融合多模型的粒子滤波运动目标实时跟踪算法

提出了一种融合多模型的粒子滤波跟踪新算法(MMGPF),并将其应用于行人与汽车跟踪。此跟踪算法特点在于:(1)将Camshift跟踪算法和AdaBoost分类器的输出作为观测值优化建议概率分布;同时,改进粒子滤波的算法结构,有效地提高了粒子滤波的采样效率;在不影响跟踪性能的情况下,大幅减少了跟踪所需粒子数。(2)用两种描绘子提高对似然性的估计。(3)采用两种有效措施提高算法的实时性。通过多模型融合,有效地解决了目标跟踪过程中由于目标相互遮挡、目标消失再重现、光照变化和目标与背景颜色相近所造成的跟踪丢失。行人和汽车的跟踪试验结果证明该算法具有鲁棒性和实时性。     

Optimal control of brakes and steering for autonomous collision avoidance using modified Hamiltonian algorithm

ABSTRACT

This paper considers the problem of collision avoidance for road vehicles, operating at the limits of friction. A two-level modelling and control methodology is proposed, with the upper level using a friction-limited particle model for motion planning, and the lower level using a nonlinear 3DOF model for optimal control allocation. Motion planning adopts a two-phase approach: the first phase is to avoid the obstacle, the second is to recover lane keeping with minimal additional lateral deviation. This methodology differs from the more standard approach of path-planning/path-following, as there is no explicit path reference used; the control reference is a target acceleration vector which simultaneously induces changes in direction and speed. The lower level control distributes vehicle targets to the brake and steer actuators via a new and efficient method, the Modified Hamiltonian Algorithm (MHA). MHA balances CG acceleration targets with yaw moment tracking to preserve lateral stability. A nonlinear 7DOF two-track vehicle model confirms the overall validity of this novel methodology for collision avoidance.     

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.     

基于最大相关准则图像分割的结构化道路路径识别和跟踪算法

为提高基于视觉导航的智能车辆对结构化道路车道标识线的识别和跟踪精度,同时消除车流、阴影和光照不均匀等不利因素的影响,提出一种基于最大相关准则的图像分割算法及基于感兴趣区域的车道标识线跟踪算法:首先,对图像进行滤波和光线补偿等前期处理,采用最大相关准则的图像分割算法对道路图像进行阈值分割;然后,根据车道的结构特征及先验知识提取车道标识线的特征点,并运用最小二乘法对特征点拟合,得到车道模型的参数;最后,通过建立感兴趣区域(ROI)的方法实现对车道标识线的准确跟踪。试验结果表明,该算法具有很好的准确性、实时性和鲁棒性。     

基于强化学习DDPG的智能车辆轨迹跟踪控制

针对智能车辆在轨迹跟踪过程中的横向控制问题，提出一种基于强化学习中深度确定性策略梯度算法（Deep Deterministic Policy Gradient，DDPG）的智能车辆轨迹跟踪控制方法。首先，将智能车辆的跟踪控制描述为一个基于马尔可夫决策过程（MDP）的强化学习过程，强化学习的主体是由Actor神经网络和Critic神经网络构成的Actor-Critic框架；强化学习的环境包括车辆模型、跟踪模型、道路模型和回报函数。其次，所提出方法的学习主体以DDPG方法更新，其中采用回忆缓冲区解决样本相关性的问题，复制结构相同的神经网络解决更新发散问题。最后，将所提出的方法在不同场景中进行训练验证，并与深度Q学习方法（Deep Q-Learning，DQN）和模型预测控制（Model Predictive Control，MPC）方法进行比较。研究结果表明：基于DDPG的强化学习方法所用学习时间短，轨迹跟踪控制过程中横向偏差和角偏差小，且能满足不同车速下的跟踪要求；采用DDPG和DQN强化学习方法在不同场景下均能达到训练片段的最大累计回报；在2种仿真场景中，基于DDPG的学习总时长分别为DQN的9.53%和44.19%，单个片段的学习时长仅为DQN的20.28%和22.09%；以DDPG、DQN和MPC控制方法进行控制时，在场景1中，基于DDPG方法的最大横向偏差分别为DQN和MPC的87.5%和50%，仿真时间分别为DQN和MPC的12.88%和53.45%；在场景2中，基于DDPG方法的最大横向偏差分别为DQN和MPC的75%和21.34%，仿真时间分别为DQN和MPC的20.64%和58.60%。     

基于当前统计模型的地图匹配车辆跟踪研究

为提高车辆跟踪系统的精度，减小差分全球定位系统（DGPS）的定位误差，通过分析行驶在城市道路上的车辆运动过程及其相应的运动模型，提出采用当前统计模型作为车辆运动模型。通过地图辅助位置择近和速度择角算法来修正卡尔曼滤波，为运行在道路上的车辆确定地图匹配估计。实际运行结果表明；整个车辆跟踪系统的精度有明显的提高。     

