Skip to main content

Encapsulación de soluciones basadas en IA para acelerar su adopción

PLEC2023-010240

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D+i en Líneas Estratégicas
Call year 2023
State Plan 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACION

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Semantic diverse redundancy for robust multi-modal safety-critical AI tasks

  • Caro Roca, Martí
  • Brando Guillaumes, Axel
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
Dual Modular Redundancy (DMR) and Triple Modular Redundancy (TMR), often combined with diversity techniques, are widely used in safety-critical systems to achieve fault detection and/or tolerance. Traditional redundancy expects bit-identical outputs, flagging any mismatch as an error. However, emerging AI-based functionalities, such as camera and LiDAR-based object detection, are intrinsically stochastic and require only semantic correctness, allowing some variation in the outputs (e.g., slightly different confidence scores). In this work, we extend our previous semantic redundancy approach – originally developed for camera-based object detection – to LiDAR-based systems, which operate on 3D point cloud data. We propose software-only DMR and TMR schemes that introduce data-level diversity through domain-specific input transformations, preserving semantic meaning while increasing robustness. Our findings demonstrate that the method can be generalized to different AI tasks, helping improve safety and reliability in the different AI components of safety-critical systems., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595. This work is also part of the project PLEC2023- 010240 funded by MICIU/AEI/10.13039/501100011033. M. Caro has received funding under grant PRE2022-102440 funded by MICIU/AEI/10.13039/501100011033 and, by ESF+., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Event monitor validation in high-integrity systems

  • Pujol Torramorell, Roger
  • Vilardell Moreno, Sergi
  • Mezzetti, Enrico|||0000-0002-1886-2931
  • Hassan, Mohamed
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
  • Cazorla Almeida, Francisco Javier
Platforms for modern embedded systems equip an increasing number of high-performance features to provide the required levels of performance. Timing analysis solutions handle the complexity of these platforms by relying on hardware event monitors (HEMs) that provide insightful information about resource utilization and, hence, contention among tasks. As a result, HEMs have become a key element to warrant a safe timing behavior of a system, for which reason they must be validated. While some initial works target HEMs validation, they consider one HEM at a time and focus on those HEMs for which an expert can establish an expected value for relatively small code snippets. In this paper, we propose a methodology for the validation of those HEMs for which a specific expected value cannot be established a priori even for simple cases and, instead, needs to be validated in conjunction with other HEMs. Our method also deals with the natural variability of the HEMs' values in high-performance platforms when collected in different experiments. We illustrate the effectiveness of our proposed technique for validating HEMs related to cache coherence in a relevant platform in the avionics domain., This work has been supported by project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033, the European Union’s Horizon Europe research and innovation programme under grant agreement No. 101070238, and the Natural Sciences and Engineering Research Council of Canada (NSERC). The authors also thank the support of the Department of Research and Universities of the Government of Catalonia to the Smart and Safe Autonomous Systems (SSAS) Research Group, (Code: 2021 SGR 00637)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Safety-relevant AI-based system robustification with neural network ensembles

  • Aldomà Coll, Adrià|||0009-0005-1270-0295
  • Brando Guillaumes, Axel
  • Cazorla Almeida, Francisco Javier
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
Functional safety requirements of AI-based safety-critical applications challenge AI models, whose accuracy can be limited. In this paper, we show how using several cooperative deep learning (DL) models helps to raise global accuracy and reject making a prediction when confidence is below a pre-established threshold., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595. This work has also been supported by the Project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033. Authors thank the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Leveraging image-based transformations to mitigate adversarial attacks in AI-based safety-critical systems

  • Caro Roca, Martí
  • Brando Guillaumes, Axel
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/10.1109/IOLTS65288.2025.11116992, Dual (DMR) and Triple Modular Redundancy (TMR) are widely used techniques to provide fault detection and/or tolerance capabilities in safety-critical systems through – often diverse – redundancy. However, these systems remain vulnerable to adversarial attacks, which can mislead the AI models and lead to severe consequences. In this paper, we propose enhanced DMR and TMR implementations for image-based object detection leveraging image transformations during inference to mitigate the impact of adversarial attacks, hence addressing safety and security concerns simultaneously. Our approach achieves up to 12.9% and 12.2% higher accuracy in adversarial scenarios compared to state-of-the-art solutions in DMR and TMR configurations, respectively., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num.101069595. This work is also part of the project PLEC2023-010240 funded by ICIU/AEI/10.13039/501100011033. M. Caro has received funding under grant PRE2022-102440 funded by MICIU/AEI/10.13039/501100011033 and, by ESF+., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Detecting low-density mixtures in high-quantile tails for pWCET estimation

  • Manau Corderroure, Blau
  • Vilardell Moreno, Sergi
  • Serra Mochales, Isabel
  • Mezzetti, Enrico|||0000-0002-1886-2931
  • Abella Ferrer, Jaume|||0000-0001-7951-4028
  • Cazorla Almeida, Francisco Javier
The variability arising from sophisticated hardware and software solutions in cutting-edge embedded products causes software to exhibit complex execution time distributions. Mixture distributions can happen, with different density (weight), as a result of inherent different features in the execution platform and multiple operational scenarios. In the context of probabilistic WCET (pWCET) analysis based on Extreme Value Theory (EVT), where identical distribution is a pre-requisite, mixtures are typically intercepted by applying stationarity tests on the full sample. Those tests, however, are instructed to detect only mixtures with sufficiently high probability (weight) and disregard low-density mixtures (which are unlikely to be preserved in the high-quantile tail of the sample) as they would prevent any form of stationarity. Nonetheless, low-density mixture distributions can persist and even exacerbate in the tail, and, when not considered, they can impair pWCET estimation in EVT-based approaches, leading to overly pessimistic or optimistic bounds. In this work, we propose TailID, an iterative point-wise approach that builds on the asymptotic convergence of the Maximum Likelihood Estimator (MLE) of the Extreme Value Index (EVI) parameter ξ to detect low-density mixture distributions on high-quantile tails and use this information to steer EVT tail selection. The benefits of the proposed method are assessed on synthetic mixture distributions and real data collected on an industrially representative embedded platform., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595 and project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033 (CAPSUL-IA). Authors also appreciate the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya. This work is partially supported by the Spanish MICIU with project PID2022-137414OB-I00., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Event monitor validation in high-integrity systems

