基于病态参数分离的机器人运动学标定测量构型优化

针对一种高灵巧性机器人及其连杆参数高敏感性与高定位精度需求,为解决机器人运动学标定随机测量构型存在绝对 定位精度低、参数辨识效果及标定结果鲁棒性较差的问题,提出一种病态参数分离与 DETMAX-改进差分进化(DETMAX-IDE) 算法的机器人运动学标定测量构型分步优化方法。 首先,建立机器人位置误差模型。 其次,建立一种可观性综合指标,评价不 同机器人标定测量构型的总体可观测性和灵敏度。 最后,分离机器人运动学位置误差模型的病态参数,建立测量构型优化目标 函数和约束条件,提出一种基于 DETMAX 算法与改进差分进化算法结合的分步迭代优化算法(简称为 DETMAX-改进差分进化 算法,简写为 DETMAX-IDE 算法),开展机器人运动学标定测量构型分步迭代优化。 通过机器人运动学标定仿真与实验,验证 了所提方法的有效性。 实验结果表明,与随机测量构型相比,所提方法对应的机器人绝对定位精度的平均值和均方差分别降低 了 62. 09% 和 62. 45% 。     

Optimization design and dynamic stability analysis of 3-DOF cable-driven parallel robot with an elastic telescopic rod

This paper proposes a three degree-of-freedoms (DOFs) cable-driven parallel robot (CDPR), which is actuated by three groups of parallel cables and tensioned by an elastic telescopic rod with a passive spring. Firstly, the architecture of the robot is briefly illustrated with emphasis on the three parallelogram arrangements of the cables and composition of elastic telescopic rod structure. This robot has larger workspace and greater tension than previous under-constrained CDPRs. Secondly, the kinematic and dynamic models of the robot are established. On this basis, the parameters of the spring installed in the telescopic rod are optimized and determined considering both the acceleration and cable force through using the differential evolution (DE) algorithm. Thirdly, dynamic stability analysis of the robot under impulsive disturbances is performed according to the Gauss principle of least constraint (GPLC). The results of a simulation case show that this robot has better dynamic stability in comparison with conventional under-constraint CDPR because of the presence of the elastic telescopic rod. The measurement results in the workspace are obtained through dynamic simulation. Finally, the experiments are performed based on numerical simulation. The feasibility of the CDPR is verified via the experiments and simulations.

     

基于遗传神经网络的机器人腕力传感器动态建模与补偿方法

介绍用于MotomamV3X机器人上的新型多维腕力传感器,比较遗传算法与人工神经网络的特点,将遗传算法的交叉和变异操作进行改进,提出一种融合改进遗传算法(Genetic algorithm, GA)的函数连接型人工神经网络(Functional link artificial neural network, FLANN),并将其用于所介绍的新型机器人腕力传感器动态建模与动态性能补偿中。介绍动态建模与动态补偿原理及改进遗传神经网络算法,给出该传感器的动态模型和动态补偿模型。该方法利用腕力传感器的动态标定数据,采用改进遗传神经网络搜索和优化模型参数,保留了遗传算法的全局搜索能力和FLANN结构简单,鲁棒性好,且具备自学习能力的特点,克服了FLANN容易陷入局部极小的缺陷,具有快的网络训练速度及高的动态建模精度。理论分析和试验结果都证实了所提出的动态建模与动态补偿方法的有效性。     

基于任务的可重构模块机器人构形确定方法

可重构机器人为柔性加工系统提供了更多的选择,但复杂多变的任务、环境以及机器人模块无穷组合,为机器人构形的确定带来了很大的难题。采用遗传算法和迭代算法分两级搜索对机器人构形组合进行优化设计:运用遗传算法对机器人构形进行进化设计,满足工作空间的可达性、环境避障、线性和角度误差、末端操作灵巧度、运动关节的动力学要求、期望模块数目等性能指标要求。运用迭代算法对构形进行运动学逆解求解,计算空间工作点的可达性适应度。根据可装配的模块机器人及模块化思想,建立各个模块的速度、加速度等动力学表达式及相关修正公式,采用补偿迭代法来自动生成动力学方程,计算运动模块驱动性能的适应度。用专家经验对搜索得到适应度较高的构形进行修正,得到满足工作要求的机器人构形。最后给出实例验证。     

