Masters Theses

Qamar Asghar

Date of Award

12-1990

Degree Type

Thesis

Degree Name

Master of Science

Major

Mechanical Engineering

Major Professor

James A. Euler

Committee Members

R. V. Dubey, S. T. Becker

Abstract

Kinematically redundant robots have been the subject of many investigations in which researchers have tried to avoid obstacles by using the available degree of redundancy. These schemes are usually based on the assumption of a well defined obstacle location and position in the robot's work space. A more complete solution to this problem includes a continuous feedback mechanism for accurate location of moving obstacles or objects in the robot's work envelope. A robot which can "see", can prove to be more dextrous in the performance of complex tasks while at the same time it requires less stringent control mechanisms than preprogrammed machines. A sensory, trainable system is also adaptable to a much larger variety of tasks, thus achieving a degree of universality that ultimately translates into lower production and maintenence cost. An obstacle avoidance scheme based on sensory feedback is presented for the kinematic control of a robotic manipulator with a single degree of redundancy, which can be extended to manipulators with multiple degrees of redundancy. An inexpensive ultrasonic sensor is used to locate obstacles in the robot's workspace and provide feedback for modification of the robot's wrist trajectory. The kinematic control scheme which uses the feedback from the sensor for local optimization of trajectory parameter is based upon the gradient projection method developed by Dubey and Euler et al. [5]. This study also explores the possibilities of using the sensor as an inexpensive feedback control loop device to duplicate the results of computer vision.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS