Integration of GPS and monocular vision for land vehicle navigation in urban area

For land vehicle navigation in urban area, Global Positioning System (GPS) receivers often suffer from the lack of positioning accuracy, availability, and continuity due to insufficient number of visible satellites and multipath errors. To mitigate this problem, this paper proposes an efficient hybrid positioning method combining a single frequency GPS receiver and a monocular vision sensor. The proposed method is advantageous in that it requires only low-cost hardware and no external map aiding. Compared with existing vision-based methods, the proposed method directly measures absolute heading angle based on the images of straight road segments. For the reason, the proposed method is resilient to multipath errors. The performance of the proposed method is evaluated by the experiments with field-collected real measurements; one with good satellite visibility and the others with poor satellite visibility. Comparison with existing positioning methods demonstrates the feasibility of the proposed method in urban area.     

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.     

基于最小临近点迹和航迹关联的多源目标融合方法

针对智能车辆多传感器的目标融合问题,提出了一种改进的基于欧氏距离与余弦相似度的点迹和航迹数据关联的车用多传感器目标跟踪融合算法。该方法需获取由毫米波雷达系统和Mobileye视觉系统检测到的目标物数据列表,并对两个传感系统检测到的目标物数据进行匹配关联;然后对目标物进行匹配跟踪,更新目标物的生命周期状态;最后对上述两个传感系统输出的目标物的数据进行融合。该算法能够融合视觉系统和雷达系统两个传感系统的优点,以达到精确感知环境信息的目的,从而解决单一传感器难以满足感知系统精度及可靠性需求的问题。     

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.     

神经元实时辨识车辆导航系统中的GPS多径误差

在城市车辆导航过程中,多径误差是近距离差分GPS等高精度定位的主要误差源。文章首次提出神经元实时辨识GPS多径误差方法,它能实时辨识当前时刻GPS接收机输出的GPS信号是否含有多径误差,从而解决GPS多径误差对城市车辆导航定位精度的影响。把该方法用到实际工程中,其结果显示能够有效的消除GPS多径误差对城市车辆导航系统定位精度的影响。     

双目视觉技术在目标测量中的应用

给出了一种用于目标空间三维距离、方位信息探测的立体视觉系统实现方法。该系统根据双目视觉原理,利用预制三维标定物对图像获取系统的内、外参数进行标定,求出投影变换矩阵,根据图像识别结果运用灰度模板、连续性假设和对极几何约束进行识别目标的特征匹配,进而获得目标的三维信息。试验结果表明,该方法可以有效地获取目标周边环境的信息。     

基于模糊逻辑的导航定位数据校正算法

为了提高GPS/DR组合定位系统的定位精度,通常采用地图匹配算法来修正定位误差.文中采用了一种基于模糊逻辑的导航定位数据校正算法,对经联合卡尔曼滤波输出的GPS/DR的定位数据进行校正.通过Matlab仿真实验,结果表明,该算法能有效地减小误差,提高组合定位系统的定位精度,改善其对航线跟踪的质量.     

A Novel Optimal Routing for Navigation Systems/Services Based on Global Navigation Satellite System Quality of Service

The positioning quality of global navigation satellite system (GNSS), or GNSS quality of service (QoS), is a major factor impacting real-time navigation performance. Commonly requested routes (i.e., shortest or fastest) may include areas with poor GNSS QoS, which can subsequently degrade navigation performance. To provide alternative routes with high or acceptable GNSS QoS along a route, a novel optimal routing for navigation systems/services based on GNSS QoS by utilizing integrated GNSS (iGNSS) QoS prediction is presented in this article. New routing criteria based on GNSS QoS are maximum availability, maximum accuracy, maximum continuity, and maximum reliability. Two experiments were conducted to compare GNSS QoS-based routes against shortest routes. In one experiment, routes were simulated, and in another, generated routes based on GNSS QoS were evaluated against GPS-based trajectories as ground truths. The results show that GNSS QoS-based routes provide routes with higher QoS, more than 50%, and longer, about 50%, than shortest routes.     

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%.     

