Ant Colony Optimization

Ant Colony Optimization
Minor Research Report

Student Name: Shivali Palicha
Student ID: 13111305

Table of Contents
Take Away / Brief Summary………………………………………………..3
List of Figures………………………………………………………………...5
Chapter 1 Introduction…………………………………………………...6
Chapter 2 Background / History……………………………………..…9
Chapter 3 Theory………………………………………………………...10
Chapter 4 Advantages and Disadvantages……………………………20
Chapter 5 Further Direction……………………………………………21
References…………………………………………………………………..23
TAKE AWAY / BRIEF SUMMARY

Figure 1

Attention of Human beings has been attracted since long time by social complex behavior exhibited by Ants. Most noticeable behavior visible to us is the formation of so called ant street.
It is Biologically Inspired

Ant Colony algorithm can be run nonstop and can adjust to changes in real time.
Optimization:

Figure 3

· Given a graph with two specified verticals X and Y , find shortest path from X to Y
Shortest path problem, Polynomial
· Given a set of cities and pairwise distances, find a shortest possible tour.
Travelling Salesperson Problem (TSP)
· Given a sequence of amino acids of a protein, find the structure of the protein.
Protein structure prediction problem
In real world, ants (initially) wander randomly, and once find food, return to their colony while laying down pheromone trails.
If other ants find path taken by previous ants, they are likely not to keep travelling random, but instead will follow the trail laid by earlier ants. They return over same path and hence reinforcing if they found food.
Over a period of time, the pheromone trails starts to evaporate, thus developing less strength to attract ants. The more and longer time it takes for an ant to travel down the path and return back again, the more time the pheromones have to evaporate. Comparing to longer path, short path gets travelled and marched over more frequently and thus the pheromone density becomes higher on shorter

Max-Min Ant System was invented in sub sequent years by Hoos and Stutzle. Ant Systems was researched and developed by Gambardella Dorigo in 1997.
Virtual trail accumulated on all given path segments. Path selected at random initially from possible paths from starting node. Ant continue journey to next node and selecting next path until reaches starting node. Finished tour is a solution and is analyzed for optimality.
Ant Colony optimization combines user given heuristics in the desire/ expectation of obtaining a more efficient & cost-effective procedure.
If one sees from the artificial intelligence (AI) perspective, ACO algorithms are one of the most victorious strands of swarm intelligence
ACO is very robust and can be applied to job shop scheduling problem with minimal of changes.
If two bridges have same length then ants tend to cross and connect and meet towards use of one of two bridges.

Chapter 3: THEORY
Shortest route is found using pheromone trails which ants deposit whenever they travel, as a form of indirect communication message.

Figure 5
ACO algorithm: The ACO algorithm comprises two Key stages: Tour construction and Pheromone