Multi-agent hierarchical architecture modeling kinematic chains employing continuous RL learning with fuzzified state space

Karigiannis, JN; Tzafestas, CS

dc.contributor.author	Karigiannis, JN	en
dc.contributor.author	Tzafestas, CS	en
dc.date.accessioned	2014-03-01T02:45:39Z
dc.date.available	2014-03-01T02:45:39Z
dc.date.issued	2008	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32316
dc.subject	Action Selection	en
dc.subject	Artificial Intelligent	en
dc.subject	Control Architecture	en
dc.subject	Degree of Freedom	en
dc.subject	Knowledge Acquisition	en
dc.subject	multi-agent architecture	en
dc.subject	Numerical Experiment	en
dc.subject	Reinforcement Learning	en
dc.subject	Robot Control Architecture	en
dc.subject	State Space	en
dc.subject	Task Model	en
dc.subject	Multi Agent System	en
dc.subject.other	Agents	en
dc.subject.other	Coordination reactions	en
dc.subject.other	Hierarchical systems	en
dc.subject.other	Kinematics	en
dc.subject.other	Knowledge acquisition	en
dc.subject.other	Manipulators	en
dc.subject.other	Motion control	en
dc.subject.other	Reinforcement	en
dc.subject.other	Robotics	en
dc.subject.other	Robots	en
dc.subject.other	4 degrees of freedoms	en
dc.subject.other	Action selections	en
dc.subject.other	Closed kinematic chains	en
dc.subject.other	Continuous problems	en
dc.subject.other	Continuous spaces	en
dc.subject.other	Control architectures	en
dc.subject.other	Control methodologies	en
dc.subject.other	Cooperation and coordinations	en
dc.subject.other	Degree of similarities	en
dc.subject.other	Degrees of freedoms	en
dc.subject.other	Dexterous manipulators	en
dc.subject.other	Hierarchical architectures	en
dc.subject.other	Hierarchical robots	en
dc.subject.other	Kinematic chains	en
dc.subject.other	Methodological frameworks	en
dc.subject.other	Multi agents	en
dc.subject.other	Multi-agent architectures	en
dc.subject.other	Numerical experiments	en
dc.subject.other	Reinforcement learning methods	en
dc.subject.other	Robotic systems	en
dc.subject.other	State spaces	en
dc.subject.other	Task models	en
dc.subject.other	Multi agent systems	en
dc.title	Multi-agent hierarchical architecture modeling kinematic chains employing continuous RL learning with fuzzified state space	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/BIOROB.2008.4762862	en
heal.identifier.secondary	http://dx.doi.org/10.1109/BIOROB.2008.4762862	en
heal.identifier.secondary	4762862	en
heal.publicationDate	2008	en
heal.abstract	In the context of multi-agent systems, we are proposing a hierarchical robot control architecture that comprises artificial intelligence (AI) techniques and traditional control methodologies, based on the realization of a learning team of agents in a continuous problem setting. In a multiagent system, action selection is important for cooperation and coordination among the agents. By employing reinforcement learning (RL) methods in a fuzzified state-space, we accomplish to design a control architecture and a corresponding methodology, engaged in a continuous space, which enables the agents to learn, over a period of time, to perform sequences of continuous actions in a cooperative manner, in order to reach their goal without any prior generated task model. By organizing the agents in a nested architecture, as proposed in this work, a type of problem-specific recursive knowledge acquisition is attempted. Furthermore, the agents try to exploit the knowledge gathered in order to be in position to execute tasks that indicate certain degree of similarity. The agents correspond in fact to independent degrees of freedom of the system, and achieve to gain experience over the task that they collaboratively perform, by exploring and exploiting their state-to-action mapping space. A numerical experiment is presented in this paper, performed on a simulated planar 4 degrees of freedom (DOF) manipulator, in order to evaluate both the proposed hierarchical multiagent architecture as well as the proposed methodological framework. It is anticipated that such an approach can be highly scalable for the control of robotic systems that are kinematically more complex, comprising multiple DOFs and potentially redundancies in open or closed kinematic chains, particularly dexterous manipulators. © 2008 IEEE.	en
heal.journalName	Proceedings of the 2nd Biennial IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2008	en
dc.identifier.doi	10.1109/BIOROB.2008.4762862	en
dc.identifier.spage	716	en
dc.identifier.epage	723	en