Ultra-Wideband (UWB) technology is an emerging low-cost solution for localization in a generic environment. However, UWB signal can be affected by signal reflections and non-line-of-sight (NLoS) conditions between anchors; hence, in a broader sense, the specific geometry of the environment and the disposition of obstructing elements in the map may drastically hinder the reliability of UWB for precise robot localization. This work aims to mitigate this problem by learning a map-specific characterization of the UWB quality signal with a fingerprint semi-supervised novelty detection methodology. An unsupervised autoencoder neural network is trained on nominal UWB map conditions, and then it is used to predict errors derived from the introduction of perturbing novelties in the environment. This work poses a step change in the understanding of UWB localization and its reliability in evolving environmental conditions. The resulting performance of the proposed method is proved by fine-grained experiments obtained with a visual tracking ground truth.

Semi-Supervised Novelty Detection for Precise Ultra-Wideband Error Signal Prediction / Albertin, Umberto; Navone, Alessandro; Martini, Mauro; Chiaberge, Marcello. - ELETTRONICO. - (2024), pp. 3719-3724. (Intervento presentato al convegno 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE) tenutosi a Bari (Italia) nel 28 August 2024 - 01 September 2024) [10.1109/CASE59546.2024.10711427].

Semi-Supervised Novelty Detection for Precise Ultra-Wideband Error Signal Prediction

Albertin, Umberto;Navone, Alessandro;Martini, Mauro;Chiaberge, Marcello
2024

Abstract

Ultra-Wideband (UWB) technology is an emerging low-cost solution for localization in a generic environment. However, UWB signal can be affected by signal reflections and non-line-of-sight (NLoS) conditions between anchors; hence, in a broader sense, the specific geometry of the environment and the disposition of obstructing elements in the map may drastically hinder the reliability of UWB for precise robot localization. This work aims to mitigate this problem by learning a map-specific characterization of the UWB quality signal with a fingerprint semi-supervised novelty detection methodology. An unsupervised autoencoder neural network is trained on nominal UWB map conditions, and then it is used to predict errors derived from the introduction of perturbing novelties in the environment. This work poses a step change in the understanding of UWB localization and its reliability in evolving environmental conditions. The resulting performance of the proposed method is proved by fine-grained experiments obtained with a visual tracking ground truth.
2024
979-8-3503-5851-3
File in questo prodotto:
File Dimensione Formato  
Semi-Supervised_Novelty_Detection_for_Precise_Ultra-Wideband_Error_Signal_Prediction.pdf

accesso riservato

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 768.85 kB
Formato Adobe PDF
768.85 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
CASE_2024___UWB_Novelty_Detection_with_Autoencoders (4).pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 684.69 kB
Formato Adobe PDF
684.69 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2994395