Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning - Inria - Institut national de recherche en sciences et technologies du numérique
Communication Dans Un Congrès Année : 2019

Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning

Bernhard K. Aichernig
  • Fonction : Auteur
  • PersonId : 869582
Roderick Bloem
  • Fonction : Auteur
  • PersonId : 1067256
Masoud Ebrahimi
  • Fonction : Auteur
  • PersonId : 1067257
Martin Horn
  • Fonction : Auteur
  • PersonId : 1067258
Franz Pernkopf
  • Fonction : Auteur
  • PersonId : 873399
Wolfgang Roth
  • Fonction : Auteur
  • PersonId : 1067259
Astrid Rupp
  • Fonction : Auteur
  • PersonId : 1067260
Martin Tappler
  • Fonction : Auteur
  • PersonId : 1067261
Markus Tranninger
  • Fonction : Auteur
  • PersonId : 1067262

Résumé

Models play an essential role in the design process of cyber-physical systems. They form the basis for simulation and analysis and help in identifying design problems as early as possible. However, the construction of models that comprise physical and digital behavior is challenging. Therefore, there is considerable interest in learning such hybrid behavior by means of machine learning which requires sufficient and representative training data covering the behavior of the physical system adequately. In this work, we exploit a combination of automata learning and model-based testing to generate sufficient training data fully automatically.Experimental results on a platooning scenario show that recurrent neural networks learned with this data achieved significantly better results compared to models learned from randomly generated data. In particular, the classification error for crash detection is reduced by a factor of five and a similar F1-score is obtained with up to three orders of magnitude fewer training samples.
Fichier principal
Vignette du fichier
482770_1_En_1_Chapter.pdf (489.85 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02526358 , version 1 (31-03-2020)

Licence

Identifiants

Citer

Bernhard K. Aichernig, Roderick Bloem, Masoud Ebrahimi, Martin Horn, Franz Pernkopf, et al.. Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning. 31th IFIP International Conference on Testing Software and Systems (ICTSS), Oct 2019, Paris, France. pp.3-21, ⟨10.1007/978-3-030-31280-0_1⟩. ⟨hal-02526358⟩
129 Consultations
207 Téléchargements

Altmetric

Partager

More