软体机械臂水下自适应鲁棒视觉伺服

水下仿生软体机器人在水底环境勘测, 水下生物观测等方面具有极高的应用价值. 为进一步提升仿章鱼臂软体机器人在特殊水下环境中控制效果, 提出一种自适应鲁棒视觉伺服控制方法, 实现其在干扰无标定环境中的高精度镇定控制. 基于水底动力学模型, 设计保证动力学稳定的控制器; 针对柔性材料离线标定过程繁琐、成本高, 提出材料参数自适应估计算法; 针对水下特殊工作条件, 设计自适应鲁棒视觉伺服控制器, 实现折射效应的在线补偿, 并通过自适应未知环境干扰上界, 避免先验环境信息的求解. 所提算法在软体机器人样机中验证其镇定控制性能, 为仿生软体机器人的实际应用提供理论基础.     

基于2D三焦点张量的移动机器人视觉伺服镇定控制

针对单目视觉移动机器人系统, 本文提出了一种基于二维三焦点张量(2D trifocal tensor, 2DTT)的视觉伺服镇定控制方法. 具体而言, 首先描述了2D三焦点张量的导出过程, 并给出了基于图像特征点的估计方法. 在此基础上根据2D三焦点张量的元素, 设计了一种反馈线性化控制器以实现机器人的位置镇定, 以及一种比例控制器来实现姿态镇定, 因而在场景信息与平移信息均未知情况下完成了移动机器人的视觉镇定控制. 通过理论分析证明了本文设计的镇定控制算法具有指数收敛性能. 相比现有方法, 这种基于2D 三焦点张量的方法在图像特征识别方面具有更强的鲁棒性, 并且在平面场景与立体场景情况下均适用. 最后利用仿真与实验结果验证了本文提出的视觉伺服方法的优良性能.     

Model-less visual servoing using modified simplex optimization

This article describes a robot positioning task with respect to a static target by visual servoing. The vision system is uncalibrated, and the kinematic model of the robot may be totally unknown. The displacements of the robot at joint level are generated in real time in order to minimize the objective function. The objective function includes the quadratic error between the current and the desired target images. A simplex method is used to minimize the objective function, and a Newton-like method is also used near convergence. We successfully validated this method with simulations under the graphic library OpenGL. This work was presented in part at the 10th International Symposium on Artificial Life and Robotics, Oita, Japan, February 4–6, 2005     

机器人视觉伺服系统的研究

机器人伺觉伺服系统及到多学科内容。针对机器人视觉伺服系统主要的三方面内容;系统的结构方式,图象处理,控制方法,介绍了该领域的研究现状及所取得的成就。最后分析了今后的发展趋势。     

视觉伺服机器人对运动目标操作的研究

机器人视觉伺服系统是机器人领域一重要的研究方向,它的研究对于开发手眼协调的机器人在工业生产、航空航天等方面的应用有着极其重要的意义．本文着眼于视觉伺服机器人操作运动目标这一问题,分析了建立此类系统的控制结构并指明其特点;同时,详细地阐述了三个组成环节：视觉图像处理、预测及滤波、视觉控制器的研究方法和现状．最后,分析了今后的研究趋势．     

直线特征的交互矩阵求取

直线特征在视觉跟踪、视觉伺服中具有重要作用, 但目前的直线交互矩阵的求取受到制约, 需要已知含有直线的平面在摄像机坐标系中的方程参数. 为摆脱含有直线的平面参数的约束, 本文利用两点的极坐标推导出直线的交互矩阵, 并给出直线交互矩阵求取方法. 经分析得知, 对于与摄像机光轴接近垂直的直线, 其在成像平面上的角度变化主要受摄像机姿态变化的影响, 对摄像机的位置变化不敏感. 对于与摄像机光轴平行的直线, 其在成像平面上的角度变化受摄像机旋转以及垂直于光轴平移 的影响较大. 实验结果验证了本文方法的有效性.     

