[1]刘佰龙,张汝波,史长亭.随机扰动下多源群体觅食系统建模与仿真[J].智能系统学报,2008,3(4):342-348.
L IU Bai-long,ZHANG Ru-bo,SH I Chang-ting.Modeling and simulating the foraging system in multi-source groups with random disturbances[J].CAAI Transactions on Intelligent Systems,2008,3(4):342-348.
点击复制
L IU Bai-long,ZHANG Ru-bo,SH I Chang-ting.Modeling and simulating the foraging system in multi-source groups with random disturbances[J].CAAI Transactions on Intelligent Systems,2008,3(4):342-348.
随机扰动下多源群体觅食系统建模与仿真
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
3
期数:
2008年第4期
页码:
342-348
栏目:
学术论文—智能系统
出版日期:
2008-08-25
- Title:
- Modeling and simulating the foraging system in multi-source groups with random disturbances
- 文章编号:
- 1673-4785 (2008) 04-0342-07
- 关键词:
- 群集智能; 自组织行为; 觅食模型; Starlogo仿真
- Keywords:
- swarm intelligence; self2organization; foraging; modeling; Starlogo simulation
- 分类号:
- TP18
- 文献标志码:
- A
- 摘要:
- 群集智能是指复杂的集体智能来自简单个体之间以及个体同环境之间的相互作用. 通常对群集智能的研究主要借助于群居生物行为的观察. 蚁群觅食行为是研究简单个体产生复杂行为的一个典型的例子. 首先建立群体觅食宏观序参数模型. 模型考虑了食物源的量和分布以及环境噪声对个体决策的随机影响. 给出2个食物源下系统模型的数值解,表明在较大的噪声影响下,系统有一定的概率会脱离最优解,到达次优解. 在Starlogo仿真平台下的实验结果表明,觅食蚂蚁的数量同任务完成时间以及碰撞频率之间呈现出幂指数关系. 这对自组织系统和群集智能的研究有一定的理论意义,并可以用来指导设计更加有效、适应、可靠的智能系统.
- Abstract:
- Swarm intelligence ( SI) is artificial intelligence based on observable collective behavior of decentralized, self2organized systems, otherwise known as social animals, and is gaining more attention from researchers. SI sys2 tems are typ icallymade up of a population of simp le agents interacting locallywith one another and with their envi2 ronment. The agents follow very simp le rules, and although there is no centralized control structure dictating how individual agents should behave, local interactions between such agents lead to the emergence of comp lex global be2 havior. Specially, the foraging behavior of ant colonies have be viewed as a p rototyp ical examp le of how comp lex group behavior can arise from simp le individual behaviors. In order to study the feature of self2organization in SI, first a macroscop ic serial parametric model for flock foragingwas established, in which the richness and distribution of food sourceswere considered as well as the stochastic effects of a noisy environment. Numerical solutions were given for systematic modelswith two food sources. These showed that, in an environmentwith great noise, the op ti2 mal solution may not be found and a second2best solution may instead be reached. Simulations on the Starlogo p lat2 form showed a power law relationship between the number of ants and comp letion time aswell as the flux of forag2 ers. The work p resented here may imp rove the understanding of self2organization and swarm intelligence. It can al2 so be used to design more efficient, adap tive, and reliable intelligent systems
- 参考文献/References:
-
[ 1 ]BONABEAU E, DOR IGO M, THERAULAZ G. Swarm in2 telligence: from natural to artificial systems [M ]. New York:Oxford University Press, 1999.
[ 2 ] PRATT S C. Recruiment and other communication behavior in the ponerine ant ectatomma ruidum[ J ]. Ethology, 1989, 81: 3132331.
?[ 3 ]MARK RUSSELL EDELEN. Swarm intelligence and stig2 mergy: robotic imp lementation of foraging behaviorp [D ]. Washington,DC: Graduate School of the University ofMary2 land, 2003.
[ 4 ] HOLLDOBLER B, W ILSON E O. The ants [M ]. Cam2 bridge, Mass,USA: Harvard University Press, 1990.
