Accéder directement au contenu Accéder directement à la navigation
Nouvelle interface
Communication dans un congrès

Reachability Set Analysis of Closed-Loop Nonlinear Systems with Neural Network Controllers

Arash Sadeghzadeh 1 Pierre-Loic Garoche 1 
1 LII - ENAC - Equipe Informatique Interactive
ENAC - Ecole Nationale de l'Aviation Civile
Abstract : A forward reachability analysis method for the safety verification of nonlinear systems controlled by neural networks is presented. The proposed method relies on abstracting the activation functions in the neural networks (NN) by quadratic constraints (QCs) resorting to local sector bounds. To tackle the system nonlinearity, the nonlinear model is embedded into a linear parameter varying (LPV) representation. An outer-approximation of the forward reachable set of the closed-loop system is obtained using semidefinite programming. A numerical example clearly demonstrates the applicability of the proposed method. Comparison with some available methods reveals that the provided approach may potentially lead to less conservative results.
Type de document :
Communication dans un congrès
Liste complète des métadonnées
Contributeur : Loetitia MOYA Connectez-vous pour contacter le contributeur
Soumis le : lundi 14 novembre 2022 - 15:25:59
Dernière modification le : mercredi 16 novembre 2022 - 16:13:06





Arash Sadeghzadeh, Pierre-Loic Garoche. Reachability Set Analysis of Closed-Loop Nonlinear Systems with Neural Network Controllers. 2022 American Control Conference (ACC), Jun 2022, Atlanta, United States. pp.2289-2294, ⟨10.23919/ACC53348.2022.9867572⟩. ⟨hal-03851617⟩



Consultations de la notice