Accurate and robust pose estimation is a prerequisite for autonomous driving but remains challenging in complex urban environments due to sensor degradation, such as Global Navigation Satellite System (GNSS) blockage, Light Detection and Ranging (LiDAR) geometric degeneracy, and wheel slip. Furthermore, conventional fusion frameworks typically employ fixed noise parameters, failing to adapt to time-varying sensor reliability. To address these issues, this paper proposes a tightly-coupled adaptive pose estimation framework integrating GNSS, LiDAR, Inertial Measurement Unit (IMU), and vehicle chassis sensors via Factor Graph Optimization (FGO). First, a novel Two Degrees of Freedom (2-DoF) vehicle dynamics pre-integration factor is constructed using wheel speed and steering angle measurements, constraining lateral and heading drift during external perception outages. Second, a hierarchical confidence assessment mechanism is designed. This mechanism quantifies heterogeneous sensor reliability in real-time and dynamically adjusts fusion weights through a designed noise mapping model combined with a progressive recovery strategy. Extensive experiments on a real-world vehicle in complex scenarios and the M2DGR dataset demonstrate the system’s effectiveness. Compared to the standard LiDAR-Inertial Odometry (LIO) baseline, the proposed method reduces the translation and rotation Root Mean Square Error (RMSE) by 15.1% and 20.4%, respectively, while maintaining a real-time latency of 30–55 ms.

Aiming at the problem of large trajectory tracking control errors for multi-axle vehicles on soft terrain, this study comprehensively considers complex wheel-terrain interaction relationships on soft terrain and develops a time-varying parameter-based trajectory tracking control algorithm for soft terrain using MPC (Model Predictive Control). Specifically, it establishes calculation methods for tire longitudinal/lateral forces on soft terrain and fits corresponding force curves. A vehicle dynamics model and trajectory tracking error model adapted to soft terrain conditions are constructed to enhance model accuracy. Considering load transfer during vehicle movement on soft terrain and the correlation between tire lateral forces and wheel loads, time-varying parameter matrices are established to accurately characterize vehicle dynamic characteristics under such conditions. By incorporating time-varying lateral force proportionality coefficients for wheel-terrain interaction, an MPC-based control algorithm is designed. Simulation tests of multi-axle vehicle trajectory tracking demonstrate the effectiveness of the proposed algorithm, showing significant improvement in control accuracy compared with conventional control methods that neglect parameter time-variation characteristics.

The vehicle localization system stands out as one of the most crucial components for intelligent vehicles. In urban canyon driving conditions, especially during GNSS (Global Navigation Satellite System) outages, an onboard sensor-based INS (Inertial Navigation System)/Chassis sensor fusion localization system emerges as the sole self-localization method that does not necessitate prior information and remains impervious to external environments. To improve the performance of the onboard sensor-based localization system, a parallel INS-based vehicle localization framework is proposed, which considers sensor mounting errors and vehicle lateral motion during steering. Firstly, the INS mechanism, fundamental to the proposed framework, is introduced. A course angle-aided method for estimating the mounting angle of the IMU (Inertial Measurement Unit) frame and the vehicle frame is conducted, and the calibrated mounting angles contribute to the collaboration of the proposed parallel INS. Subsequently, a detailed description of the proposed parallel INS-based localization framework is presented. In this framework, the first INS is responsible for outputting localization results, while the second INS is dedicated to lateral velocity estimation, providing velocity measurements for INS error correction. Finally, comprehensive field tests were conducted to verify the proposed method, demonstrating superior performance compared to traditional methods. In our experiment, the average position accuracy is improved by around 70% during sharp turning, and improved by over 34% during the whole GNSS outage period.

Intelligent transportation systems (ITS) localization is of significant importance as it provides fundamental position and orientation for autonomous operations like intelligent vehicles. Integrating diverse and complementary sensors such as global navigation satellite system (GNSS) and 4D-radar can provide scalable and reliable global localization. Nevertheless, multi-sensor fusion encounters challenges including heterogeneity and time-varying uncertainty in measurements. Consequently, developing a reliable and unified multi-sensor framework remains challenging. In this paper, we introduce UniMSF, a comprehensive multi-sensor fusion localization framework for ITS, utilizing factor graphs. By integrating a multi-sensor fusion front-end, alongside outlier detection&noise model estimation, and a factor graph optimization back-end, this framework accomplishes efficient fusion and ensures accurate localization for ITS. Specifically, in the multi-sensor fusion front-end module, we tackle the measurement heterogeneity among different modality sensors and establish effective measurement models. Reliable outlier detection and data-driven online noise estimation methods ensure that back-end optimization is immune to interference from outlier measurements. In addition, integrating multi-sensor observations via factor graph optimization offers the advantage of plug and play. Notably, our framework features high modularity and is seamlessly adapted to various sensor configurations. We demonstrate the effectiveness of the proposed framework through real vehicle tests by tightly integrating GNSS pseudorange and carrier phase information with IMU, and 4D-radar.

