Radar Health Monitoring for Highway Vehicle Applications

Monitoring the health of the radar sensor on a highway vehicle poses a special challenge. This is because the radar measures the distance to other independent vehicles on the highway and the motion of these other vehicles may be completely unknown to the fault detection system. Traditional observer-based approaches to fault diagnostic system design cannot be used. A number of new approaches are therefore explored in this paper in an attempt to create a reliable fault detection system for the radar. These include: (a) Use of inter-vehicle communication; (b) Use of a geographic database of pre-identified roadside radar targets; (c) Detection of abrupt failures using fuzzy logic and a knowledge of vehicle acceleration abilities; (d) Use of a redundant sensor that is inexpensive but of poor quality. The performance of each of these approaches is evaluated. Experimental results indicate that a combination of approaches (c) and (d) would provide the most reliable method for radar health monitoring. This combination would work effectively even in the absence of inter-vehicle communication in a realistic highway environment.     

Design and Evaluation of Robust Cooperative Adaptive Cruise Control Systems in Parameter Space

This paper is on the design of cooperative adaptive cruise control systems for automated driving of platoons of vehicles in the longitudinal direction. Longitudinal models of vehicles with simple dynamics, an uncertain first order time constant and vehicle to vehicle communication with a communication delay are used in the vehicle modeling. A robust parameter space approach is developed and applied to the design of the cooperative adaptive cruise control system. D-stability is chosen as the robust performance goal and the feedback PD controller is designed in controller parameter space to achieve this D-stability goal for a range of possible longitudinal dynamics time constants and different values of time gap. Preceding vehicle acceleration is sent to the ego vehicle using vehicle to vehicle communication and a feedforward controller is used in this inter-vehicle loop to improve performance. Simulation results of an eight vehicle platoon of heterogeneous vehicles are presented and evaluated to demonstrate the efficiency of the proposed design method. Also, the proposed method is compared with a benchmark controller and the feedback only controller. Time gap regulation and string stability are used to assess performance and the effect of the vehicle to vehicle communication frequency on control system performance is also investigated.     

Improving multi-channel wave-based V2X communication to support advanced driver assistance system (ADAS)

The advent of vehicle-to-everything (V2X) communication has opened the opportunity to design advanced driver assistance systems (ADAS) that collect information from sensors in neighboring vehicles and roadside infrastructure. IEEE and ETSI have designed network protocol standards for V2X communications. Despite the differences between the vehicular wireless communication architecture defined by ETSI and the IEEE protocol stack, the two standards have multichannel operations as a main commonality, with some channels dedicated to safety-critical applications and others to nonsafety services. Some recent studies have demonstrated that these standards might not provide sufficient channel utilization for reliable exchange of information in mid- and heavily congested scenarios. In this paper, we propose and evaluate the performance of a driver-assistance system to reduce the connectivity gaps between vehicles and roadside units (RSUs). This cooperative system of multi-service channel allocation will improve radio channel utilization. We also show that the required latency for this inter-vehicle communication can be obtained using the IEEE-WAVE standards and dedicated short-range communication (DSRC) proposed for vehicular environments. Simulation results show that the proposed scheme can improve the average throughput by up to 15 % in various traffic density conditions compared with the dynamic channel allocation method.     

自动驾驶障碍车辆行为预测技术研究

为了使自动驾驶车辆可以像有经验的驾驶员一样对周围车辆的行为做出准确的判断,通过车辆周围传感器来感知障碍车辆的相对位置信息,并结合自身车辆的高精定位信息,获得障碍车辆的精确位置,通过应用隐马尔可夫模型建立不同驾驶行为的预测模型,最终通过模型的预测来判断障碍车辆的可能驾驶意图,辅助自动驾驶车辆进行有效的驾驶决策,更好的规划安全高效的行驶路线。     

基于抽样车辆轨迹数据的信号控制交叉口排队长度分布估计

排队长度是评价信号控制交叉口运行状态的重要参数之一。现有大多数基于抽样车辆轨迹数据的排队长度估计方法可以实现周期级排队长度估计，但是需要信号配时、渗透率或车辆到达分布等实践中难以获取的输入信息。此外，这类方法在低渗透率条件下往往难以确保估计结果的准确性和可靠性，极大地限制了其实用性。因此，提出一种抽样车辆轨迹数据驱动的时段级信号控制交叉口排队长度分布估计方法，可不依赖任何交通流理论模型和前述输入信息实现排队估计。首先，通过理论推导可以证明时段内抽样车辆的停车位置分布和排队长度分布之间可互相转化；然后，提出一种扩展的核密度估计方法来拟合并平滑抽样车辆停车位置分布，从而有效地适应不同日期和周期的轨迹叠加所带来的波动，提高方法的适用性；最后，基于前述推导和拟合的停车位置分布实现时段排队长度分布、平均排队长度和百分位排队长度估计。分别采用仿真和实证数据对上述方法进行验证和评价。结果表明，通过叠加5 d相同时段的抽样轨迹数据，15 min的平均排队长度估计误差仅为1.59 veh，相对误差仅为9%。同时，面向不同分析时长，只要给定超过100 veh抽样车辆的观测样本，无论渗透率高低，所提出的方法在定时或自适应信号控制交叉口都可实现时段排队长度分布的准确估计，其成果可进一步用于信号控制交叉口运行可靠性评估以及多时段定时信号控制的鲁棒优化。     

Neural-network multiple models filter (NMM)-based position estimation system for autonomous vehicles

A highly accurate and reliable vehicle position estimation system is an important component of an autonomous driving system. In generally, a global positioning system (GPS) receiver is employed for the vehicle position estimation of autonomous vehicles. However, a stand-alone GPS does not always provide accurate and reliable information of the vehicle position due to frequent GPS blockages and multipath errors. In order to overcome these problems, a sensor fusion scheme that combines the data from the GPS receiver and several on-board sensors has been studied. In previous researches, a single model filter-based sensor fusion algorithm was used to integrate information from the GPS and on-board sensors. However, an estimate obtained from a single model is difficult to cover the various driving environments, including urban areas, off-road areas, and highways. Thus, a multiple models filter (MMF) has been introduced to address this limitation by adapting multiple models to a wide range of driving conditions. An adaptation of the multiple model is achieved through the use of the model probability. The MMF combines several vehicle models using the model probabilities, which indicate the suitability of the current driving condition. In this paper, we propose a vehicle position estimation algorithm for an autonomous vehicle that is based on a neural network (NN)-based MMF. The model probabilities are determined through the NN. The proposed position estimation system was evaluated through simulations and experiments. The experimental results show that the proposed position estimation algorithm is suitable for application in an autonomous driving system over a wide range of driving conditions.     

A Comparision of Spacing and Headway Control Laws for Automatically Controlled Vehicles1

SUMMARY

This paper investigates two different longitudinal control policies for automatically controlled vehicles. One is based on maintaining a constant spacing between the vehicles while the other is based upon maintaining a constant headway (or time) between successive vehicles. To avoid collisions in the platoon, controllers have to be designed to ensure string stability, i.e the spacing errors should not get amplified as they propagate upstream from vehicle to vehicle. A measure of string stability is introduced and a systematic method of designing constant spacing controllers which guarantee string stability is presented. The constant headway policy does not require inter-vehicle communication to assure string stablity. Also, since inter-vehicle communication is not required it can be used in systems with mixed automated-nonautomated vehicles, e.g for AICC (Autonomous Intelligent Cruise Control). It is shown in this paper that for all the autonomous headway control laws, the desired control torques are inversely proportional to the headway time.     

基于网联车辆轨迹数据的周期排队长度估计

近年，基于网联车辆轨迹数据的交通管控与服务研究方兴未艾。其中，信号控制交叉口排队长度估计备受关注。然而，在低渗透率条件下，单个周期内轨迹稀少且提供的交通信息十分有限。现有研究仅以当前周期内网联车辆轨迹数据为输入，难以获得准确且可靠的周期级排队长度估计结果。因此，融合利用历史网联车辆轨迹数据提供的车辆到达和停车位置信息以及当前周期内实时观测的网联车辆排队信息，提出一种基于最大后验概率的周期最大排队长度估计方法。首先，依据历史轨迹数据的停车位置信息，估计排队长度的先验分布；其次，依据历史轨迹数据的车辆到达信息，估计周期内车辆的历史到达分布，并结合周期内最后1辆排队网联车辆的到达时刻与停车位置，构建排队长度似然函数；最后，基于贝叶斯理论，结合前述先验分布与似然函数，推导周期排队长度的后验分布，并采用最大后验概率方法实现周期最大排队长度的估计。仿真结果表明：所提方法在不同饱和度和渗透率条件下，均优于现有的方法；即使在车辆轨迹数不超过1 veh·周期^-1的低渗透率条件下，所提方法的平均绝对估计误差也不超过2 veh·周期^-1。实证结果表明：在渗透率仅为8.96%的条件下，所提方法的平均绝对误差为2.12 veh·周期^-1，平均相对估计误差为12.4%，同样优于现有同类方法。     

