Date of Award

11-6-2015

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Mechanical, Automotive, and Materials Engineering

First Advisor

Minaker, Bruce

Keywords

Control, Dynamics, Kinematics, Reconfigurable, Robot

Rights

CC BY-NC-ND 4.0

Abstract

Robots with predefined kinematic structures are successfully applied to accomplish tasks between the robot and the environment. For more sophisticated and future applications, it is necessary to extend the capabilities of robots, and employ them in more complex applications, which generally require accurate and more changeable structural properties during the interaction with the environment. The central focus of this research was to propose a robot with new properties to address the reconfigurability problem, including its feasible solutions using model based control strategies. First, these reconfigurable robots have to combine as many properties of different open kinematic structures as possible and can be used for a variety of applications. The kinematic design parameters, i.e., their Denavit-Hartenberg (D{H) parameters, were modeled to be variable to satisfy any configuration required to meet a specific task. By varying the joint twist angle parameter (a configuration parameter), the presented model is reconfigurable to any desired open kinematic structure, such as Fanuc, ABB and SCARA robots. The joint angle and the offset distance of the D-H parameters are also modeled as variable parameters (a reconfigurable joint). The resulting reconfigurable robot hence encompasses different kinematic structures and has a reconfigurable joint to accommodate any required application in medical technology, space exploration and future manufacturing systems, for example. Second, a methodology was developed to automate model generation for n-DOF Global Kinematic Model (n-GKM). Then, advanced model based control strategies were employed to increase performance as compared to less structured approaches. An algorithm was developed to select a relevant kinematic structural robot configuration for any predefined geometric task. The main contribution of this research is that it combines a kinematic structural design with control design methods to optimize robot capability and performance. This combination has been established by developing an algorithm to select the optimal kinematic structure and the most applicable control approach to perform a predefined geometric task with high tracking performance.

Share

COinS