Positioning system is a key module of autonomous driving. As for LiDAR SLAM system, it faces great challenges in scenarios where there are repetitive and sparse features. Without loop closure or measurements from other sensors, odometry match errors or accumulated errors cannot be corrected. This paper proposes a construction method of LiDAR anchor constraints to improve the robustness of the SLAM system in the above challenging environment. We propose a robust anchor extraction method that adaptively extracts suitable cylindrical anchors in the environment, such as tree trunks, light poles, etc. Skewed tree trunks are detected by feature differences between laser lines. Boundary points on cylinders are removed to avoid misleading. After the appropriate anchors are detected, a factor graph-based anchor constraint construction method is designed. Where direct scans are made to anchor, direct constraints are constructed. While in the position where the anchor is not directly observed, the prior information in the structured road is used to construct a constraint group to improve the overall positioning. When a previously established anchor point is observed at any remote location, the global trajectory can be corrected without waiting for loop closure to occur. We verify the improvement theoretically and experimentally. The result of experiment shows that under the LeGO-LOAM-based factor graph framework, the odometry is improved significantly. The pose correction is realized by long-distance anchor detection when loop closure is not fully realized. Our method can be easily plugged in various factor graph-based systems and has good adaptability and scalability.

Using vehicle chassis information can significantly improve the accuracy of GNSS/INS fusion system when GNSS signals are unavailable. The IMU mounting angles play an important role when fusing vehicle chassis information with GNSS/INS system. However, the IMU mounting angles cannot be directly measured conveniently. This paper proposed a new method to simultaneously improve the estimation accuracy of both the vehicle heading angle and the IMU heading mounting angle by leveraging GNSS course angle as an additional measurement. Firstly, an error state Kalman filter is constructed with state variables including the attitude errors, velocity errors, position errors, gyro, and acceleration bias errors, IMU mounting angle and vehicle velocity scale factor. The GNSS course angle is augmented to the Kalman filter as a measurement when the vehicle travels in straight line. Then, the observability analysis of the GNSS/INS/Onboard sensors fusion system is carried out and the results show that the observability of the system using GNSS course angle is better than that of only using vehicle velocity and non-holonomic constraint (NHC) as measurements if the heading mounting angle and pitch mounting angle are not zero. Finally, the experiment results show that the accuracy of the vehicle heading angle and the IMU heading mounting angle can be improved by 14% judging from the lateral velocity error of vehicle compared to that of only using vehicle velocity and NHC as measurements.

Vehicle localization is essential for intelligent and autonomous vehicles. To improve the accuracy of vehicle stand-alone localization in highly dynamic driving conditions during GNSS (Global Navigation Satellites Systems) outages, this paper proposes a vehicle localization system based on vehicle chassis sensors considering vehicle lateral velocity. Firstly, a GNSS/On-board sensors fusion localization framework is established, which could estimate vehicle states such as attitude, velocity, and position. Secondly, when the vehicle has a large lateral motion, nonholonomic constraint in the lateral direction loses fidelity. Instead of using nonholonomic constraint, we propose a vehicle dynamics/kinematics fusion lateral velocity estimation algorithm, which combines the advantage of vehicle dynamic model in low dynamic driving conditions and the advantage of kinematic model in highly dynamic driving conditions. Thirdly, vehicle longitudinal velocity estimated by WSS (Wheel Speed Sensor) and lateral velocity estimated by proposed method are as measurements for the localization system. All information is fused by an adaptive Kalman filter. Finally, vehicle experiments in U-turn maneuver and left-turn maneuver at a traffic intersection are conducted to verify the proposed method. Four different methods are compared in the experiments, and the results show that the estimated position accuracy of our method is below half a meter during a 5s GNSS outage and could keep a sub-meter-level during a 20s GNSS outage while the vehicle has a relatively large lateral motion.