Modeling of Double Lane Change Maneuver of Vehicles

Lane change maneuver is one of most riskiest driving tasks. In order to increase the safety level of the vehicles during this maneuver, design of lane change assist systems which are based on dynamics behavior of driver-vehicle unit is necessary. Therefore, modeling of the maneuver is the first step to design the driver assistance system. In this paper, a novel method for modeling of lateral motion of vehicles in the standard double-lane-change (DLC) maneuver is proposed. A neuro-fuzzy model is suggested consisting of both the vehicle orientation and its lateral position. The inputs of the model are the current orientation, lateral position and steering wheel angle, while the predicted lateral position and orientation of the vehicle are the outputs. The efficiency of the proposed method is verified using both simulation results and experimental tests. The simulation and experimental maneuvers are performed in different velocities. It is shown that the proposed method can effectively reduce the undesirable effects of environmental disturbances and is significantly more accurate in comparisons with the results in the recent available papers. This method can be used to personalize the advanced driver assistance systems.     

Towards lane-level traffic monitoring in urban environment using precise probe vehicle data derived from three-dimensional map aided differential GNSS

Today's urban road transport systems experience increasing congestion that threatens the environment and transport efficiency. Global Navigation Satellite System (GNSS)-based vehicle probe technology has been proposed as an effective means for monitoring the traffic situation and can be used for future city development. More specifically, lane-level traffic analysis is expected to provide an effective solution for traffic control. However, GNSS positioning technologies suffer from multipath and Non-Line-Of-Sight (NLOS) propagations in urban environments. The multipath and NLOS propagations severely degrade the accuracy of probe vehicle data. Recently, a three-dimensional (3D) city map became available on the market. We propose to use the 3D building map and differential correction information to simulate the reflecting path of satellite signal transmission and improve the results of the commercial GNSS single-frequency receiver, technically named 3D map-aided Differential GNSS (3D-DGNSS). In this paper, the innovative 3D-DGNSS is employed for the acquisition of precise probe vehicle data. In addition, this paper also utilizes accelerometer-based lane change detection to improve the positioning accuracy of probe vehicle data. By benefitting from the proposed method, the lane-level position, vehicle speed, and stop state of vehicles were estimated. Finally, a series of experiments and evaluations were conducted on probe data collected in one of the most challenging urban cities, Tokyo. The experimental results show that the proposed method has a correct lane localization rate of 87% and achieves sub-meter accuracy with respect to the position and speed error means. The accurate positioning data provided by the 3D-DGNSS result in a correct detection rate of the stop state of vehicles of 92%.     

Autonomous Lateral Control of Vehicles in an Automated Highway System

Lateral control of vehicles in IVHS requires the installation of on-board sensors as well as the installation of roadway hardware such as cables, magnets, etc. Existing control approaches in PATH require road curvature and vehicle lateral position (with respect to the center of the lane) information. Hence these approaches rely on roadway sensors to obtain relative lateral position. These methods will necessitate infrastructural changes to the highway.

This paper introduces the concept of autonomous lateral control or auto-tracking. The method allows us to use only line-of-sight sensor information to effect vehicle control. We present a detailed vehicle model. Controllers have been proposed to demonstrate the effectiveness of the proposed auto-tracking scheme. We also examine the possibilities of using this method for lane change purposes in an automated highway system.     

基于人-车交互的行人轨迹预测

针对行人轨迹预测具有复杂、拥挤的场景和社会交互问题,基于长短时记忆网络（Long Short-term Memory Network, LSTM）对行人与车辆、行人与其他行人的交互进行建模,提出一种基于人-车交互的行人轨迹预测模型（VP-LSTM）。该模型同时考虑了行人与行人的交互、行人与车辆的交互,更适用于复杂的交通场景。所构建的VP-LSTM包括3个输入,以行人的方向和速度作为历史轨迹序列输入,行人与行人的相对位置作为人-人交互信息输入,行人与车辆的相对位置作为人-车交互信息输入。该方法首先设计扇形人-人交互邻域和圆形人-车交互邻域来准确捕捉对被预测行人有相互作用的行人和车辆;其次建立3种不同的LSTM编码层来编码历史行人轨迹序列、人-人、人-车社交信息;然后定义人-人、人-车交互的防碰撞函数和方向注意力函数作为人-车、人-人社交信息的权重,进一步提高社会信息的精度;再将人-人、人-车交互信息输入到注意力模块中筛选出对行人影响大的社会信息;最后将筛选后的社会信息与行人历史轨迹序列一起输入到LSTM神经网络中进行行人轨迹预测,并在构建的DUT人-车交互数据集上验证提出的网络。研究结果表明：提出的方法能够准确地预测出交通场景中,人-车交互行人未来一段时间内的运动轨迹,有效提高了预测精度,提高了智能驾驶决策的准确性。     

