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。     

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.     

Precise and reliable positioning based on the integration of navigation satellite system and vision system

In this paper, we propose a precise and reliable positioning method for solving common problems, such as a navigation satellite’s signal occlusion in an urban canyon and the positioning error due to a limited number of visible navigation satellites. This is an integrated system of the navigation satellites system and a vision system. In general, the navigation satellite positioning system has a fatal weakness in that it can not calculate a position coordinate when its signal is occluded by some obstacle. For this reason, positioning by using the navigation satellites system can not be used for a variety of applications. Therefore, we propose as a method to integrate both the navigation satellites system and the vision system. Some target objects that have accurate position coordinates, for example, in an outdoor shaded area like an urban canyon, are installed into the vision system. When the vision system recognizes a target object it loads the accurate coordinate of that target object. Then, it measures the distance by using the disparity from the camera sensor to the target object. These distance and object coordinate data are used for positioning with the navigation satellites system’s data. This integrated system can be used for the positioning solution where the user is in unfavorable conditions. This paper shows that the algorithm of integrated system and the numerical test performed. The results indicate that the reliable and stable positioning can be obtained by introducing the vision system to the satellite navigation system.     

Effect of number of wheels on high speed UGV traversability: Online terrain assessment approach

This paper addresses the problem of determining the maximum allowable speed (V _max ) of Unmanned Ground Vehicles (UGVs) traversing off-road heterogeneous rough/complex terrain. The maximum speed is calculated based on online terrain assessment and the vehicle’s characteristics; this speed achieves the high speed navigation without exceeding a predefined allowable range of transmitted force and moments (i.e., moments in the roll and pitch directions) to the vehicle’s frame. The proposed system enables UGV’s to change their speed autonomously and transit between terrains with different characteristics (e.g., pavement, rocky) safely. This paper proposes a computationally inexpensive approach to process acquired data and assess the terrain roughness. The proposed Roughness Index (RI) is used to represent the terrain roughness on a scale from 0.0 (highly rough to be traversed by a particular vehicle) to 1.0 (perfectly flat/smooth surface). A general vehicle model (workable for any vehicle size and wheel configuration) is presented in this paper. A closed form expression of the maximum allowable vehicle speed is developed. Simulation results obtained on real vehicles (e.g., military tank LEOPARD I) show the effect of increasing the number of wheels to improve the capability of a ground vehicle to traverse rough terrains at high speeds. In addition, the proposed terrain assessment system is a key module in UGVs navigation systems enabling them to be truly autonomous by providing the navigation system with the necessary information for path and speed planning.     

Lateral controller design for an unmanned vehicle via Kalman filtering

This paper proposes a lateral control system for an unmanned vehicle that is designed to improve the responsiveness of the system with the use of a PD control. The vehicle heading error can be stabilized, and the transient response characteristics can be improved using the proposed controller. A mathematical model of the vehicle dynamics using two degrees of freedom was developed for the controller design. The waypoint tracking method for autonomous navigation was tested with incorporation of the Point-to-Point algorithm with position and heading measurements received from GPS receivers via Kalman filtering. The performance of the designed controller was verified through experiments with a real vehicle.     

基于行动感测的实时坑洞侦测系统与方法

提出一个以行动感测和云端运算为基础的实时坑洞侦测方法,通过分析内嵌于手机内的加速度传感器数据,并利用欧拉角进行手机加速度正规化的动作,改善搜集加速度数据时需要固定手机姿态角的限制.结合坑洞侦测算法进行坑洞判定,以及使用空间内插法改善全球定位系统(GPS)定位误差,将坑洞信息回传至云端服务器.再将此坑洞信息实时传输至每个使用者的行动式手持装置中,并搭配扩增实境(AR)导航功能,让用户可以避开坑洞路段.实验表明,所提出的基于行动感测的实时坑洞侦测方法将可以准确判断出坑洞.      

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.     

Accurate INS position solution during GPS outages or degraded GPS

Temporary degraded GPS (DGPS) position loss, in circumstances such as an overhead bridge, can be alleviated by an inertial navigation system (INS) that uses onboard sensors, such as yaw and speed sensors, to determine vehicle position. This paper introduces a post-processing DGPS/INS integration approach based on using the INS solution during DGPS outages or periods of low accuracy DGPS position solutions. In this approach, the INS solution initialization is performed using the DGPS solution before DGPS position solution loss, and measurements from the Inertial Measurement Unit (IMU). The final post-processed INS solution is a weighted average of the INS forward and backward solutions. This work constitutes two parts: the INS initialization methods for different degrees of freedom vehicle positioning models, and the developed weighting model necessary to combine the forward and the backward solutions. The former part is essential in obtaining acceptable INS initial states for both the stand-alone INS or any post-processing or real time INS/GPS integrated system. The latter part is based on the use of the complementary error behaviours of the backward and the forward solutions, and can be used as a survey method with acceptable position solutions accuracies. Applying the forward/backward INS combined solution method on real data shows that the resultant INS solution accuracy is 35 cm or less over a 1000 m road segment. This method is used to survey freeways interchange road segments where 50% of the surveyed distance has no DGPS solution or has a degraded DGPS solution. The average achieved accuracy over the whole freeways interchange is around 40 cm over a 23 km distance.     

