Author

Scott Prince

Date of Award

2010

Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Mechanical Engineering

Advisor

Mohsen Shahinpoor

Second Committee Member

Senthil Vel

Third Committee Member

Rick Eason

Abstract

In robotics, the solutions to the inverse kinematics equations of open-chain articulated robotic manipulators require solving a set of 12 trigonometric transcendental equations which often posses multiple non-unique solutions. Closed form analytic solutions to the inverse kinematics equations are often time consuming to derive and are specific to a particular robotic geometry. In this thesis, a numerical algorithm using particle swarm optimization (PSO) is utilized to obtain solutions of the inverse kinematics equations for robotic manipulators. The solution method uses a constricted global topology PSO algorithm to obtain an approximate solution to the problem and then utilizes a damped least squares algorithm (DLS) to further refine the final solution. The various tuning parameters and their individual effects on the algorithm are also discussed, and experimental results performed on simulations of some common robotic manipulator configurations are also presented.

Share