Vehicle distance estimation using a mono-camera for FCW/AEB systems

For robust vision-based forward collision warning (FCW) and autonomous emergency braking (AEB) systems, not only reliable detection performance including high detection rate and low false positives but also accurate measurement output of a target vehicle is required. Especially, in order to reduce false alarm or activation of FCW/AEB systems, the systems require the precise measurement output of a target object, such as position, velocity, acceleration, and time-to-collision (TTC). In this study, we developed a measurement estimation algorithm of a target vehicle using a monocular camera. This method estimates two cases of vehicle widths for a target vehicle by using the detected lane information and a pin-hole camera model. After that, the position, velocity, acceleration, and TTC of a target vehicle are estimated by using a Kalman filter for the each estimated vehicle width. To improve robustness, the both estimation results using the detected lane information and the pinhole camera model are fused. This estimation algorithm was evaluated and compared with the state-of-the-art technology. As a result, the proposed measurement output estimation method can improve the performance of the FCW/AEB systems.     

交通视频辅助的桥梁动态称重方法研究

为了进一步提升既有的桥梁动态称重技术，提出一种交通视频辅助的新型桥梁动态称重方法。首先介绍基于深度神经网络的计算机视觉目标检测技术和一种计算机视觉坐标转换方法，实现从交通监控视频中实时地探测与定位桥上行驶的车辆和车轴。然后引入桥梁应变分解方法和应变影响面识别方法，建立车重、车辆位置与桥梁应变之间的映射关系，从而建立一种综合利用时间和空间冗余信息对车辆进行称重的方法。该方法构建超定的影响面加载方程组，使用最小二乘法求解该方程组以得到桥上行驶车辆的轴重和总重。最后总结出一套交通视频辅助的桥梁动态称重方法框架。为验证以上方法，在某连续大箱梁桥的缩尺模型以及实桥上进行试验。试验包含单车、双车、跟车、并行、直行、变道、匀速、变速等复杂交通工况。模型试验结果表明：该方法的车辆总重识别误差均值为-2.02%，标准差为4.77%；车辆轴重的识别误差均值为4.77%，标准差为17.50%。实桥试验结果表明：该方法的车辆总重识别误差均值为0.21%，标准差为1.53%；车辆轴重的识别误差均值为-3.59%，标准差为42.67%。除此以外，所提出的方法还可用于识别桥上车辆的数量、类型、轴数、实时位置、运动轨迹、行驶速度等多粒度交通信息。     

A proposed vision and vehicle-to-infrastructure communication-based vehicle positioning approach

It is essential to obtain accurate location of vehicles for new applications of Intelligent Transportation Systems. To remedy the defects of present Global Positioning System and vehicle-to-infrastructure (V-I) positioning technology, a new positioning approach based on vision and V-I communication is proposed. This approach aims at lane-level positioning with lower cost than conventional ones. In this approach, the position of the vehicle is represented by its lateral position (the lane number) and longitudinal position (the distance from entrance of the road) in a course coordinate system along the road; the specific lane the vehicle is occupying (the lane number) can be judged using the information of lane lines detected by vision systems; then the distance to the vehicle is obtained by a Road Side Unit (RSU) during the V-I communication; and the longitudinal position is calculated. The error of the approach on typical operating conditions is analyzed, indicating that the new approach can achieve the accuracy of less than 0.31 m for straight road and 0.58 m for typical arc road with ultra-wideband communication and ranging technologies and rational arrangement of RSUs. The feasibility of this approach is presented.     

