基于粒子滤波的智能机器人定位算法

自主定位是智能机器人的关键性技术。针对轮式智能机器人在使用里程计、激光雷达进行定位过程中存在较大误差的问题,联合双目摄像机和激光雷达数据,提出基于粒子滤波的自适应蒙特卡洛(AMCL)优化定位算法。预测阶段,利用双目摄像机和激光雷达数据改善提议分布,减少滤波过程中重采样的粒子数,用更少的粒子数来估计机器人的后验概率分布。在激光雷达匹配点云时,提出一种分组阶梯式阈值判断法,在不降低点云匹配效果的情况下,有效降低现有的迭代最近点(ICP)匹配算法的计算量。为了验证改进算法的性能,在四轮智能机器人平台上进行实验。结果表明:改进的AMCL优化定位算法可以有效提高机器人的定位精度,具有较好的实用性。     

基于粒子滤波重采样与变异操作的改进粒子群算法

针对标准粒子群优化(PSO)算法在求解过程中存在求解精度低、搜索后期收敛速度慢等问题,提出一种基于粒子滤波重采样步骤与变异操作相结合的改进PSO算法——RSPSO。该算法充分利用重采样中具有较大权值的粒子被保留和复制、较小权值的粒子被舍弃的特点,并利用已有的变异操作方法克服粒子匮乏的缺点,大大增强了PSO算法中后期搜索阶段的局部搜索能力。在不同基准函数下对RSPSO算法和标准PSO算法以及文献中其他改进算法进行对比。实验结果表明, RSPSO算法的收敛速度较快,同时其搜索精度和解的稳定性均有所提高,且能够全局地解决多峰问题。     

A novel efficient algorithm for mobile robot localization

Being autonomous is one of the most important goals in mobile robots. One of the fundamental works to achieve this goal is giving the ability to a robot for finding its own correct position and orientation. Different methods have been introduced to solve this problem. In this paper, a novel method based on the harmony search (HS) algorithm for robot localization through scan matching is proposed. Simulation results show that the proposed method in comparison with a genetic algorithm-based approach has better accuracy and higher performance. Furthermore a new hybrid algorithm based on harmony search and differential evolution (DE) algorithms is proposed and evaluated on different benchmark functions. Finally the hybrid algorithm has been applied for mobile robot localization and it outperformed the HS-based approach.     

Particle Filter for Collaborative Multi-Robot Localization Tolerant to Recognition Error

Statistical algorithms using particle filters have been proposed previously for collaborative multi-robot localization. In these algorithms, by synchronizing each robot's belief or exchanging the particles of the robots, fast and accurate localization is attained. However, there algorithms assume correct recognition of other robots and the effects of recognition error are not considered. If the recognition of other robots is incorrect, a large amount of error in localization can occur. This paper describes this problem. Furthermore, in order to cope with the problem, an algorithm for collaborative multi-robot localization is proposed. In the proposed algorithm, the particles of a robot are exchanged with those of other robots according to measurement results obtained by the sending robot. At the same time, some particles remain in the sending robot. Received particles from other robots are evaluated using measurement results obtained by the receiving robot. The proposed method copes with recognition error by using the remaining particles, and increases the accuracy of estimation by twice evaluating the exchanged particles of the sending and receiving robots. These properties of the proposed method are argued mathematically. Simulation results show that incorrect recognition of other robots does not cause serious problems in the proposed method.     

腰部外骨骼机器人线性自抗扰控制器参数优化

针对腰部外骨骼机器人线性自抗扰控制器参数难以调整的问题,本文提出一种基于天牛须搜索的改进粒子群算法(PSO)。建立腰部外骨骼机器人模型,采用线性自抗扰控制器,进一步引入改进的PSO对其进行参数优化。该算法通过混沌初始化种群,提高粒子执行效率;采用非线性策略调整惯性因子和学习因子,加强粒子的搜索能力;引入天牛须搜索算法与PSO结合,并采用自适应权重,使得粒子可对周边环境进行较好地判断,避免粒子陷入局部最优。分别通过6个测试函数和建立系统评价指标进行仿真实验,结果表明所提出的算法有更好的收敛精度,优化后的控制器具有更好的控制性能。     

A novel kinematic parameters calibration method for industrial robot based on Levenberg-Marquardt and Differential Evolution hybrid algorithm

The poor absolute positioning accuracy of industrial robots is the main obstacle for its further application in precision grinding of complex surfaces, such as blisk, blade, etc. Based on the established kinematic error model of a typical industrial robot FANUC M710ic/50, a novel kinematic parameters calibration method is proposed in this paper to improve the absolute positioning accuracy of robot. The pre-identification of the kinematic parameter deviations of robot was achieved by using the Levenberg-Marquardt algorithm. Subsequently, these identified suboptimal values of parameter deviations were defined as central values of the components of initial individuals to complete accurate identification by using Differential Evolution algorithm. The above two steps, which were regarded as the core of this Levenberg-Marquardt and Differential Evolution hybrid algorithm, were used to obtain the preferable values for kinematic parameters of the robot. On this basis, the experimental investigations of kinematic parameters calibration were conducted by using a laser tracker and numerical simulation method. The results revealed that the robot positioning error decreased from 0.994 mm, initial positioning error measured by laser tracker, to 0.262 mm after calibration with this proposed hybrid algorithm. The absolute positioning accuracy has increased by 40.86% than that of the Levenberg-Marquardt algorithm, increased by 40.31% than that of the Differential Evolution algorithm, and increased by 25.14% than that of the Simulated Annealing algorithm. This work shows that the proposed kinematic parameters calibration method has a significant improvement on the absolute positioning accuracy of industrial robot.     

A collaborative localization tolerant to recognition error by double-check particle exchange

Statistical algorithms using particle filters for collaborative multi-robot localization have been proposed. In these algorithms, by synchronizing every robot’s belief or exchanging particles of the robots with each other, fast and accurate localization is attained. These algorithms assume correct recognition of other robots, and the effects of recognition errors are not discussed. However, if the recognition of other robots is incorrect, a large amount of error in localization can occur. This article describes this problem. Furthermore, an algorithm for collaborative multi-robot localization is proposed in order to cope with this problem. In the proposed algorithm, the particles of a robot are sent to other robots according to measurement results obtained by the sending robot. At the same time, some particles remain in the sending robot. Particles received from other robots are evaluated using measurement results obtained by the receiving robot. The proposed method is tolerant to recognition error by the remaining particles and evaluating the exchanged particles in the sending and receiving robots twice, and if there is no recognition error, the proposed method increases the accuracy of the estimation by these two evaluations. These properties of the proposed method are argued mathematically. Simulation results show that incorrect recognition of other robots does not cause serious problems in the proposed method.     

FastSLAM算法的仿生优化改进研究

针对机器人导航标准的快速同步定位与地图构建算法(FastSLAM)在重采样过程中存在采样粒子集的贫化以及粒子多样性的缺失导致机器人的定位与建图的精度下降的问题,提出一种基于改进的蝴蝶算法来优化FastSLAM中的粒子滤波部分。改进的算法将机器人的最新时刻的观测和状态信息融入到蝴蝶算法的香味公式中,并在蝴蝶位置更新的过程加入自适应香味半径和自适应蝴蝶飞行调整步长因子,来减少算法的运算时间以及提高预测精度,同时引入偏差修正指数加权算法对粒子的权值进行优化组合,对组合后部分不稳定的粒子进行分布重采样,保证粒子的多样性。通过仿真验证了该算法在估计精度与稳定性方面优于FastSLAM,因此在移动机器人运动模型的定位与建图中具有较高的定位精度与稳定性。     

Unscented FastSLAM: A Robust and Efficient Solution to the SLAM Problem

The Rao--Blackwellized particle filter (RBPF) and FastSLAM have two important limitations, which are the derivation of the Jacobian matrices and the linear approximations of nonlinear functions. These can make the filter inconsistent. Another challenge is to reduce the number of particles while maintaining the estimation accuracy. This paper provides a robust new algorithm based on the scaled unscented transformation called unscented FastSLAM (UFastSLAM). It overcomes the important drawbacks of the previous frameworks by directly using nonlinear relations. This approach improves the filter consistency and state estimation accuracy, and requires smaller number of particles than the FastSLAM approach. Simulation results in large-scale environments and experimental results with a benchmark dataset are presented, demonstrating the superiority of the UFastSLAM algorithm.      