Vision-based fusion of robust lane tracking and forward vehicle detection in a real driving environment

With the goal of developing an accurate and fast lane tracking system for the purpose of driver assistance, this paper proposes a vision-based fusion technique for lane tracking and forward vehicle detection to handle challenging conditions, i.e., lane occlusion by a forward vehicle, lane change, varying illumination, road traffic signs, and pitch motion, all of which often occur in real driving environments. First, our algorithm uses random sample consensus (RANSAC) and Kalman filtering to calculate the lane equation from the lane candidates found by template matching. Simple template matching and a combination of RANSAC and Kalman filtering makes calculating the lane equation as a hyperbola pair very quick and robust against varying illumination and discontinuities in the lane. Second, our algorithm uses a state transfer technique to maintain lane tracking continuously in spite of the lane changing situation. This reduces the computational time when dealing with the lane change because lane detection, which takes much more time than lane tracking, is not necessary with this algorithm. Third, false lane candidates from occlusions by frontal vehicles are eliminated using accurate regions of the forward vehicles from our improved forward vehicle detector. Fourth, our proposed method achieved robustness against road traffic signs and pitch motion using the adaptive region of interest and a constraint on the position of the vanishing point. Our algorithm was tested with image sequences from a real driving situation and demonstrated its robustness.     

基于车联网BSM数据与路侧视频融合的港口集装箱卡车碰撞危险辨识方法

大型港口集装箱码头运输车辆调度频繁,堆场过道和交换区等区域视距狭窄,容易导致港口集装箱卡车与设施、作业人员和车辆发生擦碰事故。为提高智能集装箱卡车在港口密集区域的轨迹跟踪精度和行车安全感知能力,提出了一种车联网条件下融合车载终端基本安全消息（Basic Safety Messages,BSM）数据和路侧视频数据的集装箱卡车碰撞风险辨识方法。采用YOLOv5s算法提取视频监控范围内的目标车辆和作业人员,根据目标集卡大尺寸特点设计非极大值抑制锚框来提高目标识别准确度。运用透视变换原理将目标像素坐标转换成地理坐标,并应用Deep-SORT算法匹配每帧图像的车辆轨迹信息。应用交互式多模型方法（interactive multi-model,IMM）融合视频轨迹信息和车载单元（on-board units,OBU）定位数据,减小了目标机动过程中的观测误差。基于集卡融合轨迹结果,提出了1种新型的轨迹冲突风险评估模型,能够根据目标集卡与周围目标轨迹的相对运动状态实时感知车辆碰撞危险,该碰撞危险检测结果在实际场景中可通过路侧设备对车载终端和作业人员终端实时播发预警信息。针对集卡跟踪误差的实验结果表明：...     

基于预测恒定车头时距策略的自适应巡航控制研究

为了进一步提高车辆跟车过程中的跟踪性、安全性、舒适性和燃油经济性，针对已有间距策略表现过于保守或反应过于激烈等不足之处，提出了一种预测恒定车头时距策略。该策略考虑了相对加速度，建立了一种预测型期望车间距模型，进而应用于模型预测控制的多目标自适应巡航控制系统中，能进一步提高模型预测控制对多个控制目标的综合协调能力。搭建上层控制器、下层PID控制器、油门制动切换、逆纵向动力学模型。在多工况下仿真，通过建立性能评判指标对多目标进行量化分析。结果表明，所提出的间距策略在保证安全性的前提下，提升了自适应巡航控制系统的综合性能。在不同驾驶风格的车头时距下，跟踪性、舒适性和燃油经济性均有良好表现。     

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。     

面向车道变换的路径规划及模型预测轨迹跟踪

为解决智能车辆在车道变换过程中的路径规划和路径跟踪问题,首先,利用梯形加速度法设计了车道变换虚拟理想轨迹,该路径规划方法的适应性取决于车道变换时间、横向加速度及变化率等关键变量的约束条件,因而对各关键变量之间的数学关系进行了定量计算,并绘制了不同工况下的车道变换虚拟理想轨迹,用于分析各关键变量对路径规划的影响;其次,建立了线性离散的车辆动力学预测模型,综合分析了车辆模型的控制输入、状态变量以及道路结构参数等约束条件,构建了多约束模型预测控制（MMPC）系统用于车道变换路径跟踪,并基于Hildreth二次规划算法对其目标函数进行了求解,获得前轮转向角控制量,从而保证智能车辆在车道变换过程中的路径跟踪性能及操纵稳定性能;最后,利用MATLAB和Carsim软件对提出的多约束模型预测控制系统进行联合仿真,并构建单约束模型预测控制（SMPC）系统与其进行性能比较,分别对车道变换时间为3 s和6 s时的车道变换性能进行比较分析。结果表明：当车道变换时间为6 s时,2种控制系统都能较好地实现车道变换功能;当车道变换时间为3 s时,与SMPC控制系统相比较,MMPC控制系统能够在有效跟踪期望行驶路径的同时改善车辆的操纵稳定性,从而提高车辆在路径跟踪过程中的主动安全性能。     

