Adapting Robot Architecture for Pick and Place

Master thesis (2012)

Authors

Contributors

P. Lambert (mentor)
Department
Precision and Microsystems Engineering (Mechanical, Maritime and Materials Engineering) (TU Delft)
Copyright: © 2012 Wu, H.
More Info
expand_more

Published Date

11-09-2012

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Mechanical, Maritime and Materials Engineering

Department

Precision and Microsystems Engineering

Abstract

Pick and place robots are very popular in industry as they provide speedy and precise handling applications. They are especially practical in repetitive work as they can work long time with a faster and more accuracy motion than humans. Compared with serial manipulator, parallel robots have the advantages of high accuracy, high rigidity and high payload/inertia ratio, and thus more suitable for a pick and place application. In this thesis, a 5 DOF parallel robot designed for haptic application is approaching to achieve pick and place operation by improving its tracking speed and accuracy through inverse dynamic control. Pick and place robots have different performance criteria than haptic devices. One main difference is that they operate at higher speed and accuracy than haptic devices. Therefore the parallel robot is first calibrated by using mechanical constraint method to reduce the position error from kinematic computation. Inverse kinematic model based iteration method is used to find out real structure parameters after fabrication. Then an inverse dynamic model of the 5 DOF parallel robot is built up based on virtual work principle, which considers two situations and is able to be adapted dependent on whether an object is captured by grippers. Also, grasping force is setted by this model and therefore could be easily changed for different application by humans. A simmechanics model in matlab is used to compare with this numerical inverse dynamic model and proves its correctness. The inverse dynamic model provides feedforward compensation in high speed movement and reduces the tracking error than normal position control. In order to obtain higher stiffness in grasping performance while keeps fast response and low overshoot, a nonlinear control strategy is introduced to replace normal PID controller. Simulations and experiments are carried out to test tracking performance under dynamic control, the results show a limited error and a realization of pick and place operation finally.