Slow waltzing with REEM-C: a physical-social human-robot interaction study of robot-to-human communication

• Main Authors: Charbonneau, Marie, Chavez, Francisco Javier Andrade, Mombaur, Katja
• Format: Journal Article
• Language: English
• Published: 09-08-2024
• Subjects: Computer Science - Robotics
• DOI: 10.48550/arxiv.2408.05301

Humans often work closely together and relay a wealth of information through
physical interaction. Robots, on the other hand, have not yet been developed to
work similarly closely with humans, and to effectively convey information when
engaging in physical human-robot interaction (pHRI). This currently limits the
potential of physical human-robot collaboration to solve real-world problems.
This paper investigates the question of how to establish clear and intuitive
robot-to-human communication, while ensuring human comfort during pHRI. We
approach this question from the perspective of a leader-follower scenario, in
which a full-body humanoid robot leads a slow waltz dance by signaling the next
steps to a human partner. This is achieved through the development of a
whole-body control framework combining admittance and impedance control, which
allows for different communication modalities including haptic, visual, and
audio signals. Participant experiments allowed to validate the performance of
the controller, and to understand what types of communication work better in
terms of effectiveness and comfort during robot-led pHRI.