定位系统是智能车辆环境感知系统的重要组成部分.设计了智能车辆轮胎半径自适应在线估计算法以提高车辆速度估计精度,从而在GNSS(Global Navigation Satellites System)不可用时提升IMU(inertial measurement unit)/WSS(wheel speed sensor)组合定位系统的精度.首先,在GNSS信号良好时,考虑车轮动态设计了多模型融合的轮胎有效滚动半径自适应算法,以准确估计轮胎有效滚动半径;然后,基于自适应误差状态卡尔曼滤波设计了多传感器融合组合定位算法.实车试验结果表明,所设计的算法在初始轮胎半径有不足2％的误差时,丢失GNSS 40s可将定位精度提高30％以上.

In the development of automotive electronics, nearly every automobile is equipped with an inertial measurement unit (IMU). However, the yaw misalignment of the IMU is inevitable when mounted to a vehicle's body. It is difficult to measure directly, so its estimation is required to acquire accurate data from the IMU. This study proposes a method for the IMU and automotive onboard sensors to estimate the yaw misalignment autonomously. In order to estimate the IMU yaw misalignment, first, the attitude and velocity integration method in the vehicle level frame is presented. In addition, the attitude error dynamics consisting of yaw misalignment, pitch, and roll, and velocity error dynamics consisting of longitudinal and lateral velocities, are derived. Then, on the basis of the error dynamics and observation equations, the degree of the observability of the yaw misalignment is analyzed through the piece-wise constant system (PWCS) and singular value decomposition (SVD) theory. Next, a Kalman filter is implemented to estimate the yaw misalignment and the velocity error. Finally, an experimental test in straight line acceleration and deceleration maneuvers is conducted to verify the yaw misalignment estimation method. When the longitudinal or lateral acceleration varies, the yaw misalignment is observable and can be estimated without aids from external information. After compensating for the yaw misalignment, the accuracy of the state estimation result, such as lateral velocity, can be improved significantly when the IMU is integrated with other sensors.

Vehicle slip angle (VSA) estimation is of paramount importance for connected automated vehicle dynamic control, especially in critical lateral driving scenarios. In this paper, a novel kinematic-model-based VSA estimation method is proposed by fusing information from a global navigation satellite system (GNSS) and an inertial measurement unit (IMU). First, to reject the gravity components induced by the vehicle roll and pitch, a vehicle attitude angle observer based on the square-root cubature Kalman filter (SCKF) is designed to estimate the roll and pitch. A novel feedback mechanism based on the vehicle intrinsic information (the steering angle and wheel speed) for the pitch and roll is designed. Then, the integration of the reverse smoothing and grey prediction is adopted to compensate for the cumulative velocity errors during the relatively low sampling interval of the GNSS. Moreover, the GNSS signal delay has been addressed by an estimation-prediction integrated framework. Finally, the results confirm that the proposed method can estimate the VSA under both the slalom and double lane change (DLC) scenarios.

In this paper, a VSA estimation method is proposed based on fusing the vehicle kinematics and dynamics. First, the vehicle-kinematic-based (VK-based) VSA estimation method is provided by the global navigation satellite system (GNSS) and inertial navigation system (INS) integration system (GNSS/INS integration system). The heading error in GNSS/INS integration is not well observable and its estimation accuracy cannot be guaranteed when the acceleration in the horizontal plane is small and varies little. To improve the heading error estimation accuracy, a vehicle-dynamic-model-based (VDM-based) VSA estimation method is given and based on this method, a novel augmented heading estimator for the GNSS/INS integration system is designed. Besides, an intuitive heading error weighting strategy is presented to determine the heading error between the heading error from GNSS/INS integration system and that from the augmented heading error estimator. Finally, the proposed method is validated by a comprehensive test.

Improving the performance of low-cost inertial sensors is meaningful for popularization of automated driving technologies. Researchers usually utilize thermal chamber to calibrate the temperature drift errors (TDEs) of expensive inertial sensors and compensate them before using them. However, it is unacceptable for low-cost inertial sensors considering the massive calibration efforts. This paper proposes a novel method to model TDE as a state so that it can be estimated online with other states together, which can adaptively compensate different TDEs without preparation. A multi-sensor fusion system for estimating yaw angle with this idea is studied. The observability of this system is analyzed and the result shows that TDE is independent from others states when there is change of temperature. Experiments are carried out and the results reveal that the performance of inertial sensor assisted with the proposed method is better than that in normal system.