SK6弧焊机器人运动学正解问题研究

根据SK6弧焊机器人的杆件参数,建立了SK6弧焊机器人的运动学方程;考虑焊枪结构,首次建立了SK6弧焊机器人焊枪末端的运动学方程.该问题的解决,对于SK6弧焊机器人离线编程系统中机器人的仿真是非常关键的,对研究其它机器人也有一定的参考意义.     

基于混合群智能优化的机器人立体视觉标定

准确的立体视觉模型是机器人高精密视觉定位的基础,而传统的单一非线性优化算法难以实现稳定和高精度的机器人立体视觉标定。结合遗传算法全局搜索能力强和粒子群算法局部搜索能力强的特点,提出了一种基于混合群智能优化的机器人立体视觉三步标定方法。针对非线性视觉模型,标定第一步和第二步分别对两个摄像机模型单独作线性初值求解和初次非线性优化,第三步对双目立体视觉模型作联合非线性优化,直接线性变换、遗传算法、粒子群算法分别作用于标定的三个步骤,每一步计算的结果被用作下一步的初始化。仿真试验分析与实际试验结果表明,相对于传统的优化标定方法和使用单一群智能优化算法的标定方法,该方法在噪声环境下具有更高的准确性和鲁棒性,能够更好满足机器人精密视觉操作的需求。     

Local POE model for robot kinematic calibration

A robot kinematic calibration method based on the local frame representation of the product-of-exponentials (Local POE) formula is introduced. In this method, the twist coordinates of the joint axes are expressed in their respective local (body) frames. The advantages of this new approach are threefolds: (1) revolute and prismatic joints can be uniformly expressed in the twist coordinates based on the line geometry; (2) the twist coordinates of the joint axes can be set up with simple values because the local frames can be arbitrarily defined on the links; (3) the kinematic parameters described by the twist coordinates vary smoothly that makes the method robust and singularity-free. By assuming that the kinematic errors exist only in the relative initial poses of the consecutive link frames, the kinematic calibration models can be formulated in a simple and elegant way. The calibration process then becomes to re-define a set of new local link frames that are able to reflect the actual kinematics of the robot. This method can be applied to robot manipulators with generic open chain structures (serial or tree-typed). The simulation and experiment results on a 4-DOF SCARA type robot and a 5-DOF tree-typed modular robot have shown that the average positioning accuracy of the end-effector increases significantly after calibration.     

MICRAS,Microcomputer interactive codes for robot analysis and simulationMICRAS: Logiciel interactif sur micro-ordinateur pour l'analyse et la simulation des robots

With the growth and acceptance of interactive computer graphics, contemporary robotic research trends are focusing on the development of robot simulation packages. This paper describes a new microcomputer interactive package for the analysis and simulation of industrial robots. The package represents an extension of the state-of-the-art developments in the technology of industrial robots. The novel features are derived from the use of the microcomputer for analysis and simulation. The package includes all aspects of the kinematic analysis of the six general geometric structures of industrial robots.MICRAS is a package of microcomputer interactive codes for robot analysis and simulation. The program WORKSPACE is used for obtaining the value and plot of the workspace of two- and three-link revolute-jointed arms. DIRKIN is a program that generates the forward solution for any robot with up to six degrees of freedom. The data input section provides the robot geometric parameters and joint angles interactively. The program REVKIN is constructed to give the inverse solution of the six major kinematic structures of robots and display all possible configurations for each robot. COJOINT is a program that implements coordinated joint motion of a robot with defined range of motion of each joint. The modules with prefix CAT constitute the program for the simulation of robot movements for a specified cartesian path of the end effector. Finally, DIRDIF is a program for the differential kinematic analysis of robots.     

