Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Human Comfortability Index Estimation in Industrial Human-Robot Collaboration Task

Celal Savur, Jamison Heard, Ferat Sahin

TL;DR

This work tackles measuring human comfort in industrial human-robot collaboration by deriving a Comfortability Index (CI) and Uncomfortability Index (unCI) from physiological signals. It combines a circumplex emotion model with arousal-valence coordinates and a KDE-based distribution to map psycho-physiological cues into CI/unCI, supported by two ML pipelines (Random Forest and Functional Neural Network) to estimate underlying emotions. The approach is validated through a controlled assembly task with varied robot behaviors, showing that CI aligns with calm affect and unCI with anxiety, while KDE provides a non-parametric alternative. Real-time estimation (CIES) demonstrates feasibility for adaptive safety and comfort-aware HRC, though limitations include lab conditions and artifact sensitivity, suggesting directions for broader deployment and more direct affect measurements.

Abstract

Fluent human-robot collaboration requires a robot teammate to understand, learn, and adapt to the human's psycho-physiological state. Such collaborations require a computing system that monitors human physiological signals during human-robot collaboration (HRC) to quantitatively estimate a human's level of comfort, which we have termed in this research as comfortability index (CI) and uncomfortability index (unCI). Subjective metrics (surprise, anxiety, boredom, calmness, and comfortability) and physiological signals were collected during a human-robot collaboration experiment that varied robot behavior. The emotion circumplex model is adapted to calculate the CI from the participant's quantitative data as well as physiological data. To estimate CI/unCI from physiological signals, time features were extracted from electrocardiogram (ECG), galvanic skin response (GSR), and pupillometry signals. In this research, we successfully adapt the circumplex model to find the location (axis) of 'comfortability' and 'uncomfortability' on the circumplex model, and its location match with the closest emotions on the circumplex model. Finally, the study showed that the proposed approach can estimate human comfortability/uncomfortability from physiological signals.

Human Comfortability Index Estimation in Industrial Human-Robot Collaboration Task

TL;DR

This work tackles measuring human comfort in industrial human-robot collaboration by deriving a Comfortability Index (CI) and Uncomfortability Index (unCI) from physiological signals. It combines a circumplex emotion model with arousal-valence coordinates and a KDE-based distribution to map psycho-physiological cues into CI/unCI, supported by two ML pipelines (Random Forest and Functional Neural Network) to estimate underlying emotions. The approach is validated through a controlled assembly task with varied robot behaviors, showing that CI aligns with calm affect and unCI with anxiety, while KDE provides a non-parametric alternative. Real-time estimation (CIES) demonstrates feasibility for adaptive safety and comfort-aware HRC, though limitations include lab conditions and artifact sensitivity, suggesting directions for broader deployment and more direct affect measurements.

Abstract

Fluent human-robot collaboration requires a robot teammate to understand, learn, and adapt to the human's psycho-physiological state. Such collaborations require a computing system that monitors human physiological signals during human-robot collaboration (HRC) to quantitatively estimate a human's level of comfort, which we have termed in this research as comfortability index (CI) and uncomfortability index (unCI). Subjective metrics (surprise, anxiety, boredom, calmness, and comfortability) and physiological signals were collected during a human-robot collaboration experiment that varied robot behavior. The emotion circumplex model is adapted to calculate the CI from the participant's quantitative data as well as physiological data. To estimate CI/unCI from physiological signals, time features were extracted from electrocardiogram (ECG), galvanic skin response (GSR), and pupillometry signals. In this research, we successfully adapt the circumplex model to find the location (axis) of 'comfortability' and 'uncomfortability' on the circumplex model, and its location match with the closest emotions on the circumplex model. Finally, the study showed that the proposed approach can estimate human comfortability/uncomfortability from physiological signals.
Paper Structure (23 sections, 4 equations, 15 figures, 2 tables)

This paper contains 23 sections, 4 equations, 15 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. AV Domain Transformation
  5. Circumplex Model
  6. Kernel Density Estimation (KDE)
  7. Emotion Estimation from Physiological Signals
  8. Random Forest Regressor
  9. Functional Neural Network
  10. Experiments
  11. Subjective Data Collection
  12. Data Labelling
  13. Data Collection Devices
  14. Data Collection System
  15. Results and Discussion
  16. ...and 8 more sections

Figures (15)

  • Figure 1: Proposed physiological computing system where "R" represents a robot and "H" represents a human.
  • Figure 2: Arousal and Valence circumplex model and a few discrete emotional classes and their locations Russell1980Toisoul2021.
  • Figure 3: Two approaches: the first approach is a circumplex model and the second one is a Kernel Density Estimation (KDE) model.
  • Figure 4: Given a $z$ axis on the circumplex model, how CI or unCI is calculated.
  • Figure 5: Emotion estimation from physiological signal.
  • ...and 10 more figures