Kinematic and dynamic applications for multiple robotic manipulator systems
Abstract
In this thesis, three problems are addressed. The first problem is the manipulability measure for multiple manipulator systems. First the object Jacobian matrices for each manipulator and the entire system are developed. This leads to the derivation of the force and velocity manipulability ellipsoids for each manipulator and the entire system. Then the force and velocity manipulability ellipsoids for the end-effectors are developed. All the formulations for deriving the ellipsoids for the entire system and the end-effectors are performed without calculating a generalized inverse. The second problem is the self-motion for multiple manipulator systems. The definitions of self-motion for multiple manipulator systems are introduced. A kinematic model which contains the coupling constraint and the object constraint is presented. The solution of the self-motion problem is obtained at the joint velocity level. Examples show that manipulability measure for multiple manipulator systems can be used as the performance function to search for an optimal position or configuration in different self-motion cases. The third problem is the minimum-time control problem for multiple manipulators handling a rigid object along specified paths. Parameterized dynamic equations are used to derive the solutions of the problem. Then the largest admissible region for the velocities of the object is first found by the optimization methods, and the optimal minimum-time trajectory is constructed in the admissible region. A system matrix for a multiple manipulator system is developed and the redundant degrees of freedom (DOF) for the dynamic system have been investigated. If the dynamic system has a redundancy, additional constraints may be added, then the minimum-time solution can be found with the additional constraints. Once the optimal trajectory has been constructed, input torques can be computed. In a redundant system, input torques may be selected to achieve a secondary goal. Examples are given to illustrate how to solve the minimum-time problem with or without additional constraints.
Degree
Ph.D.
Advisors
Koivo, Purdue University.
Subject Area
Engineering|Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.