激光与单目视觉融合的移动机器人运动目标跟踪

针对室内环境下的移动机器人运动目标跟踪问题,提出一种基于激光与单目视觉传感信息融合的机器人定位和目标运动估计方法.首先,利用激光传感信息实现对目标的检测,并完成机器人定位与环境建图;然后,设计一种基于单目视觉传感器的目标位置估计算法,获得目标的距离和角度信息;为了实现两类传感信息的有效融合,将激光与单目视觉进行联合标定,得到二者的相对位姿关系,基于此,将激光与单目视觉提取的目标距离和角度通过具有最优重要性函数和权重的粒子滤波器进行融合,实现对目标运动状态的准确估计.实验结果表明该方法具有良好的跟踪性能.     

基于t-分布粒子滤波器的目标跟踪

针对贝叶斯跟踪中目标状态的预测分布和后验分布,利用序列蒙特卡洛方法,基于多变量t-分布提出了一种新的粒子滤波算法,称之为t-分布粒子滤波器．为了根据样本估计目标状态的概率分布,提出了一种新的ECME算法,并嵌入到t-分布粒子滤波器中．理论分析表明,在t-分布条件下,t-分布粒子滤波器是在样本数量上的渐近最优估计器．在机动目标跟踪实验中,比较了t-分布粒子滤波器、无色卡尔曼滤波(Unscented Kalman filter)及自助式粒子滤波器(Bootstrap particle filters)的跟踪精度．     

一种基于AMPF和FastSLAM的复合SLAM算法

为了改进快速同时定位和地图创建(FastSLAM)算法的粒子集性能、提高估计精度,提出基于AMPF和FastSLAM的复合SLAM算法.将辅助边缘粒子滤波器(AMPF)与FastSLAM架构相结合,用AMPF估计机器人位姿,单个粒子的位姿提议分布用无轨迹卡尔曼滤波估计.设计与AMPF和FastSLAM架构均兼容的采样方法和粒子数据结构,在FastSLAM框架下用扩展卡尔曼滤波递归估计地图.实验表明,该算法的粒子集性能比FastSLAM 2.0算法好,并且它的位姿估计精度高于FastSLAM 2.0算法.此外,粒子数较少时,该算法的估计精度较高,从而可适当减少粒子数目来提高算法的计算效率.     

同时机器人定位与人体跟踪的多源感知协作粒子滤波方法

提出了分布式多传感器协作的条件粒子滤波算法以解决人与机器人位置的联合概率分布估计问题．全局视觉系统中,各视角独立运行图像平面上基于粒子滤波的目标跟踪,并利用地平面单应关系实现多视角目标主轴同步融合．视觉观测进一步与机器人激光数据以顺序滤波方式异步融合,提出包含人体位置假设的激光似然场模型以提高对机器人位姿误差的鲁棒性,并引入基于Kullback-Leibler距离的自适应采样以降低描述联合分布所需的粒子数目．实验验证了该方法能够在具有观测噪声且人—机位置均不确定的情况下利用多传感器协作实现基于地图的同时机器人定位与人体跟踪．     

基于多假设跟踪的移动机器人自适应蒙特卡罗定位研究

针对移动机器人蒙特卡罗定位(Monte Carlo localization, MCL)算法在含有对称和自相似结构的环境中容易失败的问题, 提出了一种基于多假设跟踪的自适应蒙特卡罗定位改进算法. 该算法根据粒子间空间相似性采用核密度树聚类算法对粒子群进行聚类, 每簇粒子代表一个位姿假设并用一个独立的MCL算法进行跟踪, 总体上形成了一组非等权的粒子滤波器, 很好地克服了普通粒子滤波器由于粒子贫乏而引起的过度收敛问题. 同时运用该核密度树实现了自适应采样, 提高了算法的性能. 针对机器人``绑架'问题对该算法作了进一步的改进. 实验结果证明了该算法的有效性.     

室内环境下同步定位与地图创建改进算法

提出了一种室内环境下基于平方根无迹卡尔曼滤波（SRUKF）的同步定位与地图创建（SLAM）算法． 该方法在每步迭代中采用平方根无迹粒子滤波器进行机器人状态估计,并引入平方根无迹卡尔曼滤波器定位路标, 进而完成机器人状态和相应路标信息更新．将本文算法与机器人运动模型和红外标签观测模型结合进行了仿真和实 验,结果表明,本算法在同步定位和地图创建过程中提高了机器人状态和路标估计的精度及稳定性．     

一种机器人未知环境下动态目标跟踪交互多模滤波算法

为了解决机器人同时定位、地图构建和目标跟踪问题,提出了一种基于交互多模滤波(interacting multiple model filter, IMM)的方法.该方法将机器人状态、目标状态和环境特征状态作为整体来构成系统状态向量并利用全关联扩展式卡尔曼滤波算法对系统状态进行估计,由此随着迭代估计的进行,系统各对象状态之间将产生足够的相关性,这种相关性能够正确反映各对象状态估计间的依赖关系,因此提高了目标跟踪的准确性.该方法进一步和传统的IMM滤波算法相结合,从而解决了目标运动模式未知性问题,IMM方法的采用使系统在完成目标追踪的同时还能对其运动模态进行估计,进而提高了该算法对于机动目标的跟踪能力.仿真实验验证了该方法对机器人和目标的运动轨迹以及目标运动模态进行估计的准确性和有效性.     

基于深度学习网络的机器人定位误差估计与补偿研究

为了合理补偿机器人定位误差，提升作业能力，该文提出基于深度学习网络的机器人定位误差估计与补偿方法。确定机器人定位采样点，获取机器人末端定位理论位姿，以机器人末端理论位姿作为深度神经网络输入量，机器人末端定位误差作为输出量，利用遗传粒子群算法优化权值与阈值，得到机器人定位误差估计值，并对理论位姿坐标反向迭加该误差估计值，完成定位误差补偿。实验证明，该方法能够有效补偿机器人的位移偏差和关节角度偏差，精准抓取目标物体，并在不同数量采样点条件下，可使不同类型的机器人保持较高的定位精度。     

纯方位角目标跟踪及移动平台可观性控制方法

为了解决未知环境下的单目视觉移动机器人目标跟踪问题,提出了一种将目标状态估计与机器人可观性控制相结合的机器人同时定位、地图构建与目标跟踪方法。在状态估计方面,以机器人单目视觉同时定位与地图构建为基础,设计了扩展式卡尔曼滤波框架下的目标跟踪算法;在机器人可观性控制方面,设计了基于目标协方差阵更新最大化的优化控制方法。该方法能够实现机器人在单目视觉条件下对自身状态、环境状态、目标状态的同步估计以及目标跟随。仿真和原型样机实验验证了目标状态估计和机器人控制之间的耦合关系,证明了方法的准确性和有效性,结果表明:机器人将产生螺旋状机动运动轨迹,同时,目标跟踪和机器人定位精度与机器人机动能力成正比例关系。     

基于运动特征的颜色粒子滤波器

传统颜色粒子滤波器不能对跟踪状态进行自我判断,粒子容易失效,导致目标跟偏甚至跟丢。针对该问题,设计一种基于运动特征的颜色粒子滤波器。在分析目标运动特征的基础上,改进系统运动模型,建立自适应背景颜色尺度;在跟踪过程中通过最优估计速度和加速度来判断粒子滤波器的跟踪状态,建立相应的捕获机制以跟踪目标。实验结果证明,改进的颜色粒子滤波器能对与背景相似的运动目标进行有效准确的跟踪。     

Mobile robot localization based on PSO estimator

Localization is fundamental to autonomous operation of the mobile robot. A particle filter (PF) is widely used in mobile robot localization. However, the robot localization based PF has several limitations, such as sample impoverishment and a degeneracy problem, which reduce significantly its performance. Evolutionary algorithms, and more specifically their optimization capabilities, can be used in order to overcome PF based on localization weaknesses. In this paper, mobile robot localization based on a particle swarm optimization (PSO) estimator is proposed. In the proposed method, the robot localization converts dynamic optimization to find the best robot pose estimate, recursively. Unlike the localization based on PF, the resampling step is not required in the proposed method. Moreover, it does not require noise distribution. It searches stochastically along the state space for the best robot pose estimate. The results show that the proposed method is effective in terms of accuracy, consistency, and computational cost compared with localization based on PF and EKF.     

A SLAM based on auxiliary marginalised particle filter and differential evolution

FastSLAM is a framework for simultaneous localisation and mapping (SLAM) using a Rao-Blackwellised particle filter. In FastSLAM, particle filter is used for the robot pose (position and orientation) estimation, and parametric filter (i.e. EKF and UKF) is used for the feature location's estimation. However, in the long term, FastSLAM is an inconsistent algorithm. In this paper, a new approach to SLAM based on hybrid auxiliary marginalised particle filter and differential evolution (DE) is proposed. In the proposed algorithm, the robot pose is estimated based on auxiliary marginal particle filter that operates directly on the marginal distribution, and hence avoids performing importance sampling on a space of growing dimension. In addition, static map is considered as a set of parameters that are learned using DE. Compared to other algorithms, the proposed algorithm can improve consistency for longer time periods and also, improve the estimation accuracy. Simulations and experimental results indicate that the proposed algorithm is effective.     

A Robust Regression Model for Simultaneous Localization and Mapping in Autonomous Mobile Robot

Segment-based maps as sub-class of feature-based mapping have been widely applied in simultaneous localization and map building (SLAM) in autonomous mobile robots. In this paper, a robust regression model is proposed for segment extraction in static and dynamic environments. We adopt the MM-estimate to consider the noise of sensor data and the outliers that correspond to dynamic objects such as the people in motion. MM-estimates are interesting as they combine high efficiency and high breakdown point in a simple and intuitive way. Under the usual regularity conditions, including symmetric distribution of the errors, these estimates are strongly consistent and asymptotically normal. This robust regression technique is integrated with the extended Kalman filter (EKF) to build a consistent and globally accurate map. The EKF is used to estimate the pose of the robot and state of the segment feature. The underpinning experimental results that have been carried out in static and dynamic environments illustrate the performance of the proposed segment extraction method.     

