1: \begin{abstract}

2:   Connectivity maintenance plays a key role in achieving a desired global

3:   behaviour among a swarm of robots. However, connectivity maintenance in realistic environments is hampered by lack of computation resources, low

4:   communication bandwidth, robot failures and unstable links. In this

5:   paper, we propose a novel decentralized connectivity-preserving algorithm

6:   that can be deployed on top of other behaviours to enforce connectivity

7:   constraints. The algorithm takes a set of targets to be reached while

8:   keeping a minimum number of redundant links between robots, with the goal of

9:   guaranteeing bandwidth and reliability. Robots then incrementally build and maintain a communication backbone with the specified

10:   number of links. We empirically study the performance of the algorithm,

11:   analyzing its time to convergence, as well as robustness to faults injected into the backbone robots. Our results statistically

12:   demonstrate the algorithm's ability to preserve the desired connectivity

13:   constraints and to reach the targets with up to 70 percent of individual robot failures in the

14:   communication backbone.

15: 

16: \keywords{ Swarm robotics \and Connectivity Maintenance \and Fault-Tolerance.}

17: \end{abstract}

18: