基于KUKA工业机器人的定位误差补偿方法的研究

发布时间：2018-10-16 22:27

【摘要】：随着生产力水平的不断进步和科学技术的飞速发展,工业机器人作为先进、智能的工业化设备的代表,在社会生活的很多方面应用广泛,尤其在汽车产业领域得到了相当成熟和广泛的应用。车身的尺寸偏差是影响车身质量最重要的因素,以工业机器人车身激光检测系统为代表的车身尺寸检测技术已成为当前全球车企降低车身尺寸偏差,提升车身制造精度最为有效的手段之一,对于工业机器人定位精度的研究具有非常重要的工程意义和经济意义。首先,本论文以工业机器人车身激光检测系统中最常用的KUKA工业机器人为例,对其运动学问题进行了理论剖析,阐述了建立正向运动学模型和求解运动学逆解的方法,使用MATLAB Robotics Toolbox进行了运动学模型的仿真验证。其次,在运动学模型的基础上,考虑影响工业机器人定位精度最大的几何误差因素,采用微分的方法建立了几何偏差模型;在考虑车身激光检测工程应用的基础上,分析了车身定位坐标系与机器人激光测量坐标系不重合对工业机器人定位精度造成的耦合影响,并依此在几何偏差模型的基础上建立了基于车身激光检测系统的KUKA工业机器人车身定位误差模型,并对该模型进行了仿真验证,确定了其有效性和准确性。再次,在进行工业机器人定位误差补偿方法的研究时,首先基于所建立的工业机器人车身定位误差模型,使用牛顿-拉夫逊迭代算法进行了仿真验证。考虑到基于误差模型进行定位误差补偿的局限性,设计了合适的BP神经网络对KUKA机器人的误差模型进行网络逼近并进行了误差补偿,并且完成了对比仿真实验。结果表明,采用BP神经网络进行定位误差补偿比基于误差模型进行定位误差补偿的补偿精度更高。最后,考虑到传统BP神经网络收敛速率慢等问题,提出采用PSO算法与BP神经网络相结合的算法,对BP神经网络进行优化,使KUKA机器人定位误差补偿的效果更佳。
[Abstract]:With the continuous progress of productivity and the rapid development of science and technology, industrial robots, as the representatives of advanced and intelligent industrial equipment, are widely used in many aspects of social life. Especially in the field of automobile industry has been quite mature and widely used. The dimension deviation of the body is the most important factor that affects the quality of the body. The measurement technology of the car body size, which is represented by the laser inspection system of the industrial robot body, has become the current global automobile enterprises to reduce the body size deviation. One of the most effective means to improve the precision of body manufacturing is of great engineering and economic significance for the research of positioning accuracy of industrial robots. Firstly, taking the KUKA industrial robot, which is the most commonly used industrial robot in the body laser detection system of industrial robot, as an example, the kinematics problem is analyzed theoretically, and the method of establishing forward kinematics model and solving the inverse kinematics solution is expounded. The kinematics model is simulated with MATLAB Robotics Toolbox. Secondly, on the basis of kinematics model, considering the geometric error factors that affect the positioning accuracy of industrial robot, the differential method is used to establish the geometric deviation model. The coupling effect on the positioning accuracy of industrial robots caused by the non-coincidence of body positioning coordinate system and robot laser measuring coordinate system is analyzed. Based on the geometric deviation model, the body positioning error model of KUKA industrial robot based on body laser detection system is established. The simulation results show that the model is effective and accurate. Thirdly, in the research of the positioning error compensation method of industrial robot, the Newton-Raphson iterative algorithm is used to verify the error model of industrial robot body positioning. Considering the limitation of positioning error compensation based on error model, a suitable BP neural network is designed to approximate and compensate the error model of KUKA robot, and a comparative simulation experiment is completed. The results show that the compensation accuracy of positioning error based on BP neural network is higher than that based on error model. Finally, considering the slow convergence rate of traditional BP neural network, an algorithm combining PSO algorithm with BP neural network is proposed to optimize the BP neural network, so that the positioning error compensation of KUKA robot is better.
【学位授予单位】：长春工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP242

【相似文献】