EXPERIENCED TASK-BASED MULTI ROBOT TASK ALLOCATION

Yıl 2017, Cilt: 18 Sayı: 4, 864 - 875, 31.10.2017

H. Hilal Ezercan Kayır

https://doi.org/10.18038/aubtda.340101

Öz

In multi robot system applications, it
is possible that the robots transform their past experiences into useful
information which will be used for next task allocation processes by using
learning-based task allocation mechanisms. The major disadvantages of multi-robot
Q-learning algorithm are huge learning space and computational cost due to
generalized state and joint action spaces of robots. In this study, experienced
task-based multi robot task allocation approach is proposed. According to this
approach, robots believe to be experienced about the tasks most frequently
done. Robots prefer to do these tasks rather than the inexperienced ones. Then,
robots refuse to execute inexperienced tasks over time. This means that the
system has reduced learning space. The proposed approach plays a crucial role
to achieve required system performance and provides effective solutions to
learning space dimensions.  The
effectiveness of the proposed algorithm is demonstrated by simulations on multi-robot
task allocation problem.

Anahtar Kelimeler

Multi robot task allocation, Multi-agent Q-learning, Adaptive system architecture

Kaynakça

• Arkin R C. Behavior-Based Robotics. Cambridge, MIT Press, 1998.
• Mataric M J. Reinforcement learning in multi-robot domain. Autonomous Robots 1997; 4(1): 73-83.
• Ezercan Kayır H H, Parlaktuna O. Strategy planned Q-lerning approach for multi-robot task allocation. Procs of ICINCO 2014; 2: 410-416.
• Sutton R S, Barto A G. Reinforcement Learning: an Introduction. Cambridge, MIT Press, 1998.
• Yang E, Gu D. Multiagent Reinforcement Learning for Multi-Robot Systems: a Survey. Technical Reports of the Dept. of Computer Science, Univ. of Essex, 2004.
• Gerkey B P, Mataric M J. A formal analysis and taxonomy of task allocation in multi-robot systems. International Journal of Robotics Research 2004; 23(9): 939-954.
• Jones E G, Dias M B, Stentz A. Learning-enhanced market-based task allocation for oversubscribed domains. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems; 2007; San Diego, CA, USA: pp. 2308-2313.
• Dias M B, Zlot R M, Kaltra N, Stentz A. Market-based multirobot coordination: a survey and analysis. Proceedings of the IEEE 2006; 94(7): 1257-1270.
• Zlot R, Stentz A. Market-based multirobot coordination for complex tasks. Int. Journal of Robotics Research, Special Issue on the 4th Int. Conf. on Field and Service Robotics 2006; 25(1): 73-101.
• Gerkey B P, Mataric M J. Sold!: auction methods for multi robot coordination. IEEE Transactions on Robotics and Automation 2002; 18(5): 758-768.
• Mosteo A R, Montano L. Comparative experiments on optimization criteria and algorithms for auction based multi-robot task allocation. Proceedings of the IEEE International Conference on Robotics and Automation 2007; pp. 3345-3350.
• Kaleci B, Parlaktuna O, Ozkan M, Kırlık G. Market-based task allocation by using assignment problem. IEEE Int. Conf. on Systems, Man, and Cybernetics; 2008; pp. 135-14.
• Hatime H, Pendse R. A comparative study of task allocation strategies in multi-robot systems. IEEE Sensors Journal 2013; 13(1): 253-262.
• Russel S, Norvig P. Artificial Intelligence a Modern Approach. New Jersey, Prentice Hall, 2003.
• Buşoniu L, Babuška L, Schutter B. A comprehensive survey of multiagent reinforcement learning. IEEE Trans. on Systems, Man, and Cybernetics 2008; 38(2): 156-172.
• Kaelbling L P, Littman M L, Moore A W. Reinforcement learning: a survey”, Journal of Artificial Intelligence Research 1996; 4: 237-285.
• Watkins C J C H. Learning from delayed rewards. PhD Thesis, University of Cambridge, UK, 1989.
• Watkins C J, Dayan P. Q-learning, Machine Learning 1992; 8.
• Hu J, Wellman M P. Nash Q-learning for general sum games. Journal of Machine Learning Research 2003; 4: 1039-1069.
• Boutlier C. Planning learning and coordination in multiagent decision processes. Proceedings of the 6th Conference on Theoretical Aspects of Rationality and Knowledge, TARK '96; 1996; pp. 195-210.
• Ezercan Kayır H H. ContQL-MRTA: Continuous-learned Q-learning based task allocation approach in multi robot systems. TOK 2015; 10-12 Eylül 2015; Denizli, Turkey. (article in Turkish with an abstract in English).