Efficiency Assessment of Performance of Decentralized Autonomous Multi-Robot Systems
Arvin Agah* and George A. Bekey**
* Bio-Robotics Division, Mechanical Engineering Laboratory, AIST- MITI, 1-2 Namiki, Tsukuba, 305 JAPAN
(Institute for Robotics and Intelligent Systems, Computer Science Department University of Southern California, Los Angeles, CA 90089, U.S.A. )
** Institute for Robotics and Intelligent Systems, Computer Science Department University of Southern California, Los Angeles, CA 90089, U.S.A.
This paper presents a new methodology for the efficiency assessment of task performance of decentralized autonomous multi-robot system. This formulation considers the types of the robots (in terms of sensing, action, and control), the description of the environment (the world), the facts (rules) of the world, and the specifications of the tasks to be performed. The performance efficiency is defined in terms of the total time required for completing the task, total energy requirement, and the comparison of the final results and the desired results. A robot colony simulator was used to perform a number of experiments, measuring the task performance efficiency, of a colony of simulated robots which perform specific tasks in a virtual world. The experimental results are presented in this paper. This paper also describes a team of four robots designed and fabricated in hardware. The physical robots were used successfully to validate the results from the simulated colony.