Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems
Cooperation in multi-vehicle systems has gained great interest, as it has potential and requires proving safety conditions and integration. To localize themselves, vehicles observe the environment using sensors with various technologies, each prone to faults that can degrade the performance and reli...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/17/7351 |
| _version_ | 1797581848145035264 |
|---|---|
| author | Zaynab El Mawas Cindy Cappelle Mohamad Daher Maan El Badaoui El Najjar |
| author_facet | Zaynab El Mawas Cindy Cappelle Mohamad Daher Maan El Badaoui El Najjar |
| author_sort | Zaynab El Mawas |
| collection | DOAJ |
| description | Cooperation in multi-vehicle systems has gained great interest, as it has potential and requires proving safety conditions and integration. To localize themselves, vehicles observe the environment using sensors with various technologies, each prone to faults that can degrade the performance and reliability of the system. In this paper, we propose the coupling of model-based and data-driven techniques in diagnosis to produce a fault-tolerant cooperative localization solution. Consequently, prior knowledge can guide a discriminative model that learns from a labeled dataset of appropriately injected sensor faults to effectively identify and flag erroneous readings. Going further in security, we conduct a comparative study on learning techniques: centralized and federated. In centralized learning, fault indicators generated by model-based techniques from all vehicles are collected to train a single model, while federating the learning allows local models to be trained on each vehicle individually without sharing anything but the models to be aggregated. Logistic regression is used for learning where parameters are established prior to learning and contingent upon the input dimensionality. We evaluate the faults detection performance considering diverse fault scenarios, aiming to test the effectiveness of each and assess their performance in the context of sensor faults detection within a multi-vehicle system. |
| first_indexed | 2024-03-10T23:13:23Z |
| format | Article |
| id | doaj.art-6bab4b9f60a0425aa9ebd1d9d8e54087 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T23:13:23Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6bab4b9f60a0425aa9ebd1d9d8e540872023-11-19T08:48:42ZengMDPI AGSensors1424-82202023-08-012317735110.3390/s23177351Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot SystemsZaynab El Mawas0Cindy Cappelle1Mohamad Daher2Maan El Badaoui El Najjar3CRIStAL, Centre de Recherche en Informatique Signal et Automatique de Lille, University of Lille, CNRS, UMR 9189, F-59000 Lille, FranceCRIStAL, Centre de Recherche en Informatique Signal et Automatique de Lille, University of Lille, CNRS, UMR 9189, F-59000 Lille, FranceComputer Science Department, Beirut Arab University, Beirut 1107, LebanonCRIStAL, Centre de Recherche en Informatique Signal et Automatique de Lille, University of Lille, CNRS, UMR 9189, F-59000 Lille, FranceCooperation in multi-vehicle systems has gained great interest, as it has potential and requires proving safety conditions and integration. To localize themselves, vehicles observe the environment using sensors with various technologies, each prone to faults that can degrade the performance and reliability of the system. In this paper, we propose the coupling of model-based and data-driven techniques in diagnosis to produce a fault-tolerant cooperative localization solution. Consequently, prior knowledge can guide a discriminative model that learns from a labeled dataset of appropriately injected sensor faults to effectively identify and flag erroneous readings. Going further in security, we conduct a comparative study on learning techniques: centralized and federated. In centralized learning, fault indicators generated by model-based techniques from all vehicles are collected to train a single model, while federating the learning allows local models to be trained on each vehicle individually without sharing anything but the models to be aggregated. Logistic regression is used for learning where parameters are established prior to learning and contingent upon the input dimensionality. We evaluate the faults detection performance considering diverse fault scenarios, aiming to test the effectiveness of each and assess their performance in the context of sensor faults detection within a multi-vehicle system.https://www.mdpi.com/1424-8220/23/17/7351fault tolerancecooperative localizationmachine learningfederated learninginformation theorydata fusion |
| spellingShingle | Zaynab El Mawas Cindy Cappelle Mohamad Daher Maan El Badaoui El Najjar Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems Sensors fault tolerance cooperative localization machine learning federated learning information theory data fusion |
| title | Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems |
| title_full | Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems |
| title_fullStr | Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems |
| title_full_unstemmed | Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems |
| title_short | Comparative Analysis of Centralized and Federated Learning Techniques for Sensor Diagnosis Applied to Cooperative Localization for Multi-Robot Systems |
| title_sort | comparative analysis of centralized and federated learning techniques for sensor diagnosis applied to cooperative localization for multi robot systems |
| topic | fault tolerance cooperative localization machine learning federated learning information theory data fusion |
| url | https://www.mdpi.com/1424-8220/23/17/7351 |
| work_keys_str_mv | AT zaynabelmawas comparativeanalysisofcentralizedandfederatedlearningtechniquesforsensordiagnosisappliedtocooperativelocalizationformultirobotsystems AT cindycappelle comparativeanalysisofcentralizedandfederatedlearningtechniquesforsensordiagnosisappliedtocooperativelocalizationformultirobotsystems AT mohamaddaher comparativeanalysisofcentralizedandfederatedlearningtechniquesforsensordiagnosisappliedtocooperativelocalizationformultirobotsystems AT maanelbadaouielnajjar comparativeanalysisofcentralizedandfederatedlearningtechniquesforsensordiagnosisappliedtocooperativelocalizationformultirobotsystems |