Pantograph mechanism as a non-traditional manipulator structureMecanisme a pantographe comme structure non traditionnelle de manipulateur robotique

This paper presents a study on the design, motion and force transmission characteristics of a non-traditional robot structure, the pantograph mechanism. Although the pantograph mechanisms has been known for about four hundred years, the comprehensive study of its potential advantages and limitations when used as a robot is believed to be new. In this paper, the kinematic linearity and dynamic nonlinearity between inputs and outputs are investigated. It is also proven that the two rectilinear input motions which exist in a pantograph structure are not coupled, and therefore are suitable for being adopted as generalized coordinates in control. In addition, a set of physically meaningful dimensionless parameters has been identified. Based on these parameters, kinematic properties for obtaining optimum motion range and workspace are suggested.     

Optimal design of a 3-leg 6-DOF parallel manipulator for a specific workspace

Researchers seldom study optimum design of a six-degree-of-freedom(DOF) parallel manipulator with three legs based upon the given workspace. An optimal design method of a novel three-leg six-DOF parallel manipulator(TLPM) is presented. The mechanical structure of this robot is introduced, with this structure the kinematic constrain equations is decoupled. Analytical solutions of the forward kinematics are worked out, one configuration of this robot, including position and orientation of the end-effector are graphically displayed. Then, on the basis of several extreme positions of the kinematic performances, the task workspace is given. An algorithm of optimal designing is introduced to find the smallest dimensional parameters of the proposed robot. Examples illustrate the design results, and a design stability index is introduced, which ensures that the robot remains a safe distance from the boundary of sits actual workspace. Finally, one prototype of the robot is developed based on this method. This method can easily find appropriate kinematic parameters that can size a robot having the smallest workspace enclosing a predefined task workspace. It improves the design efficiency, ensures that the robot has a small mechanical size possesses a large given workspace volume, and meets the lightweight design requirements.     

MENDED GENETIC BP NETWORK AND APPLICATION TO ROLLING FORCE PREDICTION OF 4-STAND TANDEM COLD STRIP MILL

In order to make good use of the ability to approach any function of BP (back propagation)network and overcome its local astringency, and also make good use of the overall search ability of GA(genetic algorithms), a proposal to regulate the network's weights using both GA and BP algorithms issuggested. An integrated network system of MGA (mended genetic algorithms) and BP algorithms hasbeen established. The MGA-BP network's functions consist of optimizing GA performance parameters,the network's structural parameters, performance parameter, and regulating the network's weightsusing both GA and BP algorithms. Rolling forces of 4-stand tandem cold strip mill are predicted by theMGA-BP network, and good results are obtained.     

基于起跳稳定性的仿蝗虫八杆跳跃机器人设计

通过模仿蝗虫起跳过程中的后足胫节末端的直线运动轨迹和起跳稳定性，提出了一种仿蝗虫八杆跳跃机器人设计方法。建立运动学模型并利用天牛须搜索算法对机构的尺寸参数进行优化设计，优化后的八杆跳跃机构的等效胫节末端轨迹和蝗虫后足胫节起跳过程的末端轨迹吻合，基本为一条斜直线。基于拉格朗日方程对八杆跳跃机构的起跳动力学进行建模，并分析了身体所在杆的质心位置对起跳过程的影响，发现通过增大等效跗节杆的质量能提高起跳动力学稳定性。根据运动学和动力学分析的结果，设计了仿蝗虫八杆跳跃机器人，并利用高速摄像机搭建实验平台记录其起跳过程，验证了等效跗节杆的质量对起跳动力学稳定性具备有效的调节效果。     

Kinematic parameter calibration method for industrial robot manipulator using the relative position

A new calibration method for industrial robot system calibration on a manufacturing floor is presented in this paper. To calibrate the robot system, a laser sensor to measure the distance between robot tool and measurement surface is attached to the robot end-effector and a grid is established in the floor. Given two position command pulses for a robot manipulator and using the position difference between two command pulses, the relative position measurement calibration method will find the real robot kinematic parameters. The procedures developed have been applied to an industrial robot. Finally, the effects of the models used to calibrate the robot are discussed. This calibration method represents an effective, low cost and feasible technique for the industrial robot calibration in lab. projects and industrial environments.     

