Automatic detection of myocontrol failures based upon situational context information

Heiwolt K, Zito C, Nowak M, Castellini C, Stolkin R (2019)

Publication Type: Conference contribution

Publication year: 2019

Journal

IEEE ... International Conference on Rehabilitation Robotics : [proceedings] Institute of Electrical and Electronics Engineers Inc.

Publisher: IEEE Computer Society

Book Volume: 2019-June

Pages Range: 398-404

Conference Proceedings Title: IEEE International Conference on Rehabilitation Robotics

Event location: Toronto, ON CA

ISBN: 9781728127552

DOI: 10.1109/ICORR.2019.8779478

Abstract

Myoelectric control systems for assistive devices are still unreliable. The user's input signals can become unstable over time due to e.g. fatigue, electrode displacement, or sweat. Hence, such controllers need to be constantly updated and heavily rely on user feedback. In this paper, we present an automatic failure detection method which learns when plausible predictions become unreliable and model updates are necessary. Our key insight is to enhance the control system with a set of generative models that learn sensible behaviour for a desired task from human demonstration. We illustrate our approach on a grasping scenario in Virtual Reality, in which the user is asked to grasp a bottle on a table. From demonstration our model learns the reach-to-grasp motion from a resting position to two grasps (power grasp and tridigital grasp) and how to predict the most adequate grasp from local context, e.g. tridigital grasp on the bottle cap or around the bottleneck. By measuring the error between new grasp attempts and the model prediction, the system can effectively detect which input commands do not reflect the user's intention. We evaluated our model in two cases: i) with both position and rotation information of the wrist pose, and ii) with only rotational information. Our results show that our approach detects statistically highly significant differences in error distributions with p<0.001 between successful and failed grasp attempts in both cases.

Authors with CRIS profile

Claudio Castellini

Involved external institutions

University of Birmingham GB United Kingdom (GB) Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) / German Aerospace Center DE Germany (DE)

How to cite

APA:

Heiwolt, K., Zito, C., Nowak, M., Castellini, C., & Stolkin, R. (2019). Automatic detection of myocontrol failures based upon situational context information. In IEEE International Conference on Rehabilitation Robotics (pp. 398-404). Toronto, ON, CA: IEEE Computer Society.

MLA:

Heiwolt, Karoline, et al. "Automatic detection of myocontrol failures based upon situational context information." Proceedings of the 16th IEEE International Conference on Rehabilitation Robotics, ICORR 2019, Toronto, ON IEEE Computer Society, 2019. 398-404.

BibTeX: Download