基于不同重采样算法的RFID指纹定位

为满足移动机器人精确定位的需求,提出一种基于不同重采样算法的粒子滤波指纹定位法。定位阶段首先利用机器人运动学建立运动模型作为粒子预测分布, 并将当前的观测信息和环境指纹融入, 以改善滤波效果, 减少所需粒子数;然后给出精致重采样(ER)算法,以提高粒子的细化能力,减少粒子匮乏效应并提高定位精度;最后分析不同重采样算法对定位精度的影响,且从不同的实验角度进一步验证定位算法的精确性以及可靠性。实验结果表明, 该算法在定位精度和鲁棒性方面都有显著提高。     

基于VSLAM的自主移动机器人三维同时定位与地图构建

移动机器人在探索未知环境且没有外部参考系统的情况下,面临着同时定位和地图构建（SLAM）问题。针对基于特征的视觉SLAM（VSLAM）算法构建的稀疏地图不利于机器人应用的问题,提出一种基于八叉树结构的高效、紧凑的地图构建算法。首先,根据关键帧的位姿和深度数据,构建图像对应场景的点云地图;然后利用八叉树地图技术进行处理,构建出了适合于机器人应用的地图。将所提算法同RGB-D SLAM（RGB-Depth SLAM）算法、ElasticFusion算法和ORB-SLAM（Oriented FAST and Rotated BRIEF SLAM）算法通过权威数据集进行了对比实验,实验结果表明,所提算法具有较高的有效性、精度和鲁棒性。最后,搭建了自主移动机器人,将改进的VSLAM系统应用到移动机器人中,能够实时地完成自主避障和三维地图构建,解决稀疏地图无法用于避障和导航的问题。     

改进的均值漂移和粒子滤波混合跟踪方法

为提高粒子滤波视觉目标跟踪算法的准确性和实时性,提出一种基于均值漂移和粒子滤波的混合跟踪算法。将相异性较小的粒子进行聚类,利用均值漂移算法迭代各个聚类中的代表点,通过减少参与均值漂移迭代的粒子数来降低运算复杂度;根据跟踪情况自适应调整采样粒子数目和过程噪声分布,以提高跟踪精度和减少运算时间。实验结果表明,所提算法平均每帧计算时间不到传统混合跟踪法的一半,而且跟踪精度也有所提高。     

基于激光扫描测距的机器人粒子滤波定位技术研究

传统的粒子滤波即时定位与地图构建(SLAM)算法在构建地图和目标进行自主定位时,粒子数量大,占用的内存高,重采样之后容易出现粒子匮乏现象,为了提高机器人自主定位的效率,提出了一种改进的重采样策略和粒子更新策略,融入系统模型.在装有机器人操作系统(ROS)的旅行家移动机器人上进行测试,实验结果表明:方法能够有效提升粒子滤波定位的效率.     

层次环形拓扑结构的动态粒子群算法

粒子群算法（PSO）的拓扑结构决定粒子之间的信息交互方式,是影响算法性能的关键因素。为提高算法性能,提出了一种层次环形拓扑结构的动态粒子群算法（HRPSO）,粒子组成的环被分配在规则树中,算法运行时,环在层次中动态移动。通过6个标准测试函数优化,比较了HRPSO与几种基准算法的性能,实验结果证明HRPSO在精确性和稳定性上具有优势。     

