Transcript Swarm Intelligence

SWARM INTELLIGENCE
Sumesh Kannan
Roll No 18
Introduction

Swarm intelligence (SI) is an artificial intelligence technique
based around the study of collective behavior in decentralized,
self-organized systems.

Introduced by Beni & Wang in 1989.

Typically made up of a population of simple agents.

Examples in nature : ant colonies, bird flocking, animal herding
etc.
Intelligent Agents

An agent is anything that can be viewed as perceiving its
environment through sensors and acting upon that
environment through effectors.
Rational Agents

Rationality - expected success given what has been perceived.

Rationality is not omniscience.

Ideal rational agent should do whatever action is expected to
maximize its performance measure, on the basis of the
evidence provided by the percept sequence and whatever
built-in knowledge the agent has.

Factors on which Rationality depends




Performance measure (degree of success).
Percept sequence (everything agent has perceived so far).
Agents knowledge about the environment.
Actions that agent can perform.
Structure of IA

Agent = Program + Architecture

A Simple Agent Program.
Simple Reflex Agents

Follows Condition-Action Rule.

Needs to perceive its environment completely.
Model Based Agents

Need not perceive the environment completely.

Maintains an internal state.

Internal states should be updated.
Goal Based Agents

Makes decisions to achieve a goal.

More flexible.
Utility Based Agents

A complete specification of the utility function allows rational
decisions in two kinds of cases.


Many goals, none can be achieved with certainty.
Conflicting goals.
Environment

Accessible vs. Inaccessible

Deterministic vs. Non-deterministic

Episodic vs. Non-episodic

Static vs. Dynamic

Continuous vs. Discreet
An Environment Procedure
Ant Colony Optimization (ACO)

First ACO system- Marco Dorgo,1992

Ants search for food.

The shorter the path the greater the pheromone left by an ant.

The probability of taking a route is directly proportional to the
level of pheromone on that route.

As more and more ants take the shorter path, the pheromone
level increases.

Efficiently solves problems like vehicle routing, network
maintenance, the traveling salesperson.
Particle Swarm Optimization (PSO)

Population based Stochastic optimization technique.

Developed by Dr. Eberhart & Dr. Kennedy in 1995.

The potential solutions, called particles, fly through the
problem space by following the current optimum particles.

Applied in many areas: function optimization, artificial neural
network training, fuzzy system control etc.
Swarm Robotics

Most important application area of Swarm Intelligence

Swarms provide the possibility of enhanced task performance,
high reliability (fault tolerance), low unit complexity and
decreased cost over traditional robotic systems

Can accomplish some tasks that would be impossible for a
single robot to achieve.

Swarm robots can be applied to many fields, such as flexible
manufacturing systems, spacecraft, inspection/maintenance,
construction, agriculture, and medicine work
Applications

Massive (Multiple Agent Simulation System in Virtual
Environment) Software.


Developed Stephen Regelous for visual effects industry.
Snowbots

Developed Sandia National laboratory.
References
http://en.wikipedia.org
http://www.swarmbots.com
http://www.siprojects.com
Thank you