Date of Award

12-1990

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering

Department

Department of Electrical and Computer Engineering

First Advisor

Steven K. Rogers, PhD

Second Advisor

Curtis Spenny, PhD

Abstract

An experimental digital platform is developed as an environment on which to evaluate digital control strategies for dexterous manipulation with a pneumatically actuated tendon-driven manipulator. This environment is used to begin the study of advanced control methods that are suitable for providing the tracking accuracy required for grasping and dexterous manipulation with a pneumatically actuated tendon-driven end-effector. The digital platform consists of a PC/AT-386 and a single board MC68020 microcomputer in a VME chassis with shared RAM between the two processors to control the Utah/MIT Dexterous Hand (UMDH). The MC68020 controls the A/D and D/A access for the UMDH, while the ARCADE-HAND experimental control environment is hosted on the PC/AT-386 for user interface and control determinations. An Adaptive Model-Based Control (AMBC) algorithm is implemented and experimentally evaluated on the UMDH. Tracking performance is compared to the PD baseline controller of the ARCADE HAND environment and evaluated for the requirements of human finger emulation. The evaluation includes compensation for unknown dynamics of the UMDH system, adaptability to unknown payloads, and multiple trajectory tracking capabilities. The superior tracking of the AMBC algorithm demonstrates the potential of technique for emulation of human finger movement.

AFIT Designator

AFIT-GE-ENG-90D-50

DTIC Accession Number

ADA230606

Comments

The author's Vita page is omitted

Download

Share

COinS