swarm仿真相关资料资源-CSDN文库

共3个文件

pdf：2个

doc：1个

swarm

4星 · 超过85%的资源需积分: 9 172 浏览量 2010-08-07 21:40:02 上传评论收藏 981KB RAR 举报

资源推荐

资源详情

资源评论

收起资源包目录

swarm相关.rar （3个子文件）

swarm.pdf 365KB

swarm中文手册.doc 994KB

jSimpleBug.pdf 89KB

The Swarm Simulation System:

A Toolkit for Building Multi-agent

Simulations

Nelson Minar



Roger Burkhart

Chris Langton

Manor Askenazi

http://www.santafe.edu/pro jects/swarm/

June 21, 1996

Abstract

Swarm is a multi-agent software platform for the simulation of complex adaptive

systems. In the Swarm system the basic unit of simulation is the

swarm,

a collec-

tion of agents executing a schedule of actions. Swarm supp orts hierarchical mo del-

ing approaches whereby agents can be composed of swarms of other agents in nested

structures. Swarm provides ob ject oriented libraries of reusable comp onents for build-

ing mo dels and analyzing, displaying, and controlling exp eriments on those mo dels.

Swarm is currently available as a b eta version in full, free source co de form. It requires

the GNU C Compiler, Unix, and X Windows. More information ab out Swarm can be

obtained from our web pages, http://www.santafe.edu/pro jects/swarm/.



Santa Fe Institute 1399 Hyde Park Rd. Santa Fe, NM 87501.

<nelson@santafe.edu>

Deere & Company John Deere Road Moline, IL 61265

<roger@ci.deere.com>

Santa Fe Institute 1399 Hyde Park Rd. Santa Fe, NM 87501.

<cgl@santafe.edu>

Santa Fe Institute 1399 Hyde Park Rd. Santa Fe, NM 87501.

<manor@santafe.edu>

Swarm has b een developed at the Santa Fe Institute with the support of Grant No. N00014-95-1-1000

from the Oce of Naval Research and GrantNo. N00014-94-1-G014 from the Naval Research Lab oratory,

both acting in coop eration with the Defense Advanced Research Pro jects Agency. Swarm beneted from

earlier support from The Carol O'Donnell Foundation, Mr. and Mrs. Michael Grantham, and the National

Science Foundation. SFI also gratefully acknowledges invaluable supp ort from Deere & Company for this

pro ject.

1 Computational Approaches to Complex Systems

In the sciences, esp ecially in the study of complex systems, computer programs have come

to play an imp ortant role as scientic equipment. Computer simulations | experimental

devices built in software | have taken a place as a companion to physical experimental

devices. Computer models provide manyadvantages over traditional exp erimental metho ds,

but also have several problems. In particular, the actual pro cess of writing software is a

complicated technical task with much ro om for error.

Early in the development of a scientic eld scientists typically construct their own ex-

p erimental equipment: grinding their own lenses, wiring-up their own particle detectors,

even building their own computers. Researchers in new elds have to be adept engineers,

machinists, and electricians in addition to being scientists. Once a eld b egins to mature,

collab orations b etween scientists and engineers lead to the development of standardized, re-

liable equipment (e.g., commercially produced microscop es or centrifuges), thereby allowing

scientists to fo cus on research rather than on to ol building. The use of standardized scien-

tic apparatus is not only aconvenience: it allows one to \divide through" by the common

equipment, thereby aiding the pro duction of repeatable, comparable research results.

In complexity research, at the Santa Fe Institute and elsewhere, we rely heavily on

computers in the course of our investigations. We sp end a lot of time constructing our

own exp erimental apparatus in software, the computational equivalent to blowing our own

glassware. Unfortunately, computer mo deling frequently turns go o d scientists into bad pro-

grammers. Most scientists are not trained as software engineers. As a consequence, many

home-grown computational exp erimental to ols are (from a software engineering p ersp ective)

p oorly designed. The results gained from the use of such to ols can be dicult to compare

with other research data and dicult for others to reproduce b ecause of the quirks and

unknown design decisions in the sp ecic software apparatus. Furthermore, writing software

is typically not a go od use of a highly sp ecialized scientist's time. In many cases, the same

functional capacities are b eing rebuilt time and time again by dierent research groups, a

tremendous duplication of eort.

A subtler problem with custom-built computer models is that the nal software tends to

be very sp ecic, a dense tangle of code that is understandable only to the p eople who wrote

it. Typical simulation software contains a large number of implicit assumptions, accidents

of the way the particular co de was written that have nothing to do with the actual mo del.

And with only low-level source co de it is very dicult to understand the high-level design

and essential comp onents of the mo del itself. Suchsoftware is useful to the p eople who built

it, but makes it dicult for other scientists to evaluate and repro duce results.

In order for computer mo deling to mature there is a need for a standardized set of well-

engineered software to ols usable on a wide variety of systems. The Swarm pro ject aims

to pro duce such to ols through a collaboration between scientists and software engineers.

Swarm is an ecient, reliable, reusable software apparatus for exp erimentation. If successful,

Swarm will help scientists fo cus on research rather than on to ol building by giving them a

standardized suite of software to ols that provide awell equipped software lab oratory.

2 Multi-agent Discrete Event Simulation

The mo deling formalism that Swarm adopts is a collection of indep endentagents interacting

via discrete events. Within that framework, Swarm makes no assumptions ab out the par-

ticular sort of mo del b eing implemented. There are no domain sp ecic requirements such

as particular spatial environments, physical phenomena, agent representations, or interac-

tion patterns. Swarm simulations have b een written for such diverse areas as chemistry,

economics, physics, anthropology, ecology,andp olitical science.

