Los Angeles, CA
March 31, 2009 to April 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.234
Multi-agent simulation system based on reinforcement learning algorithm is a micro-individual acts of modeling and simulation methods, which have wide applicability, distribution, intelligent and interactive features etc. Firstly, studying on reinforcement learning algorithm, and then analysis and design the multi-agent simulation system structure, multi-agent system main modules, the implementation of the definition and finally, carefully design the multi-agent simulation system software, and multi-agent simulation collective system simulation and surrounded the location gathered from the space simulation experiment, the results showed that: Construct a multi-agent simulation system based on reinforcement learning algorithm, achieve real-time simulation of multi-agene, and multi-agent to get effect quickly, and to quickly construct surrounded conduct by mobile groups, the conduct of the system to achieve the global optimum effect.
reinforcement learning, multi-agent simulation, collective siege, gathered from space
Shu Da Wang, Shuo Ning Wang, Wei Ping Zhang, "Study on Multi-agent Simulation System Based on Reinforcement Learning Algorithm", CSIE, 2009, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE, 2009 WRI World Congress on Computer Science and Information Engineering, CSIE 2009, pp. 523-527, doi:10.1109/CSIE.2009.234