You are on page 1of 27

Swarm Intelligence

RITESH GOEL reg.no:0901341154 sic no:09et0150

Guided by:Er. Aditya hota


03-09-2012

Swarming The Definition:


aggregation of similar animals, generally cruising in the same direction Termites swarm to build colonies Birds/ant swarm to find food Bees swarm to reproduce

Swarming :scientist observations

Observations:
Ants find shortest path to food source from nest. Ants deposit pheromone along travelled path which is used by other ants to follow the same path Simple rules for each individual. No central control. Individuals are benefited from the group.

So, what about intelligence?


Social behaviour increases the ability of an individual to adapt. Intelligence arises from interactions among individuals.

From the swarming observations of ant ,swarming intelligence is developed which is utilized for optimization technique(mostly).

NEED FOR NEW TECHNIQUIES


For reducing the complexity.
To design system self-organised. Traditional software and hardware is not efficient. For salesman problem.

SWARM INTELLIGENCE:INTRODUCTION
Any attempt to design algorithms or distributed problem solving devices inspired by the swarming behavior of animals is known as swarm intelligence or swarm intelligence is an artificial intelligence technique based on the study of decentralized ,selforganised systems

HISTORY:
Introduced by Gerardo beni and Jing wang in 1989. In the context of cellular robotic system.

yearly international conference eg:-ANTS 2002-from ant colony to artificial ants.

PROPERTIES
It is composed of many indivisuals. Indivisuals are relatively homogeneous. No central control

Groups Increases self-organisation.


simple rules for indivisuals.

ALGORITHMS OF SWARM INTELLIGENCE:

Ant colony optimization(ACO)

River formation dynamics

Particle swarm optimization(PSO) Stochastic diffusion search


Gravitational search algorithm(GSA)

Continued,,,,
Intelligent Water Drops Charged System Search

ANT COLONY OPTIMIZATION(ACO)


Based on actions of ant colony.

Introduced by Macro Dorigo.(in 1992)


Useful in solving Discrete optimization problem.(salesman and shortest path problem)

ACO: in satellite network

Travelling salesman problem

PARTICLE SWARM OPTIMIZATION


Population based optimization Technique Developed by Dr. Eberhart and Dr. kennedy in 1995. Based on bird flocking or fish schooling. simples rules for the indivisual

eg: PSO
Crowd Simulation: lord of rings Airlines uses swarm theory to simulate passengers boarding a plane. Wavelet based video coding. Android applications

RIVER FORMATION DYNAMICS


Gradient version of ACO It describes how river is formed. water transform the environment ,the altitudes of the places are modified and decreasing gradients are created. this gradients are followed by subsequent drops and created new gradients.

GRAVITATIONAL SEARCH ALGORITHM


Constructed based on law of gravity and mass interactions. It uses newtonion physics every mass see the situations of different masses. transeffering of information between two masses through gravitational force.

APPLICATIONS: SWARM INTELLIGENCE


By US military for contolling vechiles. VLSI floor planning is based on pso. In physical circuit design it is used to reduced the cuts. video and image processing Applications.

IN NEAR FUTURE
NASA is investigating the use of swarm technology for planetary mapping.
European Space Agency is thinking about an orbital swarm for self assembly and interferometer.

ADVANTAGES
EVOLUTIONS Ability to adjust to new situations. highly adoptable.chance of failure is minimal.

DISADVANTAGES
Non-optimal and uncontrollable Unpridictable. Non understable(rarely)

CONCLUSION

References
www.fadooengineers.com

You might also like