模型及负载不确定情形下机械臂自适应运动控制

发布时间：2018-01-26 07:36

本文关键词： 机械臂不确定性 FAT技术 “一步猜测” 自适应控制　出处：《北京理工大学》2016年硕士论文　论文类型：学位论文

【摘要】：在当今的工业生产和日常生活中,机器人扮演着日益重要的角色。随着机器人的应用场景的日益广泛,作为机器人执行机构重要组成部分的机械臂,其动力学模型具有高度非线性化、强耦合和时变的特点,这使得机械臂在工作过程中存在着许多的不确定性。为了使机械臂在面对不确定性时具有更好的智能性,本课题研究了机械臂在自身模型具有不确定性和外力负载具有不确定性的情形下,分别通过基于FAT(函数逼近技术)技术对模型不确定性的矩阵进行估计的方法和“一步猜测”思想对未知的负载进行估计的方法,并在此基础上设计出阻抗自适应控制器和离散时间自适应控制器,使得机械臂在不确定性情形下的智能性得到提高。本文主要工作概括如下:(1)针对机械臂模型中存在的不确定性,通过FAT技术估计模型中的不确定项,在此基础上设计出机械臂的阻抗自适应控制器,并对系统的稳定性进行证明。通过将该控制器应用于基于iCub机器人的双臂跟随控制实验中,验证了该算法的实际可行性。(2)针对机械臂的外力负载中存在的不确定性情形,根据“一步猜测”思想对当前时刻的外力负载进行估计,然后设计出离散时间自适应控制器。为验证算法的理论可行性,分别将其应用于两自由度和六自由度机械臂PUMA560仿真实验中,通过实验表明该算法对未知的负载估计效果和关节角轨迹跟踪效果良好。本论文的研究对于机械臂的智能性研究具有一定的探索意义,对于未来机械臂的智能性发展有所裨益。
[Abstract]:Nowadays, robots play an increasingly important role in industrial production and daily life. With the increasing application of robots, as an important part of robot actuating mechanism, robot arm is becoming more and more important. Its dynamic model is highly nonlinear, strongly coupled and time-varying. In order to make the manipulator have better intelligence in the face of uncertainty, there are many uncertainties in the working process. In this paper, the manipulator is studied under the uncertainty of its own model and the uncertainty of the external load. The method of estimating the uncertain matrix of the model based on the FAT-based technique and the method of estimating the unknown load based on the idea of "one step guess" are presented respectively. On this basis, the impedance adaptive controller and discrete time adaptive controller are designed. The main work of this paper is summarized as follows: 1) aiming at the uncertainty in the manipulator model, the uncertainty in the model is estimated by FAT technique. On this basis, the impedance adaptive controller of the manipulator is designed, and the stability of the system is proved. The controller is applied to the two-arm following control experiment based on iCub robot. The feasibility of the algorithm is verified. (2) in view of the uncertainty existing in the external force load of the manipulator, the external force load at the current moment is estimated according to the idea of "one step guess". Then a discrete-time adaptive controller is designed. In order to verify the theoretical feasibility of the algorithm, it is applied to the PUMA560 simulation experiment of two-degree-of-freedom and six-degree-of-freedom manipulators, respectively. Experiments show that the algorithm has good effect on unknown load estimation and joint angle trajectory tracking. The research in this paper has a certain exploration significance for the intelligent research of manipulator. It is beneficial to the intelligent development of robot arm in the future.
【学位授予单位】：北京理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP241

【相似文献】