改进联合滤波在GPS/DR组合定位中的应用

根据低成本车辆GPS/DR组合定位系统传感器精度低和计算能力弱的特点,提出一种改进联合卡尔曼滤波(FKF)算法,并简化主滤波器信息融合算法,稍微降低融合精度,提高计算效率。试验结果表明,提出的改进联合滤波算法具有融合精度高、容错性好、计算量小、便于工程实现等优点。     

基于GPS的公路视频影像信息系统

为了提高公路路况管理及路产管理的效率，开发了基于全球定位系统（GPS）的公路视频影像信息系统.通过视频影像的实时采集压缩和GPS定位，系统可实现基于里程和道路设施的多种视频影像搜索和定位功能，同时可根据电子地图快速定位到相应的视频图像帧.在论述了系统的体系结构后，推导了利用GPS坐标推算里程的算法，该算法通过录入关键帧对应的里程，内插出每帧的里程数，并对其进行修正.实践证明：该算法可有效地控制图像帧的里程误差，该系统可有效用于公路路况管理和公路路产管理.     

Accurate INS position solution during GPS outages or degraded GPS

Temporary degraded GPS (DGPS) position loss, in circumstances such as an overhead bridge, can be alleviated by an inertial navigation system (INS) that uses onboard sensors, such as yaw and speed sensors, to determine vehicle position. This paper introduces a post-processing DGPS/INS integration approach based on using the INS solution during DGPS outages or periods of low accuracy DGPS position solutions. In this approach, the INS solution initialization is performed using the DGPS solution before DGPS position solution loss, and measurements from the Inertial Measurement Unit (IMU). The final post-processed INS solution is a weighted average of the INS forward and backward solutions. This work constitutes two parts: the INS initialization methods for different degrees of freedom vehicle positioning models, and the developed weighting model necessary to combine the forward and the backward solutions. The former part is essential in obtaining acceptable INS initial states for both the stand-alone INS or any post-processing or real time INS/GPS integrated system. The latter part is based on the use of the complementary error behaviours of the backward and the forward solutions, and can be used as a survey method with acceptable position solutions accuracies. Applying the forward/backward INS combined solution method on real data shows that the resultant INS solution accuracy is 35 cm or less over a 1000 m road segment. This method is used to survey freeways interchange road segments where 50% of the surveyed distance has no DGPS solution or has a degraded DGPS solution. The average achieved accuracy over the whole freeways interchange is around 40 cm over a 23 km distance.     

基于蚁群算法的最短路径搜索方法研究

最短路径搜索是车载定位导航系统中很重要的一个功能,最短路径搜索问题本身也可以归结为组合优化问题.蚁群算法是基于群体的一种仿生算法,为求解复杂的组合优化问题提供了一种新思路.文章尝试采用蚁群算法来解决车载导航系统中的最短路径搜索问题,并在VC 6.0的环境下进行了仿真实验.实验结果表明,该方法能有效解决车载导航系统中的最短路径搜索问题,具有一定的理论参考价值和实际意义.     

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。     

车载导航系统陀螺的自适应校正方法

分析了低成本压电振动陀螺误差及其影响因素,在实验的基础上得出采用温度补偿陀螺误差的可行性。建立了联合卡尔曼滤波方程融合GPS和INS信息,估计定位信息和陀螺误差。提出车载GPS/INS组合导航系统中陀螺零漂误差和标度因子误差的校正过程启动条件,当条件满足时,以估计的陀螺误差为输入,采用温度误差校正表学习算法对陀螺误差模型进行训练。用道路实验数据对提出的陀螺校正算法进行验证,结果表明该算法精度高、收敛快、可操作性好。     

基于组件式GIS技术的导航电子地图设计

在车辆导航系统中,通常采用GPS来获得车辆的位置信息,并准确地在电子地图上显示和定位。可以认为通过GPS获得定位数据是GIS的一种数据采集方式。它可以实现车辆位置的实时、动态确定。利用目前较先进的组建式GIS技术开发了一套能够实时动态显示模拟GPS信息的电子地图,具有较高的实用和教学意义。     

高精度车载导航定位系统的研究

通过对车载导航系统结构、功能的分析，对车辆的高精度导航定位进行了研究。在常见的卫星导航（GPS）、航位推算（DR）等导航定位的基础上，进一步研究并实现了地图匹配算法，克服了GPS信号受阻时定位间断或失效的缺点，避免了航位推算定位误差随时间的积累。通过大量的跑车试验表明，基于GPS／DR／MM的车辆导航系统，可以高精度地实现车辆定位，进而可以实现地理信息的查询、交通诱导等功能。具有重要的实际应用价值。     

基于单线激光雷达的SLAM系统功能优化设计

地图与导航是无人驾驶与机器人领域的关键点,针对单线激光SLAM(同步定位与地图构建)系统功能优化问题,利用DOE(实验设计),提出了基于单线激光雷达的SLAM系统功能优化方法。首先,基于SLAM小车平台,利用Gmapping、Hector、Karto、Cartographer四种算法对室内环境进行建图,分析建图结果,提出最优建图方法。其次,依据AMCL定位算法,分别利用A*和Dijkstra两种路径规划方法,通过对比定位导航效果,提出基于参数匹配的定位导航功能优化方法。在此基础上,提出系统功能匹配优化方法,完成SLAM系统功能优化设计。     

车辆运动方程定位原理的研究及其应用

利用车辆运动方程获取车辆行驶距离,车辆当前位置点处的道路方向作为方向角,根据航位推算原理实时推算车辆位置。通过实地跑车试验分别分析了该定位技术在车载导航系统和监控中心的应用。试验结果显示,在车辆导航系统中该技术能实现GPS盲区内的车辆连续定位功能;在监控中心中,该技术从理论上可以大大减少单个车辆和监控中心之间的信息流量,其实际应用还有待智能交通系统的进一步完善。     

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。