Adaptive Synchronization of Robotic Sensor Networks

Kasim Sinan Yildirim (1) , Önder Gürcan (2)

1 Ege University - EGE (Izmir, Turkey)
2 DILS (CEA, LIST) - Département Ingénierie Logiciels et Systèmes

The main focus of recent time synchronization research is developing power-efficient synchronization methods that meet pre-defined accuracy requirements. However, an aspect that has been often overlooked is the high dynamics of the network topology due to the mobility of the nodes. Employing existing flooding-based and peer-to-peer synchronization methods, are networked robots still be able to adapt themselves and self-adjust their logical clocks under mobile network dynamics? In this paper, we present the application and the evaluation of the existing synchronization methods on robotic sensor networks. We show through simulations that Adaptive Value Tracking synchronization is robust and efficient under mobility. Hence, deducing the time synchronization problem in robotic sensor networks into a dynamic value searching problem is preferable to existing synchronization methods in the literature.

Domaines: Intelligence artificielle [cs.AI]; Calcul parallèle, distribué et partagé [cs.DC]; Modélisation et simulation

Kasim Sinan Yildirim, Önder Gürcan. Adaptive Synchronization of Robotic Sensor Networks. Workshop on Robotic Sensor Networks at CPSWeek2014, Apr 2014, Berlin, Germany. ⟨cea-01806904⟩