?[ 5 ] RAMOS V, FERANDES C, ROSA A. Social cognitive map s, swarm collective percep tion and distributed search on dynamic landscapes[ J ]. Journal of NewMedia in Neu2 ral an Cognitive Science, 2005, (5) : 25231.
?[ 6 ] TSANKOYA D D, GEORGIEVA V S. From local actions to global tasks: simulation of stigmergy based foraging behavior [C ] / / IEEE International Conference on Intelligent Sys2 tems. [ S. l. ]. 2004: 532358.
[ 7 ] SUGAWARA K, WATANABE T. A study on foraging be2 havior of simp le multi2robot system [ C ] / / IEEE 2002 28 th Annual Conference of the Industrial Electronics Society. Se2 ville, Spain, 2002: 308523090.
?[ 8 ] EDELSTEIN2KESHET L, WATMOUGH J, ERMENTROUT G. Trail following in ants: individual p roperties determine population behaviour[ J ]. Behavioral Ecology and Sociobiol2 ogy, 1995, 36: 1192133.
[ 9 ] EDELSTEIN2KESHET L. Simp le models for trail2following behavior: trunk trails versus individual foragers[ J ]. Journal ofMathematical Biology, 1994, 32: 3032328.
?[ 10 ]GAZIV, PASSINO KM. Stability analysis of social fora2 ging swarms[ C ] / / IEEE Transactions on Systems, Man, and Cybernetics, 2004, 34 (1) : 5392556.
?[ 11 ]N ICOL IS S C, DENEUBOURG J L. Emerging patterns and food recruitment in ants: an analytical study [ J ]. Journal of Theoretical Biolog, 1999, 198: 5752592.
[ 12 ]BECKERS R, DENEUBOURG J L, GOSS S. Modulation of trail laying in the ant lasius niger ( hymenop tera: for2mi2 cidae) and its role in the collective selection of a food source [ J ]. Journal of InsectBehavior, 1993 (6) : 7512759.
[ 13 ]N ICOL IS S C. DETRA IN C, DEMOL IN D, et al. Op ti2 mality of collective choices: a stochastic app roach [ J ]. Bulletin ofMathematicalBiology, 2003, 65: 7952808.
[ 14 ] SUGAWARA K, SANO M, YOSH IHARAT I, et al. Wa2 tanabe foraging behavior of multi2robot system and emer2 gence of swarm intelligence[C ] / / IEEE International Con2 ference on Systems, Man, and Cybernetics. Tokyo, Ja2 pan, 1999: 2572262.
[ 15 ]BEN I G, WANG J. Distributed robotic systems and swarm intelligence[C ] / / Proceedings of IEEE International Con2 ference on Robotics and Automation. Sacramento, CA, USA , 1991.
?[ 16 ]WONG D Y, QUTUB A, HUNT C A. Modeling transport kineticswith Starlogo[C ] / / Proceedings of the 26 th Annu2 al International Conference of the IEEE EMBS. San Fran2 cisco, CA, USA, 2004: 8452848.
- 相似文献/References:
-
[1]王 龙,伏 锋,陈小杰,等.复杂网络上的群体决策[J].智能系统学报,2008,3(2):95.
WANG Long,FU Feng,CHEN Xiao-jie,et al.Collective decision-making over complex networks[J].CAAI Transactions on Intelligent Systems,2008,3():95.
备注/Memo
收稿日期: 2008-05-07.
作者简介: 刘佰龙, 男, 1983 年生, 博士研究生,主要研究方向为智能机器人、群集智能；张汝波,男, 1963年生,教授,博士生导师,中国人工智能学会智能机器人专业委员会委员、黑龙江省人工智能学会理事、《智能系统学报》编委、IEEE 会员、中国电子学会高级会员、黑龙江省计算机学会机器人专业委员会委员、黑龙江省神经科学学会人工智能与医学工程专业委员会委员,主要研究方向为智能机器人与智能控制、机器学习与计算智能、智能信息处理. 发表学术论文100多篇,被SCI、EI、 ISTP收录60余篇次,出版专著及教材5部.
史长亭,男, 1980年生,讲师,主要研究方向为智能机器人与智能控制、软件可靠性.
通信作者:刘佰龙. E-mail: lbl624@163. com.
更新日期/Last Update:
2009-05-18