SIA OpenIR  > 工艺装备与智能机器人研究室
机器人研磨抛光智能力控制方法研究
Alternative TitleResearch on Intelligent Power Control Method of Robot Grinding and Polishing
刘速杰
Department工艺装备与智能机器人研究室
Thesis Advisor李论
Keyword机器人研磨抛光 重力补偿 恒力控制 模糊PID控制 RBF神经网络PID控制
Pages73页
Degree Discipline检测技术与自动化装置
Degree Name硕士
2020-05-26
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract在工件打磨抛光领域,目前我国主要依赖人工操作。人工打磨不免面临打磨质量、生产效率、工人工作环境以及生产污染物可控性等诸多问题,技术工人短缺,雇佣费用昂贵,企业面临劳动力成本大幅上升的压力。在生产矛盾不断突出的背景下,伴随着机器人技术、计算机技术的不断发展,机器人在抛光打磨这一基础工种上的研究与应用日益受到产业及学术界的关注。然而,在机器人对工件的打磨过程中,工件安装位置、机器人与打磨工具自振以及机器人自身精度等因素会导致由工件三维模型生成的理想打磨轨迹与机器人实际运动轨迹存在偏差,从而引起打磨工具空转或工件过度打磨的问题,因此需要保持在打磨过程中的恒定感知力,实现工件表面的均匀打磨。基于此,本课题以航空发动机机匣为研究载体,对机器人在复杂曲面自动化打磨中的力控制技术进行了研究。本文简述了国内外机器人打磨恒力控制的研究现状及主要研究方法,基于航空发动机机匣复杂曲面实验工件,搭建以KUKA工业机器人和ATI六维力传感器为基础的机器人打磨柔顺控制平台,上位机、机器人、力传感器三者组建局域网络实现数据实时通信。机器人研磨抛光技术的关键在于保持打磨工具与待加工工件之间感知力的恒定,通过数学分析建立机器人末端负载重力补偿算法,消除负载重力因素对机器人末端感知力求解的影响,基于机器人多位姿运动验证重力补偿算法的有效性,在此基础上提出基于基坐标系和基于工具坐标系的两种感知力求解方案。打磨工具高速转动过程中与待打磨工件接触振动产生较强干扰信号,使得传感器采集数据波动严重,直接影响感知力计算,因此有必要在求解感知力前对传感器数据进行滤波,使得传感器数值稳定,较大程度反映感知力的真实值。采用一阶低通滤波算法对传感器信号进行滤波,并通过实验对比分析获得合理的滤波频率。控制器的设计是机器人研磨抛光恒力控制的核心,采用模糊PID控制算法和RBF神经网络PID控制算法对基本PID控制器进行优化,在已有数学模型基础上采用MATLAB仿真平台对控制算法进行分析与优化。辅助于航空发动机机匣复杂曲面研磨抛光实际工况,经实验验证证明控制算法的有效性,能够在机器人研磨抛光过程中保持感知力的恒定。
Other AbstractIn the field of workpieces grinding and polishing. At present, our country relies mainly on manual operation. Manual grinding inevitably faces many problems such as grinding quality, production efficiency, workers' working environment and controllability of production pollutants. Shortage of skilled workers and expensive hiring puts businesses under pressure for significantly higher labor costs. Against the backdrop of production contradictions and along with the continuous development of robotics and computer technology, the research and application of robots in the basic work of polishing and grinding is increasingly receiving attention from industry and academia. However, during the robot's grinding of the workpieces, factors such as the mounting position of the workpieces, the robot's vibration with the grinding tool and the robot's own accuracy can cause the ideal grinding trajectory generated by the 3D model of the workpieces to deviate from the robot's actual motion trajectory. This will cause problems with the grinding tool idling or over- grinding of the workpieces. So there is a need to maintain a constant perception during the grinding process to achieve uniform grinding of the workpieces surface. Based on this, this topic investigates the robot's force control technique in the automated polishing of complex surfaces using the aero-engine casing as a research vehicle. This paper briefly describes the current research status and main research methods of robotic grinding constant force control at home and abroad. Based on the complex curved surface experimental workpieces of aero-engine casing, a robotic grinding compliance control platform based on KUKA industrial robot and ATI six-dimensional force sensor is built. Machines, robots, and force sensors form a local area network to realize real-time data communication. The key of the robot grinding and polishing technology is to keep the contact force between the grinding tool and the workpieces to be processed constant. The gravity compensation algorithm of the robot end load is established through mathematical analysis to eliminate the influence of the load gravity factor on the solution of the robot's end contact force. Based on the robot's multi-pose motion, the effectiveness of the gravity compensation algorithm is verified. On this basis, two solutions for contact force based on the base coordinate system and the tool coordinate system are proposed. The vibration of the grinding tool in contact with the workpieces to be polished during high-speed rotation produces a strong interference signal, which makes the data collected by the sensor fluctuate seriously and directly affects the contact force calculation. Therefore, it is necessary to filter the sensor data before solving the contact force, so that the sensor value is stable and reflects the true value of contact force to a greater extent. The first-order low-pass filter algorithm is used to filter the sensor signal, and a reasonable filter frequency is obtained through experimental comparison and analysis. The design of the controller is the core of the robot grinding and polishing constant force control. The fuzzy PID control algorithm and the RBF neural network PID control algorithm are used to optimize the basic PID controller. Based on the existing mathematical model, the MATLAB simulation platform is used to analyze and optimize the control algorithm to assist in the actual grinding and polishing of the complex curved surface of the aero-engine casing. And the effectiveness of the control algorithm is proved by experiments, which can keep the contact force constant during the grinding and polishing of the robot.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/27121
Collection工艺装备与智能机器人研究室
Affiliation中国科学院沈阳自动化研究所
Recommended Citation
GB/T 7714
刘速杰. 机器人研磨抛光智能力控制方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2020.
Files in This Item:
File Name/Size DocType Version Access License
机器人研磨抛光智能力控制方法研究.pdf(3920KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘速杰]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘速杰]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘速杰]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.