Date of Award

1-7-2011

Document Type

Thesis

Degree Name

Master of Science in Mechanical Engineering (MSME)

Department

Mechanical Engineering

First Advisor

Christopher A. Kitts

Abstract

Multi-robot systems have the potential to exceed the performance of many existing robotic systems by taking advantage of the cluster’s redundancy, coverage and flexibility. These unique characteristics of multi-robot systems allow them to perform tasks such as distributed sensing, gradient climbing, and collaborative work more effectively than any single robot system. The purpose of this research was to augment the existing cluster space control technique in order to demonstrate effective gradient-based functionality, specifically, that of tracking gradient contours of specified concentration levels. To do this, we needed first to estimate the direction of the gradient and/or contour based on the real-time measurements made by sensors on the distributed robots, and second, to steer the cluster in the appropriate direction.

Successful simulation, characterization, and experimental testing with the developed testbed have validated this approach. The controller enabled the cluster to sense and follow a contour-based trajectory in a parameter field using both a kayak cluster formation and also the land based Pioneer robots. The positive results of this research demonstrate the robustness of the cluster space control while using the contour following technique and suggest the possibility of further expansion with field applications.

Download

Share

COinS