Ying Wang, Pallege G.D. Siriwardana, and Clarence W. de Silva
Multi-robot systems, machine learning, object transportation, pose estimation
In cooperative multi-robot 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 movable obstacles and it will be subject to uncertainty and unforeseen changes within
the environment. More than one robot may be required for handling heavy and large objects. This paper presents a multi-robot architecture and a machine learning approach for object transportation utilizing multiple cooperative and autonomous mobile robots. A four-layer hierarchical multi-robot architecture is presented, which employs a modiﬁed version of Q-learning for eﬀective robot coordination. As needed in the task, the paper also presents an algorithm for object pose estimation using multi-robot coordination mechanism, by utilizing the laser range ﬁnder and colour blob tracking. The developed techniques are implemented in a multi-robot system
(MRS) in laboratory. Experimental results are presented to demonstrate the eﬀectiveness of the developed MRS and its underlying methodologies.