Perpetual Energy Restoration By Multiple Mobile Robots In Circular Sensor Networks

The coverage provided by a network of battery powered sensors degrades over time and eventually disappears if energy is not restored. An important approach to energy restoration is to employ k robots that act as mobile battery chargers. These robots decide where to move next according to a predefined algorithm, called energy restoration strategy, whose effectiveness is measured in terms of: i) the number of nodes that it is able to maintain operational at any given time (Coverage Size), and the time a node battery remains depleted before getting recharged (Disconnection Time). In the case of ring networks (e.g., deployed on the border of a closed region), very simple strategies with near-optimal effectiveness exist for k = 1. In this paper we focus on recharging strategies when k > 1 robots are available. We consider two very simple strategies: 1) SUB-SEGMENT, where the ring is partitioned into segments and one robot is dedicated to each segment; 2) OVERPASS, where the robots, initially at equidistant positions, simply move around the ring charging any node in need, overpassing other robots encountered on the way.We study the two strategies running extensive simulations to assess their effectiveness, varying several network parameters. The results show, among others, that SUB-SEGMENT is always more effective than OVERPASS in terms of coverage, while for disconnection time the effectiveness depends also on other factors, like the number of sensors employed and the size of the ring. Most importantly, the results indicate that SUB-SEGMENT achieves in almost all networks an optimal effectiveness speed-up: the coverage size increases and the disconnection time decreases by a factor of k with respect to the near optimal strategy for a single robot.