基于注意力Seq2Seq网络的高速公路交织区车辆变道轨迹预测

针对交通状态复杂的高速公路交织区域,经验丰富的驾驶人能够通过正确地推断周围车辆的未来运动进行及时的车道变换,这对于实现安全高效的自动驾驶至关重要,然而目前的自动驾驶车辆往往缺乏这种预测能力。为此,基于深度学习理论,提出了一种结合注意力机制和编-解码器结构的交织区车辆强制性变道轨迹预测方法,利用Next Generation Simulation（NGSIM）数据集提取车辆变道过程中的关键特征,并引入碰撞时间（Time to Collision,TTC）和避免碰撞减速度（Deceleration Rate to Avoid a Crash,DRAC）2种风险指标,将变道车辆及其周围车辆视为一个整体状态单元,同时补全状态单元内部不同车辆在横向和纵向上的时空状态特征,从而更有效地刻画车辆间的动态交互行为;然后将不同观测车辆的连续窗口序列输入基于长短期记忆网络（Long Short-term Memory,LSTM）的编-解码器,预测交织区车辆变道的未来运动轨迹,通过添加软注意力模块,使模型能够集中聚焦于影响车辆在不同时刻下位置变化的关键信息,再现了真实交通场景下车辆的变道行为。试验验证表明：基于注意力机制的编-解码器模型与当前流行的卷积长短期记忆网络、极限梯度提升树等模型相比具有更高的轨迹预测精度,在长时域的变道轨迹拟合上有显著的优越性,为辅助和自动驾驶领域的发展提供了新思路。     

Autonomous Lateral Control of Vehicles in an Automated Highway System

SUMMARY

Lateral control of vehicles in IVHS requires the installation of on-board sensors as well as the installation of roadway hardware such as cables, magnets, etc. Existing control approaches in PATH require road curvature and vehicle lateral position (with respect to the center of the lane) information. Hence these approaches rely on roadway sensors to obtain relative lateral position. These methods will necessitate infrastructural changes to the highway.

This paper introduces the concept of autonomous lateral control or auto-tracking. The method allows us to use only line-of-sight sensor information to effect vehicle control. We present a detailed vehicle model. Controllers have been proposed to demonstrate the effectiveness of the proposed auto-tracking scheme. We also examine the possibilities of using this method for lane change purposes in an automated highway system.     

Model-independent self-tuning fault-tolerant control method for 4WID EV

To solve the problem of the existing fault-tolerant control system of four-wheel independent drive (4WID) electric vehicles (EV), which relies on fault diagnosis information and has limited response to failure modes, a modelindependent self-tuning fault-tolerant control method is proposed. The method applies model-independent adaptive control theory for the self-tuning active fault-tolerant control of a vehicle system. With the nonlinear properties of the adaptive control, the complex and nonlinear issues of a vehicle system model can be solved. Besides, using the online parameter identification properties, the requirement of accurate diagnosis information is relaxed. No detailed model is required for the controller, thereby simplifying the development of the controller. The system robustness is improved by the error based method, and the error convergence and input-output bounds are proved via stability analysis. The simulation and experimental results demonstrate that the proposed fault-tolerant control method can improve the vehicle safety and enhance the longitudinal and lateral tracking ability under different failure conditions.     

用于智能驾驶系统评价的乘员损伤模型

只有较少的交通事故数据资源被用于建立基于碰撞速度信息的乘员损伤模型，致使所得到的模型精度差。为此，提出了基于车辆变形深度的乘员损伤模型。对美国不同制造年代和车辆级别的事故数据进行聚类分析，论证出车辆变形深度与乘员损伤风险具有相关性。以车辆变形深度为自变量，通过回归分析得到乘员损伤模型。不同种类车辆的乘员损伤模型拟合精度R²约为0.9，证明了该模型的正确性。为进一步验证，以此模型为基础，评价智能驾驶系统的有效性。以自动紧急制动系统为例，对比基于变形深度和速度变化量信息2种方法的有效性计算结果。结果表明：2组结果的平均误差不超过1%，验证了基于变形深度的乘员损伤模型的准确性。该模型仅需要事故数据库中准确的变形深度信息，能够获得更多的事故数据支持，从而可以更好地适应于不同类别智能驾驶系统的评价需求。     

基于模式识别与ST-MRF相结合的车辆检测方法

车辆检测技术的主要难点是在于解决车辆之间的遮挡,以及由于光照变化引起的车辆与其阴影之间的遮挡问题,这些问题将直接影响检测的精度.针对这个问题,在原ST-MRF方法上研究了基于模式识别与ST-MRF相结合的车辆检测方法.模式识别技术分割相互遮挡的2辆车之间的边界,并识别相互遮挡车辆的边缘间隙以及边界信息,模式识别结果反馈...