Precise motion control of flexible -joint robot manipulators with an intelligent payload estimator
Abstract
Many industrial machines like mechanical and electronic assembly devices, and material-handling machines, handle a variety of payloads. In order to increase productivity and precision, industry demands high degree of automation, high-speed operation and good control performance from their machines. Two common issues must be dealt with in order to satisfy the demands from industry. They are payload variation and compliance of the driving component in a system. Flexible joint robot manipulators with variable payload mass face the same issues during high-speed operation. The objective of this research is to develop control strategies that produce fast and precise tracking motion control of robot manipulators with sudden changes in payload mass. A control strategy is proposed, and it consists of three main components: a trajectory generator, a feedback control block, and an intelligent payload estimator. The trajectory generator creates a desired trajectory such that the robot manipulator produces minimal residual vibration at the destination point. Two different control laws are proposed for the feedback control block. A feedforward inverse dynamics with a PD controller, is considered because of easy implementation of the control law. A feedback linearization is considered because of its well-defined control design methodology. Each control law is used to compute the necessary motor torque to move the robot manipulators along the desired trajectory. The intelligent payload estimator is developed based on intuitive understanding of motion of the robot manipulator due to changes in payload. It does not require an analytical model of the system. It will estimate changes of the payload and provide accurate information of the payload to the proposed controller such that the controlled robot manipulator tracks a desired trajectory with minimal error. Experiments are conducted to show the validity and the performance of the proposed control approach for the robot manipulator. The proposed inverse dynamics with a PD controller produces excellent tracking performance. In order to achieve high performance, the dynamic model of the robot manipulator must be considered in control design. Furthermore, providing accurate information about the system, i.e. payload variation, to the feedback controllers improves performance.
Degree
Ph.D.
Advisors
Meckl, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.