以赣江上游章水流域内15个雨量站1957-2016年逐日降水资料为基础,获取了12个极端降水指数.分别采用线性趋势、Modified Mann-Kendall非参数检验法、累积距平法、Pettitt突变检验法、小波分析、Hurst指数分析和克里金插值等方法对12个极端降水指数的时空变化特征进行探讨.结果表明:1957-2016年间,暴雨量及日数、强降水量及贡献率和最大湿润日数均呈现上升趋势,极强降水量及贡献率、最大干旱日数、年最大雨量、连续5日最大雨量呈现下降趋势,且其未来趋势均将得到加强;大雨量及日数呈现上升趋势,但未来趋势将可能减弱;所有极端降水指数的过去趋势均不具有显著性,同时未发生显著的突变,并且普遍存在以1988年为中间轴的33～34 a主周期;大雨量及日数、暴雨量及日数、强降水量、最大湿润日数、年最大雨量、连续5日最大雨量的高值中心均在流域西南区域的内良站附近,并呈上升趋势,低值区域主要分布在流域的东部区域,并呈现下降趋势,且其整体空间分布大致相同,均为西南向东北递减;而强降水贡献率及极强降水贡献率为东南向西北递减,最大干旱日数是从东向西的递减趋势.未来章水流域极端降水的变化趋势可能还会增强,加强流域极端降水事情的预防对流域水旱灾害防治意义重大.

Vehicle localization system is one of the most important systems of autonomous vehicles. To improve the localization accuracy during Global Navigation Satellite System (GNSS) outages, this paper presents a GNSS/Inertial Measurement System (IMU)/Wheel speed sensor (WSS) integrated localization system considering vehicle dynamics. The vehicle dynamics model and kinematics model are applied to estimate sideslip angle, which is used to calculate course angle of the vehicle so that the accurate vehicle speed in navigation coordinates could be obtained. When the GNSS measurements are available, the position measurements and heading angle measurements are fed back to the system, and all the sensor information is fused in a Kalman filter. Experiments were conducted to verify the proposed fusion method, and the results show that the consideration of vehicle dynamic characteristics is helpful to improve the localization accuracy during GNSS outages.

The sideslip angle and attitude are crucial for automated driving especially for chassis integrated control and environmental perception. In this article an inertial measurement unit (IMU)-based automated vehicle body sideslip angle and attitude estimation method aided by low-sample-rate global navigation satellite system (GNSS) velocity and position measurements using parallel adaptive Kalman filters is proposed. This method can estimate the sideslip angle and attitude simultaneously and is robust against the vehicle parameters and road friction even as the vehicle enters critical maneuvers. First, based on the acceleration and angular rate from the six-dimensional inertial measurement unit, the attitude, velocity and position (AVP) are integrated with the navigation coordinates and the AVP error dynamics and observation equations of the integration results are developed. Second, parallel innovation adaptive estimation (IAE)-based Kalman filters is designed to estimate the AVP error of the integration method to address the issues of the GNSS low sampled rate and abnormal measurements. Then the AVP error is forwarded to the AVP integration to compensate the accumulated error. To improve the heading angle estimation accuracy, the heading error is estimated by a decoupled IAE-based Kalman filter aided by GNSS heading. In addition, time synchronization of the IMU and GNSS is realized through hardware based on the pulse per second signal of the GNSS receiver and the spatial synchronization is achieved by a direct compensation method. Lastly, the sideslip angle and attitude estimation method is validated by a comprehensive experimental test including critical double lane change and slalom maneuvers. The results show that the estimation error of the longitudinal velocity and lateral velocity is smaller than 0.1 m/s (1σ), and the estimation error of the sideslip angle is smaller than 0.15° (1σ).

低速智能电动汽车近年来发展迅速,组合定位技术是其关键技术,航向角估计是组合定位技术中重要组成部分.基于低速智能电动汽车,提出了GNSS(global navigation satellite system)/IMU(inertial measurement unit)组合的航向角估计方法.介绍了GNSS/IMU松耦合条件下的航向角估计方法,提出基于IMU的航向角积分方法,推导了松耦合条件下误差动态与测量模型.针对GNSS信号质量时变问题,使用残差自适应卡尔曼滤波算法对航向角误差进行估计.针对GNSS信号质量设计了航向角误差反馈修正策略.通过在不同GNSS信号条件下进行的多组实车试验,验证了所提出的航向角估计算法的有效性.

