Organizations: Guardian Robot Project, R-IH, Riken, Kyoto, Japan · Faculty of Engineering, Monash University, Melbourne, Australia · Graduate school of engineering science, Osaka University, Osaka, Japan · Faculty of engineering, Tokyo University of Science, Tokyo, Japan
Human-robot interaction combines robotics, cognitive science, and human factors to study collaborative systems. This paper introduces a method for identifying influential robot actions using transfer entropy, a statistic that measures directed information transfer between time series. TE is effective for capturing complex, nonlinear interactions. We apply this method to analyze how robot actions affect human behavior during a conversation with a remotely controlled robot avatar. By focusing on the impact of proximity, our approach demonstrates TE's capability to identify key actions influencing human responses, highlighting its potential to improve the design and adaptability of robotic systems.
Figure 5: The workflow of the proposed method. Top: Visualisation of the timeline. Bottom: Illustration of the workflow.
Figure 6: TE measurements for one experiment. Left: Action signals, including linear and angular velocity. Center: Depth measurements with predictions from both processes. The full prediction is based on the complete window of actions, while the marginal prediction uses masked past actions. Both the mean and uncertainty ( ±3 standard deviations) of the predictions are shown. Right: Transfer entropy measurements with highlighted peaks.
Figure 7: All identified influential action sequences.
Figure 8: Averaged influential action sequences.
Figure 9: Two example sequences of Type 1 actions: The robot moves forward while the participants move backward.
Figure 10: Two example sequences of Type 2 actions: The robot moves backward while the participants catch up.
Institute of Industrial Science, The University of Tokyo, Japan · Interfaculty Initiative in Information Studies, Graduate School of Interdisciplinary Information Studies, The University of Tokyo, Japan