Hybrid Force and Vision-Based Contour Following of Planar Robots

Manipulators interacted with uncalibrated environments have limited dexterity due to constraints imposed by unknown environments. However, to perform science or industrial operations, it is necessary to be able to position and orient these manipulators on targets in order to accomplish required control tasks. This article describes how one might enhance manipulator dexterity for planar contour following tasks using hybrid force and vision-based control. The proposed control approach can guarantee task precision employing only a single-axis force sensor and an imprecisely calibrated CCD camera whose optical axis is perpendicular to the planar workspace. The goal of the autonomous task is to drive an instrument mounted on the end-effector of a planar robotic manipulator to follow a visually determined planar contour and continue tracking the contour in desired pose, contact force, and speed, all demanding time-varying, with precision. The proposed control architecture is suitable for applications that require simultaneous force and pose control in unknown environments. Our approach is successfully validated in a real task environment by performing experiments with an industrial robotic manipulator.     

基于分层运动分解的飞行机械臂视觉伺服控制

飞行机械臂系统具有主动作业能力,通过搭载视觉传感器感知周围环境,系统的自主能力将进一步提高.然而,考虑到无人机的欠驱动和整个系统的非线性特性,飞行机械臂系统的视觉伺服控制仍然是一项具有挑战性的工作.本文在充分考虑机械臂对无人机的力/力矩作用后,提出了一种基于分层运动分解的飞行机械臂视觉伺服控制方案.首先,对飞行机械臂系统的运动学和动力学模型进行分析.然后,根据所得的相机运动学模型,通过基于图像的视觉伺服控制获得相机的期望速度,进而制定无人机和机械臂的速度分配策略.在考虑机械臂运动时对无人机产生的力/力矩影响,设计了底层的飞行控制器.最后,在与现有方法的仿真对比中可以看出,所提方法具有良好的控制性能,对图像特征点位置的不确定性及图像噪声也表现了较好的鲁棒性.     

Integrating Optical Force Sensing with Visual Servoing for Microassembly

For microassembly tasks uncertainty exists at many levels. Single static sensing configurations are therefore unable to provide feedback with the necessary range and resolution for accomplishing many desired tasks. In this paper we present experimental results that investigate the integration of two disparate sensing modalities, force and vision, for sensor-based microassembly. By integrating these sensing modes, we are able to provide feedback in a task-oriented frame of reference over a broad range of motion with an extremely high precision. An optical microscope is used to provide visual feedback down to micron resolutions, while an optical beam deflection technique (based on a modified atomic force microscope) is used to provide nanonewton level force feedback or nanometric level position feedback. Visually servoed motion at speeds of up to 2 mm/s with a repeatability of 0.17 m are achieved with vision alone. The optical beam deflection sensor complements the visual feedback by providing positional feedback with a repeatability of a few nanometers. Based on the principles of optical beam deflection, this is equivalent to force measurements on the order of a nanonewton. The value of integrating these two disparate sensing modalities is demonstrated during controlled micropart impact experiments. These results demonstrate micropart approach velocities of 80 m/s with impact forces of 9 nN and final contact forces of 2 nN. Within our microassembly system this level of performance cannot be achieved using either sensing modality alone. This research will aid in the development of complex hybrid MEMS devices in two ways; by enabling the microassembly of more complex MEMS prototypes; and in the development of automatic assembly machines for assembling and packaging future MEMS devices that require increasingly complex assembly strategies.     

基于图像的双臂模糊自适应轨迹跟踪控制

在固定相机的监视下,随机位姿的目标物体在双臂系统的控制下执行轨迹跟踪控制任务时,既要考虑目标随机位姿引起的运动学和动力学不确定性问题,还要考虑运动学和动力学的协同问题.针对上述问题,分别采用自适应方法估计目标质心和特征点的位置信息,利用模糊逻辑系统逼近系统的动力学模型,使用分散控制策略处理双臂的协同问题,最后基于位置/力混合控制方法设计基于图像的双臂模糊自适应轨迹跟踪控制器,并采用李亚普诺夫方法证明系统的稳定性.仿真实验验证了所设计控制器的有效性.     