基于NMPC的智能汽车纵横向综合轨迹跟踪控制

本文中针对大曲率转弯工况下,智能汽车纵横向动力学特性的耦合和动力学约束导致轨迹跟踪精度和稳定性下降的问题,提出一种基于非线性模型预测控制(NMPC)的纵横向综合轨迹跟踪控制方法,通过NMPC和障碍函数法(BM)的有效结合,提高了跟踪精度,改善了行驶稳定性.首先建立四轮驱动-前轮转向智能汽车动力学模型和轨迹跟踪模型,采用...     

基于高精地图的物流配送车路径规划与跟踪控制

为了解决园区等场景下无人车多途经点配送问题,提出了一种基于矢量化高精地图的车道级全局路径规划、生成和跟踪控制方法。考虑配送车往返途经点顺序对行驶路径总长度的影响,基于高精地图采用A*算法计算各配送点间的最优路径,在此基础上,利用动态规划算法求解经过多个配送点的全局最优路径。应用贝塞尔曲线对规划的路径进行平滑,并根据道路曲率设定不同路径处的参考行驶速度,进而生成车道级的可用于跟踪的目标轨迹。利用车辆二自由度模型设计模型预测控制器进行轨迹跟踪,实现低速物流配送车的自主控制。在 CarSim/Prescan/Simulink联合仿真平台和实车平台上对提出的规划控制方法进行了试验。结果表明,相比传统的依据最近配送点策略确定的路径,所提出的方法搜索出的路径长度平均缩短了 6.15%。所设计的轨迹跟踪控制器能确保配送试验车与目标轨迹的横向偏差在 0.25 m 以内,航向角偏差在5°以内。     

基于小波和粒子群算法的HEV行驶状况辨识方法研究

针对混合动力汽车(HEV)行驶状况(道路坡度和整车载荷)变化难以有效识别,导致驱动系统控制策略不能有效满足驾驶员意图问题,以混联式HEV为研究对象,提出了基于小波滤波和粒子群算法的HEV行驶状况辨识方法。首先建立了汽车行驶状况辨识模型,采用最小二乘法确立了优化目标函数,其次研究了基于小波滤波和粒子群算法的HEV行驶状况辨识原理,最后进行了行驶状况粒子群智能算法辨识试验。在采集实车数据的基础上,对实车数据进行小波滤波,并运用行驶状况辨识方法对道路坡度和整车载荷进行了辨识,并对辨识结果进行小波滤波,结果表明,试验工况下整车载荷辨识的相对误差绝对平均值为2.71%,道路坡度辨识的相对误差绝对平均值为3.85%,验证了所提出方法的有效性。     

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.     

基于变权重系数的分布式驱动无人车轨迹跟踪

为了提高分布式无人车轨迹跟踪的精度，提出了基于自主与差动协调转向控制的轨迹跟踪方法。首先，在车辆三自由度模型基础上，基于模型预测控制（MPC）实时计算前轮转角以控制车辆进行自主转向轨迹跟踪。在此过程中，为了提高自主转向下车辆的轨迹跟踪精度与行驶的稳定性，考虑多种因素，利用经验公式及神经网络控制对MPC的预瞄步数和预瞄步长进行多参数调整，实现预瞄时间的自适应控制。其次，在恒转矩需求的情况下，以轨迹偏差为PID控制器的输入及左右轮毂电机转矩为输出进行差动转向控制，实现了差动转向下的轨迹跟踪控制。然后，通过设置权重系数的方法将自主与差动转向相结合。考虑到车辆横纵向动力学因素，采用模糊控制及经验公式对权重系数进行了调整，从而在提高车辆转向灵活性与轨迹跟踪效果的同时保证车辆行驶的稳定性。CarSim与Simulink联合仿真以及实车试验结果表明：与自主转向轨迹跟踪相比，采用变权重系数的协调控制可以在不同的工况下提高车辆的转向灵活性与轨迹跟踪的精度，轨迹跟踪偏差的均方根值改善率达到了11%。所提出的协调转向控制方法可为分布式驱动车辆转向灵活性的提高及轨迹跟踪精度的改善提供一种新的思路。     

