© 2018 JETIR March 2018, Volume 5, Issue 3 www.jetir.org (ISSN-2349-5162)
JETIR1803124 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org 682
FORMATION ALGORITHM IN SWARM ROBOTICS
SOFTWARE
Yazdani Hasan, Dr. M. Deepamalar
1
Research Scholar, SSSUTMS, Sehore
2
Research Guide, SSSUTMS, Sehore
ABSTRACT: Swarm behavior is interesting for humans for multiple reasons. A huge number of related papers was studied by a reading
group to gain knowledge in the field of Swarm Robotics. The main reason is that in a swarm, many simple homogeneous individuals can
meet greater capability when working together, than the individuals themselves are able to achieve. The aim of this investigation was to build
the focal point of the task to arrangement of swarm robots which is communicated in the advancement of a development calculation. An
expansive Simulink display was produced ready to recreate the conduct of 10 robots in a swarm. An arrangement calculation utilizing
potential field powers was actualized ready to make an alluring power from correspondence and an awful power from the estimating of light
force.
Keywords: Swarm Robotics, Formation Algorithm, Homogeneous system
INTRODUCTION
A swarm is both flexible and robust in the sense that it is possible to add or remove a member of the swarm, without any major faults such as
deadlocks or lack of capabilities. Hence there is no leader or specialists in the swarm, and thereby no single point of failure. Big complex
industrial robots can be very expensive, it is very interesting to see whether the swarm collaborative methods can be applied to applications as an
alternative to these. The advantages are the same as from the biology, and smaller simple robots are cheaper to produce and thereby easier and
cheaper to scale the production up and down, as the task demands it. Furthermore the swarm is able to replace a broken member by another
automatically, since the robots are homogeneous.
The main reason is that in a swarm, many simple homogeneous individuals can meet greater capability when working together, than the
individuals themselves are able to achieve. Furthermore a swarm is both flexible and robust in the sense that it is possible to add or remove a
member of the swarm, without any major faults such as deadlocks or lack of capabilities. Hence there is no leader or specialists in the swarm,
and thereby no single point of failure. Parallels can be drawn from the biology behavior to robots. Big complex industrial robots can be very
expensive, it is very interesting to see whether the swarm collaborative methods can be applied to applications as an alternative to these. The
advantages are the same as from the biology, and smaller simple robots are cheaper to produce and thereby easier and cheaper to scale the
production up and down, as the task demands it. Furthermore the swarm is able to replace a broken member by another automatically, since the
robots are homogeneous. Hence it has no single point of failure, resulting in savings in a breakdown. An example of an application for the swarm
method could be in hazardous environments like a rescue recovery or in warzones etc. When a building has crashed or is in any way too
dangerous to enter for humans, a set of swarm robots could search for surviving people using formation. By use of small simple robots, the loss
of a faulty member would not be cost full and since there is no leader it will not affect the overall mission, because all other robots can continue.
The use of swarm collaborative methods is a fairly new subject in robotics and has not really met the market yet. The testing of these methods are
mainly done in simulations, since testing on real world robots is very time consuming and costly. We will start our development by looking into
which project is already presented to learn what have been done. This will help us to solve problems and improve the solution.
This study was to increase the focus of the project to formation of swarm robots which is expressed in the developmentof a formation algorithm.
A large Simulink model was developed able to simulate the behaviour of 10 robots in aswarm. A formation algorithm using potential field forces
was implemented able to create an attractive force from communication and a repulsive force from the measuring of light intensity. From using
both communication and light intensity it was possible to control both the distance between the robots and a common phase alignment. The
development of 10 lightweight swarm robots was represented. A simple implementation of the algorithm from simulation, proved the concept by
controlling the robots to the expected behaviour.
FORMATION ALGORITHM
This paper proposes a new methodology for developing and testing swarming and formation algorithms. This methodology incorporates many
of the good features from the methodologies mentioned in Related Work. It is focused on bridging the gaps between simulation and real the
world as well as increasing understanding of swarms by combining the different approaches and streamlining analysis.