Spectro-Temporal Recurrent Neural Network for Robotic Slip Detection with Piezoelectric Tactile Sensor - CEA - Commissariat à l’énergie atomique et aux énergies alternatives Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Spectro-Temporal Recurrent Neural Network for Robotic Slip Detection with Piezoelectric Tactile Sensor

Résumé

In this paper, we present a novel tactile-based slippage-detection method for robotics applications, utilizing a single piezoelectric sensor. The method combines spectral analysis (FFT) and deep learning (GRU) for improved efficiency and adaptability. We implement an automated data-collection process with accurate and unbiased labels of slip events. The proposed method was evaluated through an ablation study, to characterize the influence of different parameters. The results showed a high classification accuracy of 98.70% at 100Hz and detection delays of 8.5 ± 23.7ms, demonstrating the relevance of our spectro-temporal pipeline. The proposed method has the potential to enhance the performance of robotic systems and increase their reliability in robotic grasping applications.
Fichier principal
Vignette du fichier
Slip_detection_AIM23_post.pdf (1.26 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

cea-04176729 , version 1 (03-08-2023)

Identifiants

Citer

Théo Ayral, Saifeddine Aloui, Mathieu Grossard. Spectro-Temporal Recurrent Neural Network for Robotic Slip Detection with Piezoelectric Tactile Sensor. 2023 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM), Jun 2023, Seattle (USA), United States. pp.Pages 573-578, ⟨10.1109/AIM46323.2023.10196263⟩. ⟨cea-04176729⟩
15 Consultations
69 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More