Abstract
This paper details the development, implementation, and experimental evaluation of an interface-aware, task-agnostic assistance system for handling unintended interface operation during shared human-robot teleoperation, specifically applied to a 7-DoF robotic arm. The system addresses a limitation of current shared-control methods by considering the impact of control interfaces on user input precision and the robot agent’s incomplete understanding of the human’s policy. Additionally, the system addresses the issue of data efficiency in a data-scarce domain where gathering extensive data is impractical. The approach is evaluated in empirical case studies involving participants with spinal cord injuries and with data-driven models that are originally derived via direct statistical modeling, and later are derived via transfer learning. The paper overviews the limitations addressed and improvements made throughout this iterative process. The personalized assistance system is found to improve safety and reduce cognitive load across participants with spinal cord injuries in real-world assistive settings with minimal training data requirements.
Keywords
Get full access to this article
View all access options for this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
