DocumentCode :
1895447
Title :
The necessity of average rewards in cooperative multirobot learning
Author :
Tangamchit, Poj ; Dolan, John M. ; Khosla, Pradeep K.
Author_Institution :
Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA, USA
Volume :
2
fYear :
2002
fDate :
2002
Firstpage :
1296
Abstract :
Learning can be an effective way for robot systems to deal with dynamic environments and changing task conditions. However, popular single-robot learning algorithms based on discounted rewards, such as Q learning, do not achieve cooperation (i.e., purposeful division of labor) when applied to task-level multirobot systems. A task-level system is defined as one performing a mission that is decomposed into subtasks shared among robots. We demonstrate the superiority of average-reward-based learning such as the Monte Carlo algorithm for task-level multirobot systems, and suggest an explanation for this superiority.
Keywords :
Monte Carlo methods; learning (artificial intelligence); multi-robot systems; Monte Carlo algorithm; average rewards; cooperative multirobot learning; dynamic environments; robot systems; task-level multirobot systems; Artificial intelligence; Centralized control; Control systems; Costs; Delay; Feedback; Learning systems; Monte Carlo methods; Multirobot systems; Robot control;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Automation, 2002. Proceedings. ICRA '02. IEEE International Conference on
Print_ISBN :
0-7803-7272-7
Type :
conf
DOI :
10.1109/ROBOT.2002.1014721
Filename :
1014721
Link To Document :
بازگشت