基于近红外视觉的机器人室外定位系统

在光照条件可变且存在电磁干扰的环境下,针对机器人室外导航任务,提出了一种基于全景近红外视 觉和编码路标的自定位系统．通过近红外光源照明,利用全景视觉识别采用条形编码格式的路标,并利用扩展卡尔 曼滤波算法（EKF）融合视觉数据和里程计数据,从而实现机器人自定位．实验证明,该方法消除了室外大范围导航 时光照变化对机器人定位结果的影响．     

Convergence and Consistency Analysis for Extended Kalman Filter Based SLAM

This paper investigates the convergence properties and consistency of extended Kalman filter (EKF) based simultaneous localization and mapping (SLAM) algorithms. Proofs of convergence are provided for the nonlinear two-dimensional SLAM problem with point landmarks observed using a range-and-bearing sensor. It is shown that the robot orientation uncertainty at the instant when landmarks are first observed has a significant effect on the limit and/or the lower bound of the uncertainties of the landmark position estimates. This paper also provides some insights to the inconsistencies of EKF based SLAM that have been recently observed. The fundamental cause of EKF SLAM inconsistency for two basic scenarios are clearly stated and associated theoretical proofs are provided.     

基于扫描匹配的室外环境SLAM 方法

针对室外环境中的机器人“绑架”问题,提出了基于地图匹配的SLAM方法．该方法舍弃了机器人里程计信息, 只利用局部地图和全局地图的图形相关性进行机器人定位．方法的核心是多重估计数据关联,并将奇异值分解应用到机器人位姿计算中．利用Victoria Park数据集将本算法与基于扩展卡尔曼滤波器的方法进行比较,实验结果证明了本文提出的算法的有效性．     

Mobile-robot pose estimation and environment mapping using an extended Kalman filter

In this paper an extended Kalman filter (EKF) is used in the simultaneous localisation and mapping (SLAM) of a four-wheeled mobile robot in an indoor environment. The robot’s pose and environment map are estimated from incremental encoders and from laser-range-finder (LRF) sensor readings. The map of the environment consists of line segments, which are estimated from the LRF’s scans. A good state convergence of the EKF is obtained using the proposed methods for the input- and output-noise covariance matrices’ estimation. The output-noise covariance matrix, consisting of the observed-line-features’ covariances, is estimated from the LRF’s measurements using the least-squares method. The experimental results from the localisation and SLAM experiments in the indoor environment show the applicability of the proposed approach. The main paper contribution is the improvement of the SLAM algorithm convergence due to the noise covariance matrices’ estimation.     

Identification of process and measurement noise covariance for state and parameter estimation using extended Kalman filter

The performance of Bayesian state estimators, such as the extended Kalman filter (EKF), is dependent on the accurate characterisation of the uncertainties in the state dynamics and in the measurements. The parameters of the noise densities associated with these uncertainties are, however, often treated as ‘tuning parameters’ and adjusted in an ad hoc manner while carrying out state and parameter estimation. In this work, two approaches are developed for constructing the maximum likelihood estimates (MLE) of the state and measurement noise covariance matrices from operating input-output data when the states and/or parameters are estimated using the EKF. The unmeasured disturbances affecting the process are either modelled as unstructured noise affecting all the states or as structured noise entering the process predominantly through known, but unmeasured inputs. The first approach is based on direct optimisation of the ML objective function constructed by using the innovation sequence generated from the EKF. The second approach - the extended EM algorithm - is a derivative-free method, that uses the joint likelihood function of the complete data, i.e. states and measurements, to compute the next iterate of the decision variables for the optimisation problem. The efficacy of the proposed approaches is demonstrated on a benchmark continuous fermenter system. The simulation results reveal that both the proposed approaches generate fairly accurate estimates of the noise covariances. Experimental studies on a benchmark laboratory scale heater-mixer setup demonstrate a marked improvement in the predictions of the EKF that uses the covariance estimates obtained from the proposed approaches.     

Robot Pose Estimation in Unknown Environments by Matching 2D Range Scans

A mobile robot exploring an unknown environment has no absolute frame of reference for its position, other than features it detects through its sensors. Using distinguishable landmarks is one possible approach, but it requires solving the object recognition problem. In particular, when the robot uses two-dimensional laser range scans for localization, it is difficult to accurately detect and localize landmarks in the environment (such as corners and occlusions) from the range scans.In this paper, we develop two new iterative algorithms to register a range scan to a previous scan so as to compute relative robot positions in an unknown environment, that avoid the above problems. The first algorithm is based on matching data points with tangent directions in two scans and minimizing a distance function in order to solve the displacement between the scans. The second algorithm establishes correspondences between points in the two scans and then solves the point-to-point least-squares problem to compute the relative pose of the two scans. Our methods work in curved environments and can handle partial occlusions by rejecting outliers.