基于镜面反射型视觉传感器的车窗装配系统

机器人视觉伺服系统是机器人领域的一个重要研究方向,它的研究对于开发手眼协调的机器人在工业生产、设备制造等方面的应用有极其重要的意义。该文介绍了一种基于镜面反射型视觉传感器的车窗装配系统,着重介绍了该系统的设备构成、设计方案。最后分析了目前研究中存在的问题及今后的研究方向。     

具有深度自适应估计的视觉伺服优化

在手眼机器人视觉伺服中,如何确定机器人末端摄像机移动的速度和对物体的深度进行有效的估计还没有较好的解决方法.本文采用一般模型法,通过求解最优化控制问题来设计摄像机的速度,同时,利用物体初始及期望位置的深度估计值,提出了一种自适应估计的算法对物体的深度进行估计,给出了深度变化趋势,实现了基于图像的定位控制.该方法能够使机器人在工作空间范围内从任一初始位置出发到达期望位置,实现了系统的全局渐近稳定且不需要物体的几何模型及深度的精确值.最后给出的仿真实例表明了本方法的有效性.     

基于任务空间的无标定视觉机械臂自适应跟踪控制

针对动力学参数不确定的无标定视觉机械臂系统,研究基于任务空间的自适应控制问题。对于控制器的设计,首先研究机械臂动力学参数不确定情况下基于任务空间的控制问题,然后设计自适应摄像机标定控制器,最后根据任务空间信息和图像空间信息的一致收敛关系统一两部分控制器,设计整个闭环系统控制信号和自适应控制律。实验结果表明了所提出的控制方法的有效性。     

Selection of features and evaluation of visual measurements during robotic visual servoing tasks

This paper presents some of the computer vision techniques that were employed in order to automatically select features, measure features' displacements, and evaluate measurements during robotic visual servoing tasks. We experimented with a lot of different techniques, but the most robust proved to be the Sum-of-Squared Differences (SSD) optical flow technique. In addition, several techniques for the evaluation of the measurements are presented. One important characteristic of these techniques is that they can also be used for the selection of features for tracking in conjunction with several numerical criteria that guarantee the robustness of the servoing. These techniques are important aspects of our work since they can be used either on-line or off-line. An extension of the SSD measure to color images is presented and the results from the application of these techniques to real images are discussed. Finally, the derivation of depth maps through the controlled motion of the handeye system is outlined and the important role of the automatic feature selection algorithm in the accurate computation of the depth-related parameters is highlighted.The views and conclusions contained in this document are those of the author and should not be interpreted as representing the official policies, either expressed or implied, of the funding agencies. Tel. (612) 625-0163; e-mail address: npapas@cs.umn.edu.     

Visual servoing in the task-function framework: A contour following task

We describe an approach to contour following unknown objects using a handeye robotic system. Relevant and sufficient feature points providing optical flow data are extracted from the edges of the target object. The desired motion of the end-effector is computed with the objective of keeping the visual features always at the same target location in the image plane. A cartesian PD controller is used to perform the desired motion by the robot's end-effector. To address thecontrol issues, we take advantage of the unifying robot control theory stated in the literature as thetask-function approach [21]. To validate our approach, we restricted our experiments to motionless objects positioned in a plane parallel to the image plane: three degrees of freedom (two of translation, one of rotation) are thus controlled.     

迭代学习神经网络控制在机器人示教学习中的应用(英文)

示教学习是机器人运动技能获取的一种高效手段.当采用摄像机作为示教轨迹记录部件时,示教学习涉及如何通过反复尝试获得未知机器人摄像机模型问题.本文力图针对非线性系统重复作业中的可重复不确定性学习,提出一个迭代学习神经网络控制方案,该控制器将保证系统最大跟踪误差维持在神经网络有效近似域内.为此提出了一个适合于重复作业应用的分布式神经网络结构.该神经网络由沿期望轨线分布的一系列局部神经网络构成,每一局部神经网络对对应期望轨迹点邻域进行近似并通过重复作业完成网络训练.由于所设计的局部神经网络相互独立,因此一个全程轨迹可以通过分段训练完成,由起始段到结束段,逐段实现期望轨迹的准确跟踪.该方法在具有未知机器人摄像机模型的轨迹示教模仿中得到验证,显示了它是一种高效的训练方法,同时具有一致的误差限界能力.