基于在线位姿测控系统的工业机器人定位精度优化

发布时间：2019-02-25 13:21

【摘要】：现代制造技术中将工业机器人作为智能制造系统的主要构成部分。随着德国“工业4.0”高端制造技术计划和《中国制造2025》国家战略计划的提出,基于工业机器人的数字化柔性加工系统在工业领域的重要性迅速上升。然而,工业机器人现有的绝对定位精度还不足以满足精密加工领域的需求。目前通过修正机器人的运动学参数误差虽然在一定程度上可以提高机器人定位精度,但机器人的运动学模型并不能完整准确地描述机器人实际位姿,并且离线标定无法反映机器人受到的环境、负载等干扰。因此,本文研究了一种通过外部测控系统实时测量机器人终端位姿,优化补偿机器人定位精度的方法。通过外部测控系统的在线测量和补偿,得到机器人终端的实时位姿信息,不仅能简化机器人复杂的参数建模过程,适应实际情况变化,且能够反映机器人的综合误差。本文利用近景摄影系统、双轴倾角测量仪和PC机组合为一套六自由度位姿在线测量系统,它能够通过视觉、电学、惯性测量相结合的方式实时获取机器人末端执行器的六自由度位姿信息。本文利用离线误差标定技术和终端测量技术相结合,离线与在线方式相结合,发展了一种机器人离线参数标定方法。研究了基于卡尔曼滤波和矩阵加权线性最小方差最优融合准则的多传感器数据融合方法,实现了机器人在线位姿测量系统的超前动态预估和冗余数据融合。同时,发展了一种工业机器人编码器解算值与双轴惯性倾角仪相结合,测量三维姿态角的方法。本文研究并分析了在线位姿测量系统的最佳工作范围和最佳工作条件。提出一种基于点云特征线重合的坐标系转换方法,能够准确快速地统一在线位姿测量系统与工业机器人的坐标系。基于KUKA KR 5 arc工业机器人编写了一套实时测量控制软件算法,在实验室环境下,实现了在线位姿测量系统与工业机器人的实时通信和反馈补偿。
[Abstract]:Industrial robot is regarded as the main component of intelligent manufacturing system in modern manufacturing technology. With the development of "Industry 4.0" High-end Manufacturing Technology Plan and "made in China 2025" National Strategic Plan in Germany, the importance of digital flexible machining system based on industrial robot is rising rapidly in the field of industry. However, the absolute positioning accuracy of industrial robots is not enough to meet the needs of precision machining. At present, although the robot positioning accuracy can be improved to some extent by modifying the kinematics parameter error of the robot, the kinematics model of the robot can not describe the actual position and pose of the robot completely and accurately. And off-line calibration can not reflect the robot environment, load and other interference. Therefore, this paper studies a real-time measurement and control system based on external measurement and control system to measure the position of robot terminal and optimize the compensation of robot positioning accuracy. Through the on-line measurement and compensation of the external measurement and control system, the real-time position and pose information of the robot terminal can be obtained, which not only simplifies the complex modeling process of the robot, adapts to the change of the actual situation, but also can reflect the comprehensive error of the robot. In this paper, a six-degree-of-freedom position and pose on-line measurement system is developed by using a close range photography system, a biaxial inclination measuring instrument and a PC machine. The six-degree-of-freedom pose information of the robot end-actuator is obtained in real-time by combining inertial measurement. In this paper, an off-line parameter calibration method for robot is developed by combining off-line error calibration technology with terminal measurement technology, off-line method and on-line method. The multi-sensor data fusion method based on Kalman filter and matrix weighted linear minimum variance optimal fusion criterion is studied. The advanced dynamic prediction and redundant data fusion of robot online pose measurement system are realized. At the same time, a method of measuring the three-dimensional attitude angle by combining the calculation value of the industrial robot encoder with the biaxial inertial dip meter is developed. In this paper, the optimum working range and working conditions of on-line pose measurement system are studied and analyzed. A coordinate system transformation method based on point cloud feature line coincidence is proposed, which can accurately and quickly unify the coordinate system between the online position and attitude measurement system and the industrial robot. Based on KUKA KR 5 arc industrial robot, a set of real-time measurement and control software algorithm is developed. In the laboratory environment, the real-time communication and feedback compensation between the on-line position and attitude measurement system and the industrial robot are realized.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP242

【相似文献】