hal-00650333
https://ensta-bretagne.hal.science/hal-00650333
[UNIV-NANTES] Université de Nantes
[ENSTA-BRETAGNE] ENSTA Bretagne
[CNRS] CNRS - Centre national de la recherche scientifique
[LINA] Laboratoire d'Informatique de Nantes Atlantique
[ENSTA-BRETAGNE-STIC] Département STIC
[ENSIETA-E3I2] Exploitation de l'Information en Environnements Incertains
[TDS-MACS] Réseau de recherche en Théorie des Systèmes Distribués, Modélisation, Analyse et Contrôle des Systèmes
[NANTES-UNIVERSITE] Nantes Université
[UNIV-NANTES-AV2022] Université de Nantes
Processing Interval Sensor Data in the Presence of Outliers, with Potential Applications to Localizing Underwater Robots
Sliwka, Jan
Jaulin, Luc
Ceberio, Martine
Kreinovich, Vladik
[SPI.AUTO] Engineering Sciences [physics]/Automatic
COMM
interval uncertainty
outliers
sensor data
underwater robots
Measurements are never absolutely accurate, the measurement result ex is, in general, different from the actual (unknown) values x of the corresponding quantity. In many practical problems, we only know upper bounds Δ on the measurement errors Δx def = ex x. In such situations, once we know the measurement result, the only conclusion that we can make about the actual value x is that this value belongs to the interval [exΔ; ex+Δ]. There exist many efficient algorithms for processing such interval data. However, these algorithms usually assume that all the measurement results are valid. In reality, due to factors such as sensor malfunction, some measurement results may be way off (outliers), for which the difference between ex and x is much larger than the upper bound Δ on the measurement error. In this paper, we overview the algorithmic problems related to processing interval sensor data in the presence of outliers. Our case study - for which we develop and analyze these algorithms - is localization of underwater robots, a problem in which a significant number of measurement results are outliers
2011-10-09
en
IEEE SMC 2011
Anchorage, United States