Distributed algorithms for self-spreading of robotic networks over unknown complex areas in GPS-denied environments

This paper presents new algorithms for area coverage by mobile robotic swarms in complex and unknown environments. The robots are silent and not aware of the team’s size, do not discern between each other, lack access to a positioning system, and cannot play distinct roles and so should be driven by a common control rule. On the positive side, they determine the relative positions of the objects, including the boundary of the handled area and obstacles, within a given finite sensing range and have access to a common direction. The proposed algorithms are attributed the physics-inspired virtual force-based (VFB) approach. To highlight their benefits, the paper reports on comparative analysis of seven methods, including the above two ones and an essential group of well-established algorithms that follow VFB approach and are suited to handle the examined scenario. The analysis is carried out using a whole range of various metrics, which capture different aspects of the performance quality. Moreover, a new criterion of coverage uniformity is introduced and justified. Extensive computer simulations in complex scenes have shown that the proposed algorithms demonstrate the best performance in terms of coverage uniformity and percentage.