摘要
为了提高多智能体系统在动态不确定性环境中的灵活性、鲁棒性和自适应性,文中提出了一种基于群智能的多智能体系统SIMAS,其主要思想是把群智能思想应用到非群体的多智能体系统中。整个SIMAS系统采用自顶向下和自底向上设计相结合的方法建立系统体系结构,应用群智能的阈值模型进行多智能体的自组织任务分配。最后通过机器人围捕仿真实验,验证了该项工作的有效性和可行性。
The introduction of the full paper discusses relevant matters and then proposes the new design mentioned in the title, which we believe is more effective than previous ones and which is explained in sections 1 and 2. Their core consists of： ＂In order to improve the flexibility, robustness and adaptability for multi agent system in dynamic and uncertain environment, SIMAS architecture is introduced to design and model swarm-intelligence based multi agent system. This architecture applies swarm intelligence theory to non-swarm multi agent system. SIMAS is established by top-down task partition and bottom-up task allocation, and the threshold model of swarm-intelligence theory is used to achieve self-organized task allocation. ＂ Simulation experiments are carried out on multi-robot pursuing platform; their results, presented in Figs. 3 through 6, and their analysis preliminarily show that our work is indeed effective.
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2012年第1期124-128,共5页
Journal of Northwestern Polytechnical University
基金
国家航空科学基金(20100753022)
西北工业大学基础研究基金(JC201121)资助
关键词
群智能
多智能体
体系结构
自组织
任务分配
algorithms, analysis, artificial intelligence, architecture, design, dynamics, efficiency, experiments, improvement, robots, models, multi agent systems, robustness （ control systems）, schematic diagrams, simulation, topology, uncertain systems
swarm intelligence, task allocation