The basic unit of a Swarm simulation is the agent. An agent is any actor in a system,

any entity that can generate events that aect itself and other agents. Simulations consist

of groups of manyinteracting agents. For example, an ecosystem simulation could consist of

agents representing coyotes, rabbits, and carrots. In an economic simulation, agents could

be companies, sto ckbrokers, shareholders, and a central bank. Simulation of discrete inter-

actions b etween agents stands in contrast to continuous system simulations, where simulated

phenomena are quantities in a system of coupled equations.

Agents dene the basic ob jects in the Swarm system, the simulated components. A

schedule of discrete events on these ob jects denes a pro cess occurring over time. In Swarm,

individual actions take place at some sp ecic time; time advances only by events scheduled

at successive times. Aschedule is a data structure that combines actions in the sp ecic order

in which they should execute. For example, the coyote/rabbit simulation could have three

actions: \rabbits eat carrots," \rabbits hide from coyotes," and \coyotes eat rabbits". Each

action is one discrete event: the schedule combines the three in a sp ecic order, e.g. \each

day, have the rabbits eat carrots, then they hide from the coyotes, then the coyotes try to

eat the rabbits". The passage of time is mo deled by the execution of the events in some

sequence.

3 Swarms

The fundamental component that organizes the agents of a Swarm mo del is an ob ject called

a \swarm." A swarm is a collection of agents with a schedule of events over those agents.

For example, a swarm could b e a collection of 15 coyotes, 50 rabbits, a garden with carrots,

and a simple schedule: the rabbits eating the carrots and hiding and the coyotes eating the

rabbits (Figure 1). The swarm represents an entire model: it contains the agents as well as

the representation of time.

In addition to being containers for agents, swarms can themselves be agents. A typical

agent is mo deled as a set of rules, responses to stimuli. But an agent can also itself be a

swarm: a collection of ob jects and a schedule of actions. In this case, the agent's b ehavior

is dened by the emergent phenomena of the agents inside its swarm. Hierarchical mo dels

can be built by nesting multiple swarms. For example, one could build a mo del of a p ond

inhabited by single celled animals. At the highest level, a swarm is created that contains

agents: the swarm represents the pond, and each agent represents one animal. The b ehavior

of cells could be dened simply as some algorithm, but a cell is itself a collection of organelles:

eat hide

Figure 1: ASwarm of Rabbits and Coyotes

a nucleus, mito chondria, endoplasmic reticulum. Another way to represent a cell is as a

swarm of agents, the organelles. Two mo dels are b eing combined: the pond as a swarm of

cells and the cell as aswarm of organelles.

The ability to build mo dels at various levels can b e very p owerful. Swarm allows users to

explicitly build and test multi-level mo dels. A swarm can explicitly represent an emergent

structure, a group of agents b ehaving cohesively as a single agent. Because swarms can

be created and destroyed as the simulation executes, Swarm can be used to mo del systems

where multiple levels of description dynamically emerge.

Another use of multiple swarms is to supp ort the mo deling of agents that themselves build

mo dels of their world. As noted in Hogeweg's MIRROR system [4], in some simulations,

esp ecially those where the agents have a cognitive comp onent, an imp ortant factor in the

system dynamics is an agent's own b eliefs ab out its world. In Swarm, agents can themselves

own swarms, mo dels that an agent builds for itself to understand its own world. For example,

in an economic simulation of a swarm of companies the researcher mightbe interested in the

theory each company has of its comp etitors' actions. To mo del this in the Swarm system,

the model builder would give each company-agent its own swarm: these private swarms

implementeach company's mo del of the world.

The formalism of the swarm is a natural way of encapsulating a simulation: a swarm

simply represents a group of agents and their schedule of activity. The modularity and

comp osability of swarms allows for a exible modeling system. Swarms can be nested to

directly represent multi-level simulations, and they can be used by the agents themselves as

mo dels of their own world.

评论收藏

内容反馈

mercybuaa

2015-04-04

有点杂乱，上传者要是能再整理一下就好了。我下了很多SWARM资料，感觉里面的内容其实都差不多，基本上都是官方提供的。
erospan

2014-12-23

还可以吧，不是很有用。

pan6pcpc

粉丝: 0
资源: 4

swarm仿真相关资料

swarm仿真工具资料

swarm:Swarms相关应用程序的仿真框架

多主体swarm仿真软件包

Swarm仿真中文手册

java_swarm 仿真系统

群智能仿真平台swarm教程

swarm经济仿真源代码.

swarm2.1仿真工具

论文研究-基于PSO的组合拍卖模型及其Swarm仿真.pdf

swarm例程_agent2d

Swarm模型程序《复杂系统建模与仿真》人大出版社第四章介绍的模型的源程序

一个基于Agent的股票市场仿真模型的Swarm实现

Swarm+for+Java仿真及编程实现

swarm for java 经济学仿真

swarm安装教程学习参考网址

一篇关于SWARM的好文章

vaea.rar_Agent_VaEA_java仿真_swarm 平台

基于多Agent计算机仿真实验平台Swarm的综述.pdf

jSimpleBug的教程和代码

Swarm-Robotic-Formation-Control-master_swarm_formationcontrol_fo

最新资源