Laser tracker-based kinematic parameter calibration of industrial robots by improved CPA method and active retroreflector

Typical approaches to calibrate industrial robots are based on open- and closed-loop methods; the screw–axis measurement methods traditionally receive much less attention. Although the identification process does not guarantee the physical–mathematical link between the robot parameters in the first two groups of techniques, these techniques are generally more effective in reducing the global positioning error compared to the screw–axis methods. The third group of techniques acquires parameters based on the physical reality of the robot, effectively keeping the physical–mathematical link. This group is considered more appropriate than the previous two groups; however, it cannot reduce the overall error when considering the entire workspace of the robot compared to the previous groups. This paper presents a new technique to identify the kinematic parameters of an industrial robot based on a combination of techniques from the aforementioned categories. This new data acquisition technique uses a laser tracker with an active target, which maximises the angle covered by each joint and greatly simplifies the screw–axis measurement process. An identification procedure based on circle point analysis is also proposed, and the procedure evaluates the technique by obtaining initial values with a new formulation of the objective function of error based on mutual distances between the points captured in screw–axis measurements. This type of measurement also allows the eccentricity and backlash of each joint to be characterised independently such that local joint corrections could be made in combination with the identified parameters.     

基于球杆仪检测信息的并联机构运动学标定

由于并联构型装备难于实现全闭环反馈控制,使运动学标定成为一项具有显著经济价值并能非常有效提高并联构型装备精度的手段,运动学标定通常包括误差建模、测量、辨识和补偿4个环节。基于以上因素,以5自由度混联机械手TriVariant为对象,研究一种基于球杆仪检测信息的运动学标定方法。首先建立球杆仪测量值与影响末端可补偿位姿误差的几何误差源的映射关系,并给出可辨识条件。在此基础上,以误差参数辨识矩阵条件数为评价指标,探讨合理设置球杆仪安装位置和数目的方法。最后,计算机仿真和试验验证了所提出方法的可行性和有效性,并指出仍然需要解决的若干问题。     

虚拟封闭运动链法提高机器人运动学标定精度

介绍了一种基于激光器构造虚拟封闭运动链的标定方法,该方法通过末端操作器上安装的激光器向观测平板发射激光束,并保持激光投影点处于观测平板上一个固定位置,构造出一个虚拟封闭运动链。在此基础上对IRB140机器人的几何参数进行了标定,经试验验证表明,本标定方法有效地实现了机器人绝对精度的大幅度提高。     

复合运动模式四足机器人机构设计及分析

研制了一种轮足复合运动的四足智能移动机器人,该机器人可以步行前进,原地转弯,楼梯爬越,也可在良好路面利用足底轮以较高速度滚动前进。阐述了机器人的机械结构和参数,运动学分析和运动空间描述。利用ADAMS建立了三维仿真模型,进行了多种步态的仿真。得出稳定裕度合适的步态,实验验证了机器人的性能。     

仿人型跑步机器人矢状面起跳运动的实现

针对仿人型跑步机器人实现起跳动作时没有考虑腿部质量或将机器人质心固定在身体上某点的不足,提出了一种新的方法实现跑步机器人矢状面内的起跳动作:利用“虚拟腿”的概念,求出机器人质心的轨迹,然后对某些广义坐标进行规划,通过求解非线性方程组计算出机器人在每个时刻的运动学参数。最后根据动力学方程求出各个关节的驱动力矩。仿真结果表明:机器人跑步时各个关节角度和关节驱动力矩变化平稳,因此起跳动作设计合理。