doi: 10.4304/jsw.9.10.2586-2597
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centralized Multi-agent Coordination
- 1 Informatics Graduate School and Disaster Prevention Research Institute, Kyoto University, Japan
- 2 The Robotics Institute, Carnegie Mellon University, USA
- 3 Information Sciences Institute, University of Southern California, USA
Abstract
Problem: Multi-agent coordination is an importantissue in the domain of disaster emergency responseoperations where a team of agents (field units or robots) aimsto achieve a joint objective. The responsibility of the IncidentCommander (IC) is to (I) specify an effective strategycomposed of a number of threads (a set of prioritized subproblems),(II) appropriately assign/allocate agents to thesethreads as a strategic decision, and (III) release agents in atimely manner from the assigned threads to adapt a strategicdecision to a new situation. Objective: The purpose of thispaper is to present an intelligent algorithm that assists ahuman in multi-agent coordination by providing two keyfunctions: 1) automatically calculate and present a set offeasible alternatives for selecting a choice as a strategicdecision in a definite time, and 2) autonomously and ina timely manner identify a subset of assigned agents thatshould be released from their threads in order to refinea strategic decision. Method: This algorithm expands adecision tree from a state node in which a thread (or severalthreads) has received a set of new agents from either theIC or a higher thread. Each thread is associated with onelevel of a decision tree with a number of nodes. A threadcalculates a set of efficient coalitions using all the availableagents and generates a new node for each coalition to showwhat agents are allocated to the thread and what agents arereleased into a lower thread. In real-time, this algorithmcontinuously observes and monitors the task environmentto identify a subset of the assigned agents that cannotprovide efficient capabilities for their threads and should bereleased for assignment to other threads. Results: To gatherfurther insight, this paper applied this algorithm for teamcoordination to a simulated search & rescue scenario inan earthquake disaster-affected area where the team’s goalwas to rescue trapped people distributed in five operationalzones. The result was an infinite set of alternative scenariosfor a human-defined strategy. The calculated alternativeswere presented to the IC for selection according to hisintuition or for delegation to the system to determine anoptimal strategy.
Keywords
- Agent assignment problem
- incident comma
How to Cite
Reza Nourjou, Stephen F. Smith, Michinori Hatayama, and Pedro Szekely, "Intelligent Algorithm for Assignment of Agents to Human Strategy in Centralized Multi-agent Coordination," Journal of Software, vol. 9, no. 10, pp. 2586-2597, 2014. https://doi.org/ 10.4304/jsw.9.10.2586-2597
Copyright & License
Copyright © 2014 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).