基于力反馈的打磨机器人控制系统研究

发布时间：2018-05-25 08:04

本文选题：工业机器人 + 打磨加工　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：机器人技术的飞速发展,使得大量的机器人被应用到工业生产中,随着感知系统的出现,人工打磨这类传统制造工艺,也在逐步被机器人自动打磨取代。本文构建了基于力反馈的打磨机器人控制系统,目的是为了控制加工过程中的接触力,提高打磨加工质量。基于封闭的机器人控制器,研究了力/位置控制算法,设计了相应的打磨控制方案,搭建了机器人打磨控制系统,实现了接触力恒定的打磨加工。分析打磨加工精度的影响因素,提出了打磨法向恒力控制策略。对于机器人末端与工件的作用过程,建立了末端位移与接触力的数学模型,在封闭的机器人控制器下,只能控制末端位移,引入阻抗控制算法,实现了间接控制接触力;建立了基于位置的阻抗控制模型,并分析了不同目标阻抗参数的作用。通过对阻抗控制的稳态误差分析,基于在线估算环境位置和刚度,建立了自适应阻抗控制模型,并仿真验证了控制策略对环境变化的适应能力。研究了打磨轨迹规划过程,通过实时调整末端打磨轨迹,实现打磨法向力恒定。基于机器人坐标系统,分析了末端打磨工具姿态变换过程,设计了打磨位置点控制策略的实现过程。根据打磨加工特点,提出了基于试教打磨轨迹的动态轨迹规划,设计了单点循环打磨控制流程,解决阻抗模型中的参数固定问题;基于当前打磨点位置修正,下一打磨点位置补偿的轨迹调整策略,设计了自适应阻抗控制实时轨迹调整流程。提出了对于未知轮廓估算打磨轨迹的动态规划,设计了边估算轨迹边打磨的控制流程。构建了机器人打磨实验平台。搭建了硬件系统,设计了末端打磨工具的装夹方式。编写了软件系统,建立了系统间的实时通讯方案,设计了在线位置修正和力信息实时处理的运行流程。最后研究了机器人打磨力控制实验。根据打磨工艺参数,设计了单因素试验,验证了工艺参数对打磨加工的影响,并推导了切入深度与磨削力的经验公式。设计了机器人对位移的响应实验,提出了控制机器人法向速度的加工方式,推导了偏移量与法向速度的函数关系式。设计了力控制精度对比试验,验证了控制方案的有效性。设计了对未知轮廓估算轨迹的力控制试验,位置跟踪效果良好,力控制在恒定区间内,验证了控制方案可行。
[Abstract]:With the rapid development of robot technology, a large number of robots are used in industrial production. With the appearance of perceptual system, traditional manufacturing technology such as manual grinding is gradually replaced by robot automatic grinding. The control system of grinding robot based on force feedback is constructed in this paper. The purpose of this system is to control the contact force in machining process and improve the quality of grinding process. Based on the closed robot controller, the force / position control algorithm is studied, the corresponding grinding control scheme is designed, the robot grinding control system is built, and the grinding process with constant contact force is realized. The influence factors of grinding precision are analyzed, and the control strategy of grinding normal constant-force is put forward. The mathematical model of the end displacement and contact force is established for the interaction between the robot end and the workpiece. Under the closed robot controller, the terminal displacement can only be controlled, and the impedance control algorithm is introduced to realize the indirect control of the contact force. The impedance control model based on position is established, and the function of different impedance parameters is analyzed. By analyzing the steady state error of impedance control, an adaptive impedance control model is established based on the on-line estimation of environment position and stiffness, and the adaptability of the control strategy to environmental change is verified by simulation. The process of grinding trajectory planning is studied, and the normal force of grinding is constant by adjusting the trajectory of end grinding in real time. Based on the robot coordinate system, the attitude transformation process of the end grinding tool is analyzed, and the realization process of the grinding position control strategy is designed. According to the characteristics of grinding processing, the dynamic trajectory planning based on the track of grinding is proposed, and the control flow of single point circular grinding is designed to solve the problem of fixed parameters in the impedance model. The trajectory adjustment strategy of the next grinding point position compensation is proposed and the real-time trajectory adjustment flow of adaptive impedance control is designed. In this paper, the dynamic programming for the unknown contour to estimate the grinding trajectory is proposed, and the control flow of grinding while estimating the trajectory is designed. A robot grinding experiment platform is constructed. The hardware system is built and the clamping mode of the end grinding tool is designed. The software system is written, the real-time communication scheme between the systems is established, and the running flow of on-line position correction and force information real-time processing is designed. Finally, the control experiment of robot grinding force is studied. According to the grinding process parameters, a single factor test was designed to verify the influence of the process parameters on the grinding process, and the empirical formulas of the cutting depth and grinding force were deduced. The experiment of robot response to displacement is designed. The machining method of controlling the normal velocity of the robot is proposed. The functional relationship between the offset and the normal velocity is derived. A contrast test of force control accuracy is designed to verify the effectiveness of the control scheme. A force control experiment is designed to estimate the trajectory of unknown contour. The effect of position tracking is good and the force control is in a constant range. The feasibility of the control scheme is verified.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】