2. INTRODUCTION
Biologically inspired algorithms are becoming powerful in modern numerical
optimization, among these biology-derived algorithms, the multi-agent meta-
heuristic algorithms such as particle swarm optimization form hot research topics
in the start-of-the-art algorithm development in optimization and other
application.
Particle swarm optimization has many similarities with genetic algorithms, but it is
much simpler because it does not use mutation/crossover operators. Instead, it
uses the real-number randomness and the global communication among the
swarming particles.
The Genetic Algorithm transforms a population of individual objects, each with an
associated fitness value, into a new generation of the population using the Darwin
principle of individual of reproduction and survival of the fittest and naturally
occurring genetic operation such as a cross over (recombination) and mutation.
Each individual in the population represents a possible solution to a given
problem.
The firefly algorithm (FA) is a meta-heuristic algorithm, inspired by the flashing
behaviour of fireflies .
3. PROBLEM STATEMENT
To do a comparative study of nature inspired
algorithm in between Firefly Algorithm and the
Particle Swarm Optimization using Michalewicz
function for the numerical optimization. Thereafter
we have compared Particle swarm optimization,
Genetic Algorithm and Ant Colony Optimization
using the real world benchmark problem that is
Travelling salesman problem
4. BENEFITS/NOVELITY OF THE
APPLICATION
Nature-inspired algorithms are among the most powerful algorithms for
optimization.
The PSO algorithm searches the space of the objective functions by adjusting the
trajectories of individual agents, called particles, as the piecewise paths formed by
positional vectors in a quasi-stochastic manner.
The Firefly Algorithm can be modified to solve multi objective optimization
problems. In addition, the application of firefly algorithms in combination with other
algorithms may form an exciting area for further research.
Evolutionary algorithm optimizers are global optimization methods and scale well to
higher dimensional problems. They are robust with respect to noisy evaluation
functions, and the handling of evaluation functions which do not yield a sensible
result in given period of time is straightforward.
8. TESTING REQUIRED
Type of Test Will it be EXPLANATIONS Software Component
performed?
Requirement Yes Requirement testing is testing the Manual work, need to
Testing requirements whether they are plan out all the software
feasible or not. Because a project requirements, time needed
depends on a number of factors like to develop, technology to
time, resources, budget etc. Before be used etc.
we start working on a project it‘s
important to test these requirements.
Unit Yes Testing by which individual units of Manual check is required
source code are tested to determine
if they are fit for use.
Integration Yes Testing wherein individual Compiling various classes
components are combined and tested and testing them as one
as a group. single code.
9. LIMITATION OF THE APPLICATION
Genetic algorithm applications in controls which are
performed in real time are limited because of random
solutions and convergence, in other words this means that
the entire population is improving, but this could not be
said for an individual within this population. Application
doesn’t consider the camera angle.
Firefly algorithm suffers greatly from diminishing returns
once the swarm size grows past a certain point, or when
the solution space grows immensely large. Any extra
installation cost is not considered in the final cost
displayed.
10. FINDINGS
Firefly Algorithm :
Parallelization - The parts of computation that cannot be parallized are the portions that determine
the next movement of a firefly. They are dependent on the current position of the firefly. As a
stochastic algorithm, it is impossible to predict future moves of a firefly so there is no opportunity
to allow pre-cacheing of values.
Scalability : The primary drawback to FA is the required communication between fireflies.
This algorithm requires every firefly in the swarm to know the fitness of every other firefly in the
swarm at the end of each iteration of moving and updates. However, if we use a central node to act
as a communications hub, we can then drastically reduce the number of communications required at
the end of each generation of the algorithm.
Particle Swarm Optimization :
Parallelization : A particle will only have to touch its own information and the stand-
alone nature of the algorithm is only broken by the communication between a particle
and coordinating node to update the Swarm Best Fitness.
Scalability : The only information that is required to be disseminated to the rest of the
swarm is the swarm's best fitness and associated parameters. That will then allow the
swarm particles to independently update their own positions and velocity with no other
outside information required
11. CONCLUSION
We studied new firefly algorithm and analysed its similarities and differences with
particle swarm optimization. We then implemented and compared these algorithms.
Our simulation results for finding the global optima of various test functions suggest
that particle swarm often outperforms traditional algorithms such as genetic
algorithms, while the new firefly algorithm is superior to PSO in terms of both efficiency
and success rate. This implies that FA is potentially more powerful in solving NP-hard
problems which will be investigated further in future studies. The basic firefly
algorithm is very efficient, but we can see that the solutions are still changing as the
optima are approaching. It is possible to improve the solution quality by reducing the
randomness gradually. A further improvement on the convergence of the algorithm is to
vary the randomization parameter α so that it decreases gradually as the optima are
approaching. These could form important topics for further research. Furthermore, as a
relatively straightforward extension, the Firefly Algorithm can be modified to solve
multi objective optimization problems. In addition, the application of firefly algorithms
in combination with other algorithms may form an exciting area for further research.
This time we implemented genetic algorithm, and other nature inspired algorithms
which are particle swarm optimisation and ant colony optimisation on real-time
problem ,TRAVELLING SALESMAN PROBLEM which is a NP hard problem and many
algorithms have been implemented and we found out the PSO is the best out of all
three implemented.
12. FUTURE WORK
Use more varied functions for comparing the
efficiency of different evolutionary algorithms.
Implement more different evolutionary algorithms
like SFLA and Ant colony optimization and
compare the accuracy of results.
Try to implement these algorithms using more
datasets.
13. REFERENCES
[1] Bonabeau E., Dorigo M., Theraulaz G., Swarm Intelligence: From Natural to Artificial
Systems. Oxford University Press, (1999)
[2] Deb. K., Optimisation for Engineering Design, Prentice-Hall, New Delhi, (1995).
[3] Gazi K., and Passino K. M., Stability analysis of social foraging swarms, IEEE Trans. Sys.
Man. Cyber. Part B - Cybernetics, 34, 539-557 (2004).
[4] Goldberg D. E., Genetic Algorithms in Search, Optimisation and Machine Learning,
Reading, Mass.: Addison Wesley (1989).
[5] Kennedy J. and Eberhart R. C.: Particle swarm optimization. Proc. of IEEE International
Conference on Neural Networks, Piscataway, NJ. pp. 1942-1948 (1995).
[6] Kennedy J., Eberhart R., Shi Y.: Swarm intelligence, Academic Press, (2001).
[7] Passino K. M., Biomimicrt of Bacterial Foraging for Distributed Optimization, University
Press, Princeton, New Jersey (2001).
[8] Shilane D., Martikainen J., Dudoit S., Ovaska S. J., A general framework for statistical
performance comparison of evolutionary computation algo- rithms, Information Sciences:
an Int. Journal, 178, 2870-2879 (2008).