基于地空信息融合的无人机交通状态感知方法研究

现有的无人机（UAV）交通状态感知方法，主要针对宏观交通状态参数的获取，同时尚未克服UAV自运动对交通参数检测精度的影响，难以满足智能交通系统对于高精度微观交通参数的应用需求。为此，提出一种基于地空信息融合的UAV交通状态感知方法，该方法包括：地空信息融合模型、道路关键点（IKP）检测及跟踪、车辆目标检测及追踪算法和交通状态参数提取及估计。其中，地空信息融合模型利用地基信息（IKP世界坐标）与空基信息（IKP像素坐标）进行最优化融合，并通过自适应IKP追踪算法与自适应UAV位置偏移判断算法实时更新模型参数，以此克服UAV自运动对车辆轨迹精度的影响，进而获取可靠的车辆级（瞬时速度、车头间距和车头时距）与车道级（车道动态密度、车道流量和空间平均车速）交通状态参数。利用提出的感知方法获取实地拍摄视频的车辆级交通参数并进行了分布检验，同时比较了基于不同交通流模型的车道级参数估算方法。结果表明：该方法在车辆检测的mAP@0.5指数超过90%，同时提取的车辆轨迹相对完整，获取的车辆级和车道级交通状态参数也符合实际交通流状况。最后，将该模型应用于实地道路的交通拥堵检测及交通事件检测，该研究结果为UAV在现代交通感知和管理中的应用提供了一种理论和技术参考。     

A vehicle stability control strategy with adaptive neural network sliding mode theory based on system uncertainty approximation

Modelling uncertainty, parameter variation and unknown external disturbance are the major concerns in the development of an advanced controller for vehicle stability at the limits of handling. Sliding mode control (SMC) method has proved to be robust against parameter variation and unknown external disturbance with satisfactory tracking performance. But modelling uncertainty, such as errors caused in model simplification, is inevitable in model-based controller design, resulting in lowered control quality. The adaptive radial basis function network (ARBFN) can effectively improve the control performance against large system uncertainty by learning to approximate arbitrary nonlinear functions and ensure the global asymptotic stability of the closed-loop system. In this paper, a novel vehicle dynamics stability control strategy is proposed using the adaptive radial basis function network sliding mode control (ARBFN-SMC) to learn system uncertainty and eliminate its adverse effects. This strategy adopts a hierarchical control structure which consists of reference model layer, yaw moment control layer, braking torque allocation layer and executive layer. Co-simulation using MATLAB/Simulink and AMESim is conducted on a verified 15-DOF nonlinear vehicle system model with the integrated-electro-hydraulic brake system (I-EHB) actuator in a Sine With Dwell manoeuvre. The simulation results show that ARBFN-SMC scheme exhibits superior stability and tracking performance in different running conditions compared with SMC scheme.     

Vehicle handling dynamics modelling by a data-driven identification method

Modelling of vehicle handling dynamics has received a renewed attention in recent years. Different from traditional vehicle modelling, a novel data-driven identification method for vehicle handling dynamics is proposed, which can avoid the problems of the under-modelling and parameter uncertainties in the first-principle modelling process. By first-order Taylor expansion, the nonlinear vehicle system can be linearised as a slowly linear time-varying system with fourth-order. In order to identify the derived identifiable model structure, a recursive subspace method is presented. Derived by optimal version of predictor-based subspace identification (PBSID_opt) and projection approximation subspace tracking (PAST), the identification method is numerical stability and gives an unbiased estimation for the closed-loop system. Based on standard road tests, the proposed modelling method is proven effective and the obtained model has good predictive ability. Additionally, it is noted that the model obtained from the initial phase of straight driving is just a mathematical model to describe the relationship between input and output. And when the vehicle is steering, the model can converge to a stable phase quickly and represent vehicle dynamic performance.     

Robust yaw stability control for electric vehicles based on active front steering control through a steer-by-wire system

A robust yaw stability control design based on active front steering control is proposed for in-wheel-motored electric vehicles with a Steer-by-Wire (SbW) system. The proposed control system consists of an inner-loop controller (referred to in this paper as the steering angle-disturbance observer (SA-DOB), which rejects an input steering disturbance by feeding a compensation steering angle) and an outer-loop tracking controller (i.e., a PI-type tracking controller) to achieve control performance and stability. Because the model uncertainties, which include unmodeled high frequency dynamics and parameter variations, occur in a wide range of driving situations, a robust control design method is applied to the control system to simultaneously guarantee robust stability and robust performance of the control system. The proposed control algorithm was implemented in a CaSim model, which was designed to describe actual in-wheel-motored electric vehicles. The control performances of the proposed yaw stability control system are verified through computer simulations and experimental results using an experimental electric vehicle.