Go to  Advanced Search

Machine learning-based multi-robot cooperative transportation of objects

Show full item record

Files in this item

Files Size Format Description   View
ubc_2009_fall_siriwardana_pallege.pdf 3.574Mb Adobe Portable Document Format   View/Open
 
Title: Machine learning-based multi-robot cooperative transportation of objects
Author: Siriwardana, Pallege Gamini Dilupa
Degree Master of Applied Science - MASc
Program Mechanical Engineering
Copyright Date: 2009
Publicly Available in cIRcle 2009-11-09
Abstract: Multi robot cooperative transportation is an important research area in the multi robot domain. In the process of object transportation, several autonomous robots navigate cooperatively in either a static or a dynamic environment to transport an object to a goal location and orientation. The environment may consist of both fixed and removable obstacles and it will be subject to uncertainty and unforeseen changes within the environment. Furthermore, more than one robot may be required in a cooperative mode for handling heavy and large objects. These are some of the challenges addressed in the present thesis. This thesis develops a machine learning approach and investigates relevant research issues for object transportation utilizing cooperative and autonomous multiple mobile robots. It makes significant original contributions in distributed multi robot coordination and self deterministic learning for robot decision making, and comes up with an optimal solution to the action selection conflicts of the robots in the cooperative system. This will help to improve the real time performance and robustness of the system. Also, the thesis develops a new method for object and obstacle identification in complex environments using a laser range finder, which is more realistic than the available methods. A new algorithm for object pose estimation algorithm is developed, enabling a robot to identify the objects and obstacles in a multi-robot environment by utilizing the laser range finder and color blob tracking. The thesis develops a fully distributed hierarchical multi-robot architecture for enhanced coordination among robots in a dynamic and unknown environment. It strives to improve the real time performance and robustness. The system consists with three layers. By combining two popular artificial intelligence (Al) techniques such as learning and behavior based decision making, the developed architecture is expected to facilitate effective autonomous operation of cooperative multi-robot systems in a dynamically changing, unstructured, and unknown environment.
URI: http://hdl.handle.net/2429/14700

This item appears in the following Collection(s)

Show full item record

All items in cIRcle are protected by copyright, with all rights reserved.

UBC Library
1961 East Mall
Vancouver, B.C.
Canada V6T 1Z1
Tel: 604-822-6375
Fax: 604-822-3893