A Trajectory-Based Map-Matching System for the Driving Road Identification in Vehicle Navigation Systems

Driving road identification is the key issue of a vehicle navigation system that supports various services of intelligent transportation systems. The method for driving road identification is also known as map matching (MM). In spite of the development of MM algorithms, limitations still exist in obtaining the positioning data and preparing candidate roads (CRs) that may result in mismatches in some special difficult road configurations such as flyovers and parallel roads. To overcome the limitations, an integrated trajectory-based MM (tbMM) system is proposed based on the trajectory similarity evaluation method. The system can fuse the information from global positioning systems (GPS) and inertial sensors to generate the vehicle trajectory that represents the vehicle continuous movement in three dimensions. The elevation data of vehicle and roads are involved to enhance the trajectory-based matching process. Also the method employs an optimized mechanism for generating and maintaining CRs. Using the mechanism, separated road segments in a digital map are reorganized in the form of possible driving roads and the topology among them is guaranteed. Moreover, the CRs are obtained considering all the possibilities in determining the driving road so that the valuable historical information can be effectively reserved to provide more reliable matches in ambiguous situations. The tbMM system was evaluated using a number of real-world vehicle-level test datasets in urban areas in Beijing. Also a comparison test was performed to evaluate the driving road identification accuracy against existing MM algorithms. The results show that the tbMM system can provide reliable matches with about 99% accuracy in all the difficult scenarios and outperforms the existing algorithms.     

利用常规仪器完成海上打桩定位的探索与实践

结合杭州湾跨海大桥北航道桥栈桥工程实践,在打桩船未安装商用海上GPS打桩定位系统的条件下,本文提出了一种新的利用常规仪器完成海上桩基钢管桩的插打测量定位方法,并实现了快速准确地对钢管桩的插打测量定位。通过对沉桩进行的偏位检查表明,该方法完全能够满足钢管桩测量定位的精度和配合施工作业的进度。     

考虑动力学模型系统误差补偿的智能车GNSS/IMU组合定位算法

为了解决智能车动态组合定位过程中，因动力学模型与实际模型之间存在偏差导致滤波精度下降的问题，针对智能车全球导航卫星系统(GNSS)/惯性测量单元(IMU)组合定位系统，结合非线性预测滤波(NPF)和自适应滤波的优点，提出了一种考虑动力学模型系统误差实时估计和补偿的自适应非线性预测滤波(ANPF)算法。首先，根据NPF算法原理，通过最小化预测观测残差与系统误差的加权平方和，估计动力学模型系统误差；其次，结合自适应滤波原理，利用状态预测残差向量构造自适应因子，设计了一种自适应扩展卡尔曼滤波(AEKF)算法，用于估计系统状态向量，并通过自适应因子抑制动力学模型系统误差和线性化误差对系统状态估计精度的影响，克服NPF对系统状态估计精度有限的缺陷；再次，对动力学模型系统误差的估计误差和由动力学模型系统误差引起的系统噪声的等效协方差阵进行了分析和推导，以补偿动力学模型系统误差对系统状态估计的影响；最后，通过车载GNSS/IMU组合定位系统试验，从算法精度、鲁棒性和实时性方面对提出的算法和其他滤波算法的性能进行了验证和对比分析。研究结果表明:提出的自适应算法继承了NPF算法简易性和高实时性的优点，同时克服了NPF算法估计精度有限的缺陷，具有较好的滤波解算精度，水平定位精度小于1.0 m，算法单次平均执行时间约为0.013 9 ms，在精度和实时性的平衡方面显著优于其他滤波方法。     

复杂多点进出型地下道路车辆定位与导航技术应用测试研究

针对复杂地下道路环境内缺乏有效交通引导、定位导航的问题,分析探讨了现有室内定位技术的不足,研究了适用于普通手机终端的地下车行定位与导航系统。通过开发适用于复杂地下环境的射频矩阵基站、定位与导航引擎,结合普通智能手机终端,实现了具有低延时、高精度和高可靠性的地下车行导航解决方案。同时,于苏州中心开展了地下定位导航示范应用测试,结果表明：该系统能满足车辆在地下环境的定位与导航需求,实现与手机导航app对接,为驾驶人提供地上地下无缝衔接、稳定高效的一体化导航服务。     

Development of a telematics evaluation environment based on a driving simulator and its application

For developing telematics devices, traditional development methods include the unit function test, compatibility test and T-Car, which have some limitations. Telematics devices have various functions that require accounting for the interactions among three major elements of automotive electronics: the vehicle, the device unit and driver. The KAAS (KATECH Advanced Automotive Simulator) system is a virtual-reality-based test environment designed to test and analyze the three elements in one place. One of the difficult functions when constructing such VR (Virtual Reality)-based telematics test environment is to develop a test method for the LBS (Location-Based Service) function such as a car navigation demanding the GPS (Global Positioning System) satellite signals because KAAS is in a fixed laboratory. To overcome these problems, a real-time GPS simulation system, which can be integrated with KAAS, is needed because the location of the vehicle in virtual space is determined purely by the driver’s personal intention while driving virtually. This paper presents new concepts needed to construct a VR-based telematics test environment to generate a GPS RF signal, which reflects the continuously changing vehicle location during virtual driving in real-time. To construct this system, the coordinate transform must be conducted from a rectangular coordinate system that is compatible with a virtual 3D DB that is used to construct a 3D image for KAAS using a WGS84 and a longitude-latitude coordinate system compatible with a GPS simulator. Moreover, the real-time HILS (Hardware In Loop Simulation) systems and the CDMA (Code Division Multiple Access) simulation system are developed to evaluate telematics devices. Finally, we show its applications and results.     

A weight-based map-matching algorithm for vehicle navigation in complex urban networks

A map-matching algorithm is an integral part of every navigation system and reconciles raw and inaccurate positional data (usually from a global positioning system [GPS]) with digital road network data. Since both performance (speed) and accuracy are equally important in real-time map-matching, an accurate and efficient map-matching algorithm is presented in this article. The proposed algorithm has three steps: initialization, same-segment, and next-segment. Distance between the GPS point and road segments, difference between the heading of the GPS point and direction of road segments, and difference between the direction of consecutive GPS points and direction of road segments are used to identify the best segment among candidates near intersections. In contrast to constant weights applied in existing algorithms, the weight of each criterion in this algorithm is dynamic. The weights of criteria are calculated for each GPS point based on its: (a) positional accuracy, (b) speed, and (c) traveled distance from previous GPS point. The algorithm considers a confidence level on the assigned segment to each GPS point, which is calculated based on the density and complexity of roads around the GPS point. The evaluation results indicate 95.34% correct segment identification and 92.19% correct segment assignment. The most important feature of our algorithm is that the high correct segment identification percentage achieved in urban areas is through a simple and efficient weight-based method that does not depend on any additional data or positioning sensors other than digital road network and GPS.     

专用短程通信辅助的车辆卫星定位故障检测方法

为了确保卫星定位性能满足特定协作式智能交通应用需求,提高车辆定位系统的故障容错能力,针对车辆卫星定位的自主故障检测与性能优化问题,提出基于专用短程通信辅助的卫星定位故障检测方法,充分利用专用短程通信设备的测距率观测信息,实现故障检测对不同类型卫星可视条件的有效适应。基于专用短程通信多普勒观测特性,构建基于载波频偏的车间测距率观测模型;设计卫星定位与专用短程通信组合观测与解算框架;基于容积卡尔曼滤波提出适于非线性观测特征的故障检测、识别与排除算法,并叠加量测噪声方差矩阵动态调整策略,对故障检测性能进行优化;基于实测试验检验车间测距率的观测性能,并运用实车轨迹对多车协同运行及定位采集过程进行仿真,检验所提出方法的故障检测性能。研究结果表明：提出的方法有效解决了常规接收机自主完好性监测算法受卫星可视条件限制的问题,所引入的量测噪声方差矩阵调整策略提升了故障检测及故障排除性能的稳定性,在给定仿真场景中,常规卫星观测条件下阶跃故障、斜坡故障排除率相对常规方法最高可分别提升52%、18%,受限观测条件下不同水平2类故障的排除率最高分别可达100%、89%,边界观测条件下不同水平2类故障的检测率最高分别可达100%、96%。研究结果对于充分发挥车-车协同模式的核心优势、保障车辆定位性能具有重要价值。     

