE上进行训练和测试。结果表明,CRPM具有良好的预测精度和提前预测时间,预测准确率达99.67%,在风险发生前2 s预测准确率为96.73%。

电动汽车三段式驱动桥桥壳有限元分析

为了实现电动汽车驱动桥桥壳轻量化,对动力总成外壳为铝合金材料的三段式驱动桥桥壳进行有限元分析。首先,对桥壳4种工况进行受力分析,获取桥壳外部载荷;其次,用UG软件建立桥壳几何模型,简化后导入ANSYS Workbench软件建立桥壳有限元模型;最后,对4种工况下的桥壳有限元模型进行约束和施加载荷,求解得桥壳的应力和位移分布规律。结果表明:4种工况下驱动桥桥壳满足强度、刚度要求,为三段式驱动桥桥壳结构设计提供了参考。

车辆跟驰行为的主要影响因素研究综述

概述国内外车辆跟驰行为,重点分析L0—L5级车辆跟驰行为的主要影响因素,明确各因素对车辆跟驰行为的影响机制,为解决由驾驶人、车辆特性和车辆自动化等级导致的跟驰行为问题提出合理的解决思路,最后提出综合考虑驾驶人和车辆特性因素影响下L0—L5级车辆的跟驰行为,驾驶不同等级的L0—L5级车辆驾驶人驾驶风格与自动化等级之间的耦合跟驰行为,以及L4—L5级车辆的行为预测与跟踪对车辆跟驰行为的影响等是未来需要进一步研究的主要问题。

基于深度强化学习的无人驾驶路径规划研究

针对深度确定性策略梯度(Deep Deterministic Policy Gradient,DDPG)算法在训练神经网络时出现收敛不稳定、学习效率低等问题,提出了一种基于奖励指导的深度确定性策略梯度(Reward Guidance DDPG,RG_DDPG)算法。该算法在回合内创建优秀经验集合,便于指导智能汽车充分利用过往有效信息,得到稳定的控制策略;采用基于奖励的优先经验回放机制,打破数据之间的关联性,提高数据的利用率,减少搜索过程的盲目性,提高算法的收敛稳定性。基于ROS(Robot Operating System)操作系统对算法进行了验证。在Gazebo建模软件中,设计了智能汽车模型以及障碍物环境,利用决策算法规划智能汽车的安全行驶路径。数据结果验证了RG_DDPG算法在处理路径规划任务的有效性,相比于DDPG算法,改进后智能汽车的车速能够提升60.5%,获取奖励提升一倍多,算法的收敛稳定性更好。最后通过实车实验验证了该算法的实用性。