History and Technology of Swarm Robotics

Swarm robotics is the study of design and governing of multiple robots. The term ‘Swarm’ stands for a bunch of communicating individuals who want to achieve the same task. They are influenced from Swarm intelligence. The principle motivation for swarm robotics originates from the perception of social animals like ants, bees etc. It deals with the decentralized coordination of group of robots using local logic and endorses scalability by making use of local communication through radio signals or infrared. The intrinsic parallelism (to accomplish task concurrently) and redundancy make multi-robot frameworks more robust than single robot frameworks. Robustness is the probability of failure of the system is very low and the swarm robotics has high robustness. Miniaturization and price are the chief aspects in building swarm robotics. In the scope of research, working in a real environment resolves many issues rather than simulations. However due to some restraining factors the swarm robotics are yet to be implemented in real-world analysis. The Microbotics or Nanorobotics has many applications in the progressing medical field.

Before introducing Swarm robotics (SR) concept there was a system that referred to as Multi agent Robot (MAS) which was based on the concept of several robots working together. It made the foundation for communication, coordination, task planning, and disseminated agent structures in robots. Conspicuously, SR is more advanced than MAS, SR could scale up to thousands of robots whereas MAS systems could not support large numbers.

As we discussed earlier the conduct of Social insects influenced the development of SR. The comprehensive practices of insects were considered for quite a while weird and puzzling parts of science. It has been studied that the understanding of swarm is distributed to all the individuals and one robot can’t accomplish its task without the other counterparts. Organization rises up out of the collaborations between the individuals and amongst individuals and the environment. These connections are engendered all through the colony and in this manner the colony can explain errands that couldn't be tackled by a sole individual. These aggregate practices are characterized as self-organizing. It depends on the combination of these four regulations they are – positive response, negative response, uncertainty and multiple communications. Self-organizing speculations, acquired from material science, can be utilized to clarify how social insects display complex aggregate conduct that rises up out of communications of individuals behaving normally.

According to Erol Şahin, in order to get a clear picture about Swarm robotics and so as to have a better understanding and be able to separate it from other multi-robot systems the robots should follow this criteria-

1. The robots of the swarm must be independent robots, ready to detect and incite in a genuine environment.
2. The number of robots in the swarm must be vast or possibly the control rules permit it.
3. Robots must be homogeneous. There can exist distinctive sorts of robots in the swarm, yet these gatherings must not be excessively many.
4. The robots must be inadequate or wasteful regard to the fundamental task they need to understand, this is, they have to team up keeping in mind the end goal to succeed or to enhance the performance.
5. Robots have just nearby communication and detecting abilities. It guarantees the coordination is disseminated, so adaptability ends up one of the properties of the framework.

Classification

According to Dudek et al. the Swarm robots are classified based on their properties such as

• Group size - Number of robots.
• Signal Conveyance range - Maximum range.
• Conveyance topology - of the robots in the communication range, those which can be communicated with.
• Conveyance Bandwidth – Amount of information the robots can send each other.
• Collective reconfigurability – The coordination rate that can be altered.
• Course ability – The calculation model used by robots.
• Collated composition – Are Robots homogeneous or heterogeneous.

Current state of art of Swarm Robotics

Several swarm intelligence researches are influenced by the swarms occurring in the nature. The concept of Swarm intelligence was proposed in 1980s. It has now turned into an interdisciplinary frontier and focal point of numerous disciplines including Artificial intelligence, sociology etc.