Swarm and entropic modeling for landmine detection robots

Abstract

Even at the dawn of the 21st century, landmines still pose a global threat. Buried just inches below the surface, combatants and noncombatants alike are all at risk of stepping on a mine. Their very nature is such that these furtive weapons do not discriminate, making it an urgent task to tackle the problem. According to the U.S.State Department [1], based on an estimate reported just a few years ago, there are well over 100 million anti-personnel mines around the world. The existence of these passive weapons causes a disruption in the development of already impoverished regions, as well as maiming or killing countless innocent passers-by. Since the ratification of the anti-personnel mine total ban treaty in 1997, their detection, removal, and elimination have become a top priority. Nevertheless, at the current rate, given the manpower and the man-hours that could be dedicated to the removal of these sleeping arms, it would take centuries. The concerns regarding the speed of removal and safety of the disposers eventually bring us to the discussion of the proposed method. Nature already provided good solutions to manage groups of less able beings: fish schools, ant swarms, animal packs, bird flocks, and so on.With the growing desire of humans to create intelligent systems, these biosystems are being thoroughly inspected [3-10] and implemented [11-14] in various studies. In this study a robotic agent is referred to as a drone, the group of robotic agents is referred to as a swarm, and the agent with mapping abilities is referred to as the alpha drone.