The design and size given the fact that
The evolved system of multi-robots having identical properties and sometimes an identical end goal is known as a swarm of robots. The most common and discussed paradigm of the swarming phenomenon is insects or we would say, a swarm of insects. Insects that live in colonies – ants, bees, wasps, and termites, have fascinated swarm roboticists for quite a few years. Every single insect in a social colony with no intelligence and a centralized control/ supervision seems to have its own agenda, and yet an insect colony is so organized and effective that they perform complex tasks such as foraging of food, migrating etc. In a swarm of robots, each robot is simplified and compact in terms of design and size given the fact that these properties will allow accommodation of more robots per swarm. The individual robot counterparts and the combination of more robots in a swarm is valuable. The research of swarm robotics is based on the theme of simplicity and elegance that resonates in both the designs and algorithms devised for the systems of the robots. The idea that complex macro-level behaviors can emerge from simple local interactions between the agents is what the algorithms are based on. To judge the performance of the swarm robot to an individual robot is its individual entity performing a complex behavior at the macro level. The obvious improvement observed is to cover more area than an individual robot. This is an analogous, for it covers different parts of a search space at once. Another improvement observed is the swarm robotics algorithms do not require the dependence of robots on each other thus the swarm robots are fault tolerant compared to an individual robot. The rest of the swarm can continue performing their actions if a module fails, as though the module never existed, whereas if a failure occurs in a critical component of an individual robot it may become worthless. The most extremely important feature in hostile or complex environments is the robustness. Their effectiveness scales well enough with more number of members in robot swarms. To increase the effectiveness of a swarm, all that has to be done is to add, more robots. But, the improvement of the effectiveness of an individual robot is not clear, because the hardware improvement requires a software that is upgraded which is not in case of swarms. Thus, these properties of a swarm robot can make the multi-robot system suitable for application domains. Although the research of swarm robotics is still rather new and has not produced a swarm of robots that have been used in a practical application. A swarm of robots could cover different locations at once and could disperse in an environment. The maps of the environment that are accurate and developed are superimposed on one another to provide a detailed and extensive map. A swarm of robots can be used for applications such as (a) monotonous and tedious tasks, that includes carrying loads around a plant or warehouse, (b) patrolling that may involve guarding borders, detecting intruders etc. (c) searching and working in hazardous or poisonous environments such as nuclear plants and waste sites, or burning buildings. The rest of this article is organized in the following way: first, the maze generation algorithm is discussed followed by the method used for maze exploration and maze solving/goto goal. Then, the notion of multi-agent systems is induced for maze exploration. Next, the figures of merit used to measure the performance of our solution are defined, followed by results. Finally, conclusions and directions for future work are summarized.