Force Myography-Based Human Robot Interactions via Deep Domain Adaptation and Generalization

• Published in: Sensors (Basel, Switzerland) Vol. 22; no. 1; p. 211
• Main Authors: Zakia, Umme, Menon, Carlo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 29.12.2021; MDPI
• Subjects: Acclimatization; Adaptation; Adaptation, Physiological; Algorithms; applied force estimation in dynamic motion; Calibration; Collaboration; domain adaptation; domain generalization; Electrodes; force myography technique; Humans; Investigations; Machine learning; Mechanical Phenomena; Muscle contraction; Myography; Neural networks; pretrained model; Prostheses; Robotics; Robots; Sensors; Signal processing; transfer learning; pretrained model; transfer learning; force myography technique; applied force estimation in dynamic motion; domain adaptation; domain generalization
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s22010211

Estimating applied force using force myography (FMG) technique can be effective in human-robot interactions (HRI) using data-driven models. A model predicts well when adequate training and evaluation are observed in same session, which is sometimes time consuming and impractical. In real scenarios, a pretrained transfer learning model predicting forces quickly once fine-tuned to target distribution would be a favorable choice and hence needs to be examined. Therefore, in this study a unified supervised FMG-based deep transfer learner (SFMG-DTL) model using CNN architecture was pretrained with multiple sessions FMG source data (Ds, Ts) and evaluated in estimating forces in separate target domains (Dt, Tt) via supervised domain adaptation (SDA) and supervised domain generalization (SDG). For SDA, case (i) intra-subject evaluation (Ds ≠ Dt-SDA, Ts ≈ Tt-SDA) was examined, while for SDG, case (ii) cross-subject evaluation (Ds ≠ Dt-SDG, Ts ≠ Tt-SDG) was examined. Fine tuning with few “target training data” calibrated the model effectively towards target adaptation. The proposed SFMG-DTL model performed better with higher estimation accuracies and lower errors (R2 ≥ 88%, NRMSE ≤ 0.6) in both cases. These results reveal that interactive force estimations via transfer learning will improve daily HRI experiences where “target training data” is limited, or faster adaptation is required.