【作者】 王建军；

【导师】 杜付鑫；

【作者基本信息】 山东大学 ， 机械工程， 2023， 硕士

【摘要】 随着建筑业的快速发展,大型高层建筑拔地而起,建筑施工面临的工作人员老龄化、工人劳动强度大、危险性大、施工效率低且施工质量难以得到保障等问题逐渐凸显。涉及大型幕墙的安装,更是需要多名工人协同作业,危险性高,事故频发。本课题在国家自然科学基金委联合基金支持下,基于人机协同作业理念,研制幕墙安装机器人系统,并对控制系统的稳定性、协同作业的柔顺性进行改进,开展建筑施工的智能化研究,推进机器人技术在建筑施工行业的实用化。本文对比分析干挂石材幕墙行业中使用的挂件,分析干挂石材幕墙的传统安装流程与施工工艺,归纳、总结传统建筑施工中亟待解决的问题,提出一套适用于人机协同作业的幕墙安装工艺流程。采用合适的人机系统思想,设计一种适用于高空安装幕墙的人-机器人协同作业系统,并设计合理的机器人电气控制系统,为幕墙板材安装的控制奠定基础。然后,基于动态运动基元原理规划机器人的作业轨迹。根据人机协同安装幕墙策略,针对同一作业任务进行多次示教,采用快速傅里叶变换算法分析每条示教轨迹的参与权重,并规划出机器人的作业轨迹。使用动态运动基元算法,对规划出的作业轨迹进一步进行学习与泛化,使得经过规划的轨迹更具普适性。而后,搭建机器人变导纳控制模型,使用神经网络对导纳控制器参数进行预测。通过对人机协同作业系统的稳定性分析,获取导纳参数的可调节范围,并采用双六维力传感器耦合作业方式来确保机器人能够安全作业。对比分析BP神经网络与RBF神经网络在预测导纳参数中的绝对误差和相对误差。借助MATLAB/Simulink软件,对本文所提出的导纳控制模型进行仿真与结果分析。最后,进行了人机协作系统的随动性能验证实验,机器人轨迹泛化实验以及精细安装幕墙实验,验证了本文所提出方法应用于人机协同安装幕墙系统的合理性,提高了控制系统的操作柔顺性。更多还原

【Abstract】 With the rapid development of the construction industry,large high-rise buildings have sprung up.The problems faced by the construction industry,such as the aging of the staff,the high labor intensity of the workers,the high risk,the low construction efficiency,and the difficulty in ensuring the construction quality,have become increasingly prominent.The installation of large curtain walls requires more than one worker to work together,with high risk and frequent accidents.With the support of the joint fund of the National Natural Science Foundation of the People’s Republic of China,based on the concept of human-robot cooperative operation,this project develops a curtain wall installation robot system,improves the stability of the control system and the flexibility of cooperative operation,carries out intelligent research on building construction,and promotes the application of robot technology in the construction industry.This paper compares and analyzes the pendants used in the dry-hanging stone curtain wall industry,analyzes the traditional installation process and construction process of dry-hanging stone curtain walls,sums up and summarizes the problems that need to be solved in the traditional construction,and proposes a set of curtain wall installation process suitable for human-robot cooperation.A human-robot cooperative operation system suitable for the installation of curtain walls at high altitudes is designed by adopting the appropriate human-robot system idea,and a reasonable robot electrical control system is designed to lay the foundation for the control of curtain wall panel installation.Then,the robot’s operation trajectory is planned based on the principle of Dynamic Movement Primitives.According to the strategy of human-robot cooperation in curtain wall installation,the same task is taught several times,and the fast Fourier transform algorithm is used to analyze the participation weight of each teaching track,and the robot’s operation trajectory is planned.The dynamic movement primitive algorithm is used to further study and generalize the planned operation trajectory,making the planned trajectory more general.Besides,the neural network is used to analyze the operation intention of the human operator,and the variable admittance control model of the robot is built.Through the stability analysis of the human-robot cooperative operation system,the adjustable range of the admittance parameters is obtained,and the double six-dimensional force sensor is used to ensure that the robot can work safely.With the help of the neural network toolbox provided by MATLAB,the absolute error and relative error of BP neural network and RBF neural network in predicting human operation intention are compared and analyzed.With the help of MATLAB/Simulink software,the variable admittance control model proposed in this paper is simulated and the results are analyzed.Finally,the following performance verification experiment of the human-robot cooperation system,robot trajectory generalization experiment,and fine curtain wall installation experiments are carried out in the physical space,which verifies the rationality of the proposed method applied to human-robot cooperation curtain wall installation system and improves the operational flexibility of the control system.更多还原