The vehicle sideslip angle is an important state for vehicle dynamic control, which needs to be estimated as it could not be obtained directly by the vehicle. To improve the estimation accuracy of the sideslip angle based on the intelligent vehicle platform, this study proposes a novel vehicle sideslip angle estimation algorithm with the fusion of dynamic model and vision information. Firstly, to further improve the model accuracy of the vehicle during lateral acceleration conditions, a vehicle dynamic model is established considering the acceleration error compensation with the assistance of attitude information. In addition, based on the lane line information obtained from the equipped camera in intelligent vehicles, a visual geometric model is established. Owing to the measurement delay and low sampling frequency of the camera, a multi-rate sideslip angle observer with delay compensation is designed to coordinate with the inter-frequency signal of the vehicle chassis. Finally, the effectiveness of the algorithm is verified by the slalom test.

Estimation of the sideslip angle is significant for vehicle safety control systems such as electronic stability control. This paper proposes a vehicle-kinematic-model-based sideslip angle estimation method by fusing the information from an inertial measurement unit (IMU) and global navigation satellite system (GNSS) with aligning the heading from the GNSS. To estimate the velocity and attitude errors of the reduced inertial navigation system (R-INS), we first formulate the associated system error dynamics. Then, to further improve the heading estimation accuracy of the R-INS, the heading from the GNSS is aligned to the vehicle longitudinal direction by a robust regression method and adopted to estimate the heading error of the R-INS. Next, an adaptive Kalman filter is applied to estimate the errors in the R-INS to attenuate the noise influence. With the velocity in navigation coordinates and the attitude between the navigation coordinates and vehicle body coordinates from the R-INS, the velocity and sideslip angle in the vehicle body coordinates are computed. Finally, tests in straight line, double lane change (DLC), and slalom maneuvers are per-formed to verify the sideslip angle estimation and the heading alignment method. After aligning the heading from the GNSS, the sideslip angle estimation accuracy is improved, and the mean error under typical DLC and slalom maneuvers are below 0.21 degrees. (c) 2020 Elsevier Ltd. All rights reserved.

Traditional vehicle localization systems use GNSS signals for global localization or Lidar or vision based systems for Simultaneous Mapping and Localization (SLAM). However, the signals of GNSS might be contaminated or even blocked in areas where there are tall buildings or trees, especially for autonomous sweeper vehicles that usually work along the road. There are additional information that can be used to improve the accuracy and robustness of vehicle localization system when applied in autonomous sweeper vehicles. Thus, a novel vehicle localization system based on GNSS, vision, wheel speed and Inertial Measurement Unit (IMU) is proposed. When the vehicle works in areas where GNSS signals are good, accurate position information from GNSS is used. Besides, wheel speed and IMU information is fused to improve the output rate of position information. When the vehicle enters areas where GNSS signals are unavailable, the heading information from vision is used as measurement to estimate the heading angle of vehicle and then fused with wheel speed and IMU to estimate the position. Real vehicle experiments are conducted to validate the effectiveness of the proposed system.

为研究江西省桃江流域降水量特征,以桃江流域内的17个雨量站1957—2016年逐日降水资料为研究对象,运用线性趋势、Mann-Kendall趋势检验、克里金插值等方法进行时空特征分析.结果表明,时间变化特征上:桃江流域多年平均值为1603.49mm,且以2.4mm/10a的趋势增加,其第1主周期为34a,并且经历了"偏多—偏少—偏多"的周期变化;季节分配不均匀,春夏多,秋冬少,且秋、冬季变化剧烈,除了夏季以-4.8mm/10a的速率下降,其余季度均呈现上升趋势,分别为春季(7.3mm/10a)、秋季(7.1mm/10a),冬季降水(5.3mm/10a).在周期规律上,除冬季的第1主周期为34a外,其余季度为33a;月降水量呈现出"两峰一谷"的形状,降水主要集中在3-8月,10月到翌年1月的降水变化剧烈.空间分布特征:桃江流域年降水量从西南到东北呈现出"偏多-偏少-偏多"的空间分布特征,春、夏两季降水量呈现出南多北少的分布特征,秋、冬季降水量空间分布从南到北呈现出"偏多—偏少—偏多"的分布特征;年降水量北部区域偏多,主要受秋冬季节降水影响;除了冬季的高值中心出现在下游区域外,其他季节均出现在上游流域;上游流域在1-9月均为各区域降水的最大值,而10-12月,下游流域为各区域的最大值;各区域降水在月份分布上均呈现出"双峰一谷"的分布特征,3-9月,各区域降水间存在着较大的差异.研究结果可为桃江流域防洪抗旱及水生态修护提供依据.