基于高斯扰动的粒子群优化算法

针对标准粒子群优化(PSO)算法易陷入局部最优、进化后期收敛速度慢和收敛精度低的缺点,提出一种基于高斯扰动的粒子群优化算法。该算法采用对粒子个体最优位置加入高斯扰动策略,有效地防止算法陷入局部最优,加快收敛并提高收敛精度。在固定评估次数的情况下,对8个常用的经典基准测试函数在30维上进行了仿真。实验结果表明,所提算法在收敛速度和寻优精度上优于一些知名的粒子群优化算法。     

一种粒子群算法的多样性策略研究

本文提出了一种粒子群算法的多样性策略,即在搜索过程中,对部分适应值较差的粒子重新进行随机初始化。修改后的算法经过了大量测试函数上的模拟实验验证,并与其他已有算法进行了比较。实验结果表明,该算法能获得更高的收敛成功率和质量更好的解。在困难的多峰函数优化上具有很强的竞争力。     

粒子群优化的移动机器人路径规划算法

针对移动机器人在复杂环境下采用传统方法路径规划收敛速度慢和局部最优问题,提出了斥力场下粒子群优化（PSO）的移动机器人路径规划算法。首先采用栅格法对机器人的移动路径进行初步规划,并将栅格法得到的初步路径作为粒子的初始种群,根据障碍物的不同形状和尺寸以及障碍物所占的地图总面积确定栅格粒度的大小,进而对规划路径进行数学建模;然后根据粒子之间的相互协作实现对粒子位置和速度的不断更新;最后采用障碍物斥力势场构造高安全性适应度函数,从而得到一条机器人从初始位置到目标的最优路径。利用Matlab平台对所提算法进行仿真,结果表明,该算法可以实现复杂环境下路径寻优和安全避障;同时还通过对比实验验证了算法收敛速度快,能解决局部最优问题。     

基于复合前馈模糊PID的位置伺服系统研究

在现代伺服控制系统中,PID控制在响应速度和位置跟踪精度方面存在不足。针对此种不足,提出了一种基于复合前馈模糊PID的控制算法。同时,使用该算法建立了控制系统的数学模型和策略,并使用Matlab和TMS320F28379D控制器搭建了系统仿真模型进行实验验证。实验结果表明,该复合模糊控制算法具有快速响应、准确、无超调等特性,满足机器人等伺服控制场合需要。     

基于双种群交叉学习的粒子群优化算法

粒子群优化算法因其支配参数少、收敛速度快、易于实现等特点被广泛应用,但是粒子群优化算法存在精度低、容易陷入局部优化的问题。为此提出一种基于双种群交叉学习的粒子群优化算法。在该算法中,整个种群被分为普通子种群和精英子种群。普通子种群采用综合变异机制,该机制通过设置概率参数使普通子种群随机选择朝着优秀粒子的方向或者保持自身方向进行变异,以侧重寻找可能解区域。精英子种群则采用交叉学习机制,将粒子的历史最优和全局最优个体进行交叉生成范例,从而引导粒子对可能解区域进行局部搜索,还提出了一种非线性惯性权重来平衡粒子的全局勘探和局部开发能力。为了验证算法的有效性,在十六个基准问题上进行测试并与其他七种粒子群优化算法变体比较,实验结果表明该算法在求解精度和收敛速度总体排名第一,验证了该算法求解性能优于其他粒子群优化算法变体。     

基于改进支持向量回归算法的移动机器人定位

为了提高移动机器人定位精度,提出了一种基于正交编码器和陀螺仪的轮式移动机器人定位系统,建立机器人的定位模型和运动学模型。研究了支持向量回归（SVR）算法,为获得更好的鲁棒性,对目标函数误差平方进行加权,分析不同参数优化算法对支持向量机回归准确率的影响。以自制的移动机器人为实验平台,将改进的算法与最小二乘支持向量回归（LSSVR）算法、加权最小二乘支持向量回归（WLSSVR）算法进行比较,对比了用改进算法时机器人在木地板场地与瓷砖场地的定位误差情况,并对正交编码器+陀螺仪定位系统与双码盘定位系统、单码盘+陀螺仪定位系统进行比较。实验结果表明,改进的算法使机器人的定位精度明显高于对比算法,并且所提出的定位系统定位效果较好。