  • Pujol Torramorell, Roger
  • Vilardell Moreno, Sergi
  • Mezzetti, Enrico
  • Hassan, Mohamed
  • Abella Ferrer, Jaume
  • Cazorla Almeida, Francisco Javier
Platforms for modern embedded systems equip an increasing number of high-performance features to provide the required levels of performance. Timing analysis solutions handle the complexity of these platforms by relying on hardware event monitors (HEMs) that provide insightful information about resource utilization and, hence, contention among tasks. As a result, HEMs have become a key element to warrant a safe timing behavior of a system, for which reason they must be validated. While some initial works target HEMs validation, they consider one HEM at a time and focus on those HEMs for which an expert can establish an expected value for relatively small code snippets. In this paper, we propose a methodology for the validation of those HEMs for which a specific expected value cannot be established a priori even for simple cases and, instead, needs to be validated in conjunction with other HEMs. Our method also deals with the natural variability of the HEMs' values in high-performance platforms when collected in different experiments. We illustrate the effectiveness of our proposed technique for validating HEMs related to cache coherence in a relevant platform in the avionics domain., This work has been supported by project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033, the European Union’s Horizon Europe research and innovation programme under grant agreement No. 101070238, and the Natural Sciences and Engineering Research Council of Canada (NSERC). The authors also thank the support of the Department of Research and Universities of the Government of Catalonia to the Smart and Safe Autonomous Systems (SSAS) Research Group, (Code: 2021 SGR 00637)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Leveraging image-based transformations to mitigate adversarial attacks in AI-based safety-critical systems

  • Caro Roca, Martí
  • Brando Guillaumes, Axel
  • Abella Ferrer, Jaume
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/10.1109/IOLTS65288.2025.11116992, Dual (DMR) and Triple Modular Redundancy (TMR) are widely used techniques to provide fault detection and/or tolerance capabilities in safety-critical systems through – often diverse – redundancy. However, these systems remain vulnerable to adversarial attacks, which can mislead the AI models and lead to severe consequences. In this paper, we propose enhanced DMR and TMR implementations for image-based object detection leveraging image transformations during inference to mitigate the impact of adversarial attacks, hence addressing safety and security concerns simultaneously. Our approach achieves up to 12.9% and 12.2% higher accuracy in adversarial scenarios compared to state-of-the-art solutions in DMR and TMR configurations, respectively., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num.101069595. This work is also part of the project PLEC2023-010240 funded by ICIU/AEI/10.13039/501100011033. M. Caro has received funding under grant PRE2022-102440 funded by MICIU/AEI/10.13039/501100011033 and, by ESF+., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Detecting low-density mixtures in high-quantile tails for pWCET estimation

  • Manau Corderroure, Blau
  • Vilardell Moreno, Sergi
  • Serra Mochales, Isabel
  • Mezzetti, Enrico
  • Abella Ferrer, Jaume
  • Cazorla Almeida, Francisco Javier
The variability arising from sophisticated hardware and software solutions in cutting-edge embedded products causes software to exhibit complex execution time distributions. Mixture distributions can happen, with different density (weight), as a result of inherent different features in the execution platform and multiple operational scenarios. In the context of probabilistic WCET (pWCET) analysis based on Extreme Value Theory (EVT), where identical distribution is a pre-requisite, mixtures are typically intercepted by applying stationarity tests on the full sample. Those tests, however, are instructed to detect only mixtures with sufficiently high probability (weight) and disregard low-density mixtures (which are unlikely to be preserved in the high-quantile tail of the sample) as they would prevent any form of stationarity. Nonetheless, low-density mixture distributions can persist and even exacerbate in the tail, and, when not considered, they can impair pWCET estimation in EVT-based approaches, leading to overly pessimistic or optimistic bounds. In this work, we propose TailID, an iterative point-wise approach that builds on the asymptotic convergence of the Maximum Likelihood Estimator (MLE) of the Extreme Value Index (EVI) parameter ξ to detect low-density mixture distributions on high-quantile tails and use this information to steer EVT tail selection. The benefits of the proposed method are assessed on synthetic mixture distributions and real data collected on an industrially representative embedded platform., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595 and project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033 (CAPSUL-IA). Authors also appreciate the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya. This work is partially supported by the Spanish MICIU with project PID2022-137414OB-I00., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Safety-relevant AI-based system robustification with neural network ensembles

  • Aldomà Coll, Adrià
  • Brando Guillaumes, Axel
  • Cazorla Almeida, Francisco Javier
  • Abella Ferrer, Jaume
Functional safety requirements of AI-based safety-critical applications challenge AI models, whose accuracy can be limited. In this paper, we show how using several cooperative deep learning (DL) models helps to raise global accuracy and reject making a prediction when confidence is below a pre-established threshold., The research leading to these results has received funding from the European Union’s Horizon Europe Programme under the SAFEXPLAIN Project (www.safexplain.eu), grant agreement num. 101069595. This work has also been supported by the Project PLEC2023-010240 funded by MICIU/AEI/10.13039/501100011033. Authors thank the support given to the Research Group SSAS (Code: 2021 SGR 00637) by the Research and University Department of the Generalitat de Catalunya., Peer Reviewed, Postprint (author's final draft)