Francesca Della Donna
Real-Time Detection of Cognitive Underload for Adaptive Human-Robot Collaboration.
Rel. Alessandro Simeone, Maria Pia Cavatorta. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Della Produzione Industriale E Dell'Innovazione Tecnologica, 2026
Abstract
Real-Time Physiological Digital Twin for Cognitive Load Prediction: A Generative AI Framework Using Wearable Biosignals. Under the Industry 5.0 paradigm, the operator's well-being is repositioned as a design objective on a par with productivity. In human-robot collaboration (HRC), the operator's cognitive state therefore becomes a determinant of joint performance rather than an external factor. This thesis addresses a specific and under-studied risk of repetitive, low-variability HRC tasks, namely cognitive underload: the under-stimulation that erodes vigilance and engagement, as opposed to the overload that most adaptive systems target. The thesis designs, implements and evaluates a real-time closed-loop system that estimates the operator's cognitive state along an underload-optimal-overload continuum from a consumer-grade, four-channel dry-electrode EEG (xMuse) and adapts the collaboration accordingly, so as to keep the operator within a cognitively sustainable band.
The load index is the frontal theta/alpha ratio, smoothed and compared against per-subject robust thresholds derived from a short live calibration
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Informazioni aggiuntive
Corso di laurea
Classe di laurea
Ente in cotutela
Aziende collaboratrici
URI
![]() |
Modifica (riservato agli operatori) |