基于贝叶斯概率估计的智能电动车动态目标避障算法

为了实现智能电动车在中汽中心智能网联示范基地内的动态避障,首先将直角坐标系与曲线坐标系进行转换,构建以参考路径的弧长s为横坐标,横向偏移距离q为纵坐标的曲线坐标系;其次,在曲线坐标系中利用三次多项式生成满足初始位姿与子目标点位姿的候选路径,同时对标准化常量的似然函数进行定义,在此基础上利用贝叶斯定理对每条候选路径的危险等级进行概率估计;在动态避障过程中,借鉴速度障碍法对碰撞威胁进行实时检测,并建立最短避障时间和安全距离的数学模型来实现高效的动态避障,最后对行人占用车道行走与横穿马路2种典型场景进行动态避障试验。研究结果表明：在曲线坐标系中,通过横向偏移距离能够便捷地建立起一系列候选路径,克服在直角坐标系中寻找移动子目标点这个难题;在寻找安全路径方面,由于智能电动车工作环境的不确定性,利用贝叶斯定理对候选路径危险等级进行概率计算的方法可靠性更高,速度障碍法与避障数学模型的结合满足碰撞危险检测的实时性和动态避障的高效性要求。试验结果表明：采用曲线坐标系中的动态避障算法对行人占用车道和横穿马路2种场景进行了有效的避障,在路径选择上符合实际驾驶习惯,达到了智能网联示范基地动态避障的要求。     

基于双目视觉的前车距离识别方法

本文提出一种基于双目视觉系统,对道路上各对象进行特征识别的方法。首先利用Canny算子和霍夫变换等图像处理方法检测出道路区域,以提高其后的图像处理效率;然后在道路区域内部通过阴影识别方法对车辆进行识别;最后利用双目视觉几何关系,对对象物体距离进行比较。实验结果表明,通过本文所述方法可准确识别出图像中各对象及其距离关系。     

弱GNSS信号下基于EMD和LSTM的车辆位置预测方法研究

针对弱GNSS环境下组合导航（INS/GNSS）系统存在的定位偏差问题，提出一种基于经验模态分解和长短期记忆网络的车辆位置预测算法。首先，针对训练数据中噪声较大的惯导数据，提出一种融合经验模态分解与离散小波变换的降噪算法。该算法基于噪声能量估计和各阶本征模态函数的功率谱密度函数，提出一种确定混合模态函数阶数上下界的方法，并采用离散小波变换硬阈值法对混合模态函数进行滤波处理，最终利用经过处理的各阶模态函数重构原始数据以达到降噪目的。训练数据经过预处理后，采用改进的堆叠式长短期记忆网络离线训练位置预测模型，利用该训练模型可在线实时进行位置预测。针对车辆定位序贯数据预测，提出一种局部数据降噪方法，该方法利用一定长度时间窗口的历史数据，通过线性最小二乘给出当下时刻数据的预估值，并与实际量测值进行滑动平均滤波，优化位置预测的结果。在封闭场地模拟隧道环境下，对长短期记忆网络输入端进行局部数据降噪与不进行降噪处理比较，经度和纬度的归一化均方误差分别下降了13.34%和9.38%，经度和纬度的归一化平均绝对误差分别下降了8.64%和5.41%；在复杂城市交通环境下，检验提出的方法，经度和纬度的归一化均方误差分别下降了6.51%和5.66%，经度和纬度的归一化平均绝对误差分别下降了5.70%和8.23%。试验结果表明，在弱GNSS信号环境下，提出的车辆位置预测方法有效提高了车辆定位精度和稳定性。     

改进粒子滤波算法在组合导航中的应用

为了提高组合导航定位系统的定位精度和可靠性,分别对扩展卡尔曼滤波(EKF)、粒子滤波(PF)和U卡尔曼滤波(UKF)3种算法进行了分析。通过分析3种算法各自的特点,将PF算法和UKF算法的优点相结合,提出了一种新的粒子滤波算法——U粒子滤波(UPF)算法,并将其应用于GPS/DR组合导航系统中。通过对UPF算法与PF算法在GPS/DR组合导航系统中的仿真研究比较,进一步证实了UPF算法的可行性及计算的精确性。