Applications based on groups of self-organized mobile robots and - more generically - agents are becoming pervasive in communication, monitoring, traffic and transportation systems. Their advantage is the possibility of providing services without the existence of a previously defined infrastructure and with a high degree of autonomy. On the other hand, physical agents, in general, are prone to failures, adding uncertainty and unpredictability in the environments in which they operate. Therefore, a robust topology regarding failures is an imperative requirement. In this paper, we show that mechanisms based solely on connectivity maintenance are not enough to obtain a sufficiently resilient network, and a robustness-oriented approach is necessary. Thus, we propose a local combined control law that aims at maintaining the overall network connectivity while improving the network robustness via actions that reduce vulnerability to failures that might lead to network disconnection. The combined control law performance was validated from two perspectives: as a reactive and as a proactive mechanism. As a reactive mechanism, it was able to accommodate ongoing failures and postpone or avoid network fragmentation. As a proactive mechanism, the network topology was able to evolve from a potentially vulnerable topology w.r.t. failures to a more robust one.
|Data di pubblicazione:||2016|
|Titolo:||Improving the fault tolerance of multi-robot networks through a combined control law strategy|
|Autori:||Ghedini, Cinara; Ribeiro, Carlos H. C.; Sabattini, Lorenzo|
|Digital Object Identifier (DOI):||10.1109/RNDM.2016.7608289|
|Data del convegno:||13-15 September 2016|
|Nome del convegno:||Resilient Networks Design and Modeling (RNDM), 2016 8th International Workshop on|
|Luogo del convegno:||Halmstadt, Sweden|
|Titolo del libro:||Proceedings of 2016 8th International Workshop on Resilient Networks Design and Modeling (RNDM)|
|Appare nelle tipologie:||Relazione in Atti di Convegno|
I documenti presenti in Iris Unimore sono rilasciati con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia, salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris