|
Zug, Sebastian; Dietrich, André
Examination of Fusion Result Feedback for Fault-Tolerant and Distributed Sensor Systems (Konferenzbeitrag)
IEEE International Workshop on Robotic and Sensors Environments (ROSE 2010), Phoenix, AZ, USA, 2010.
(Abstract | BibTeX | Schlagwörter: Distributed Applicaton, Programming Abstraction, Smart Sensing)
@inproceedings{EOS-2010.000-ZD,
title = {Examination of Fusion Result Feedback for Fault-Tolerant and Distributed Sensor Systems},
author = {Sebastian Zug and André Dietrich},
year = {2010},
date = {2010-01-01},
booktitle = {IEEE International Workshop on Robotic and Sensors Environments (ROSE 2010)},
address = {Phoenix, AZ, USA},
abstract = {Future distributed applications will combine mobile and stationary sensor actuator entities as part of smart environments. Hence, a various number of intelligent sensor systems and sensor types will be available for mobile robots. We designed a programming abstraction – the Smart Abstract Entity – to deal with the diversity of data formats, measurement units, individual uncertainties, etc., in such applications. One focus of this approach was fault-tolerance in a hybrid manner. We combined centralized and decentralized mechanisms in our core concept. In this paper we discuss one aspect of the fault-tolerance concept – the feedback transmission of available fusion results to the sensor nodes, for a verification of its current state. We use a mobile robot scenario to examine the impact of this feedback approach for the position estimation and discuss parameters as well as methods for refinement.},
keywords = {Distributed Applicaton, Programming Abstraction, Smart Sensing}
}
Future distributed applications will combine mobile and stationary sensor actuator entities as part of smart environments. Hence, a various number of intelligent sensor systems and sensor types will be available for mobile robots. We designed a programming abstraction – the Smart Abstract Entity – to deal with the diversity of data formats, measurement units, individual uncertainties, etc., in such applications. One focus of this approach was fault-tolerance in a hybrid manner. We combined centralized and decentralized mechanisms in our core concept. In this paper we discuss one aspect of the fault-tolerance concept – the feedback transmission of available fusion results to the sensor nodes, for a verification of its current state. We use a mobile robot scenario to examine the impact of this feedback approach for the position estimation and discuss parameters as well as methods for refinement.
|