Swarm Robotics a survey and its Applications

Swarm Robotics a survey and its Applications

Swarm Robotics

Swarm Robotics is basically a conceptualization of the multiple robots acting as a coordinating system which consists of mostly simple physical Robots. The desired mass behavior of the robot system emerges from the interaction of the robots with each other as well as with environment. The approach egresses from the concept of Artificial Swarm Intelligence as well as from Entomology that is the biological study of insects like honeybees/drones, ant colonies, Bird flocking, Fish Schooling, animal herding, bacterial growth, and microbial intelligence.

Swarm Robotics a survey and its Applications

Definition

It is the study and design of robots with Swarm Intelligence, their physical body and controlling behavior. It is inspired by but not limited by the nascent behavior of the observed in the social insects called the Swarm Intelligence (SI).

Swarm Intelligence(SI)

It can be defined as the collective behavior of the decentralized, self-organized systems, natural or artificial. The concept was basically employed in the work on Artificial Intelligence(AI). The expression Swarm Intelligence was introduced by Gerardo Beni and Jing Wang in 1989, in the circumstance of cellular robotic systems. Comparatively simple individual rules can produce a galactic set of intelligent Swarm Behavior. A system of constant feed-back is the key component for communication among the swarm robots. Generally this communication is established using a local communication mode like wireless transmission technology like Radio Frequency (RF) or Infra Red (IR) technology between individual robots.

Swarm Robotics a survey and its Applications

Models of Swarm Behavior

Boids (Reynolds 1987)

Boids is an artificial life program, formulated by Craig Reynolds in 1986, which simulates the flocking behaviour of birds. The term ”boid” refers to a bird like object. The boids is an example of emergent behavior and the complexity of behavior arises from the interaction of simple robots adhering to simple rules.

The simple rules of boid world are:

Separation: guide to avoid local flock-mate crowding.

Alignment: guide towards the average heading of local flock-mate.

Cohesion: guide towards the average position i.e. Center of Mass of local flock-mates.

More complex functions can be added such as obstacle avoidance and goal seeking.

Self-propelled particles (Vicsek et al. 1995)

Self-propelled particles (SPP), also termed as the Vicsek model, was introduced in 1995 by Vicsek as a special case Reynolds’ boid model. SPP swarm robots is a collection of particles that move with a constant speed but respond to a random discomposure by adopting at each time increment of the average direction of motion of the other particles in their local neighborhood. These models are turning out to be universal and robust.

Meta-heuristics

In Computer Science and Mathematical optimization a Meta-heuristics is a higher level procedure or heuristics developed to find, generate and select an algorithm that provides a good solution to an optimization problem especially with incomplete information or limited computational capability. Metaheuristics is basically a sampled set of solutions that is to large to be completely sampled. It may take a few assumptions for a problem to be solved so they are useful for variety of problems. Evolutionary algorithms (EA), ant colony optimization (ACO),  particle swarm optimization (PSO), and their variants dominate the field of nature-inspired metaheuristics.

There are some articles and papers that are already published based on metaphor based meta-heuristics. Following is the list of published papers and articles.

Stochastic diffusion search (Bishop 1989)

-Finding Symmetry

Swarm Robotics a survey and its Applications

 

 

Ant colony optimization (Dorigo 1992)

Swarm Robotics a survey and its Applications

Particle swarm optimization (Kennedy, Eberhart & Shi 1995)

Swarm Robotics a survey and its Applications

 

Goals and Applications

A miniature model and cost are the key factors in making of the swarm robots. As a large number of robots are to be made so every individual team member should be as simple as possible. This build up should encourage a swarm intelligent behavior instead of an individual behavior.

The Swarm research is the most captivating field for researchers around the world. Such a system can be very useful for variety of applications. Each member of a swarm team must be very much less demanding regarding the resources and more power and energy efficient.

Examples of swarm Robots

An example of a swarm robot is a LIBOT that involves the low cost robot system for outdoor swarm robotics that work on GPS sensors. The system can be provided with indoor facility using Wireless Fidelity (Wi-Fi) connection as the GPS sensors become weak inside buildings.

Applications

The swarm robotics have many applications. The swarm robotics supply tasks like miniaturization and (Nano – robotics, micro- robotics) , the distributed sensing task in micro-machinery and the medical operations of human body.

The Rescue operation

The Rescue operation is a most promising application of swarm robotics. The team of robots can be sent at the rescue places where the human rescue workers cannot reach safely. These robots can detect the presence of life using infrared sensors. For this purpose the swarm individual robot should be cheap and simple in design.

Swarm Robotics a survey and its Applications

The swarm robotics can also be used in the mining and agricultural foraging i.e. searching for wild food resources.

There is another sizable set of applications can be used using the swarm micro air flying robots. It can be used in the motion detecting system in laboratory conditions.

Ariel Vehicles

Hundreds of micro Ariel vehicles can also be controlled using the Global Navigation Satellite System ( GNSS ) such as GPS system in outdoor conditions but also can be localized using Wi-Fi. One such system is developed called Shooting Star, a quad-copter designed for light shows by Intel. It is basically formed of Styrofoam plastics and has built in Light Emitting Diodes(LEDs) for display purpose. A single computer and operator can control a large number of Shooting Stars. About more than four million color combinations are made using built-in LEDs, with the systems micro-controller programming algorithms optimizing the choreography and the flight paths of the bots. These are basically Drone Displays that are the multiple lighted drones for artistic aerial show at night.

Swarm Robotics a survey and its Applications

Furthermore, the Swarm Intelligence is very useful for various Military operations. For Information sake, some developed countries are already following the Swarm Intelligence for the military applications. The U.S. military is already investigating the swarm techniques for controlling the unnamed vehicles. European Space Agency is thinking of orbital swarm system for interferometry and self-assembly. NASA is investigating the swarm technology for planetary mapping. Even in India, the Swarm Robotics will be soon adopted for various military operations. The SI is also taking into account in Medical operations for controlling the cancer tumors in body. A paper is already been published in 1992 by  M. Anthony Lewis and George A. Bekey in this regard. The Swarm Intelligence is already been used in Data Mining.

Programming of Swarm Robots

The researchers from the Sheffield Robotics have invented a method of automatically programming and controlling of about 600 swarm robots to complete a specified task simultaneously. This automatic programming system reduces human error by automatically fixing the program bugs and thus making the programming part more user friendly and reliable. The previous methods are much more tedious for programmers which undergoes traditional trial and error methods. These methods are more time consuming and unpredictable for coding in real world that often results in the undesirable behavior of the swarm robot system. This is an actually a system of great advantage in the Driver-less Car System. The research studies are still going on regarding this field.

Here at Eckovation You can go through the basics of Robotics and enhance your capability to handle the modern technologies. You can go through Robotics Online Course,  Machine Learning Course, Programming for All Course for the build up of your amazing project.

Please post your comments in the below comments section. It helps us to stay motivated! 🙂

Don't miss out!
Subscribe To Our Newsletter

Learn new things. Get an article everyday.

Invalid email address
Give it a try. You can unsubscribe at any time.

Comments

comments