An introduction to swarming robotics: application development trends

Abstract

Animals help to sustain the environment's life cycle and ecosystem. Without human intervention, these creatures carry out their 'spontaneous routine' jobs and contribute towards balance in nature. Any natural system that congregates as a result of some form of collective intelligence of nature is also known as swarm intelligence (SI). This metaphor inspires a variety of techniques to solve the problem of calculating, in most cases dealing with optimization problems and has sparked interest amongst scientists. It is very trying for a new researcher to understand the whole concept of swarming robotics (SR) and optimization algorithm (i.e. realizing the idea from animal's perception to the SR application). In addition, the existing algorithms are computationally complicated, difficult to be understood by beginners as there are too many parameters. Thus, in this paper, we simplify the three branches of the main applications which are frequently used for SI namely: (1) optimization and networks design, (2) prediction and forecasting, and (3) SR. This paper summarizes the basic understanding overview of swarming robotics and discusses their basic concepts and principles.

Morphogen diffusion algorithms for tracking and herding using a swarm of kilobots

“…Thus, to rigorously validate swarm pattern formation algorithms, it is essential to test them on large-scale real robotic systems. Note that as there exist obvious trade-offs between the cost and the capability of the robot platforms such as movement and sensing capability [8][9][10], an adequate robotic test-bed needs to be carefully chosen depending on the requirements of a specific pattern formation algorithm.…”

Morphogenetic Self-Organization of Collective Movement without Directional Sensing

“…Thus, to rigorously validate swarm pattern formation algorithms, it is essential to test them on large-scale real robotic systems. Note that as there exist obvious trade-offs between the cost and the capability of the robot platforms such as movement and sensing capability [7][8][9], an adequate robotic test-bed needs to be carefully chosen depending on the requirements of a specific pattern formation algorithm.…”