Date of Award
2009
Degree Type
Thesis
Degree Name
Master of Engineering Science
Program
Electrical and Computer Engineering
Supervisor
Dr. Mehrdad Kermani
Second Advisor
Dr. Amir Yazdani
Abstract
Industrial manipulators perform repetitive and dangerous tasks. They are widely used, however present a source for accidental collisions with human operators. Therefore, they require large isolated spaces heavily taxing factory real-estate. Thus, there exists a need to create a safe cooperative working space shared by both manipulators and humans. The purpose of this research is to provide such an environment by integrating a safety mat-style sensory system, with an implementation of a potential field trajectory planning algorithm. The safety mat sensor has been designed and constructed in a cost effective means acting as a proof of concept for future industrial applications. Both the safety mat and potential field algorithm have been integrated with a CRS F3 manipulator for conducting validation experiments. We have found that our implementation of the potential field algorithm can successfully avoid single, and multiple obstacles detected by the mat. Moreover, collision avoidance is achieved for both static and dynamic obstacles. Finally, our implementation of the potential field algorithm is capable of preventing local minima entrapment of the manipulator, a problem affecting past implementations.
Recommended Citation
Al Lawati, Mohammed A., "Path Planning of Industrial Manipulators for Dynamic Obstacles using a New Sensory System" (2009). Digitized Theses. 3843.
https://ir.lib.uwo.ca/digitizedtheses/3843