Legibility and Predictability of NICO Robot Movement

Abstract

Introduction

In the context of the rapid development of AI and robotics, the question of effective human-machine interaction becomes increasingly important. Various factors contribute to the success of communication between humans and machines, and one such crucial factor is movement design, which significantly affects legibility and predictability. Legibility refers to humans' capacity to understand and anticipate a robot's intentions and actions based on its movements [1]. A study conducted using the NICO platform reveals that fluency and trajectory are essential factors that enhance humans' ability to predict a robot's intentions [3]. Additionally, research suggests that human perception of robot intentions is influenced by the robot's movements [2]. Building upon these findings, our experimental design aims to investigate and validate these results.

Experimental Design

This study aims to investigate the impact of movement design on the ability of humans to understand and predict the intentions of a robot. The NICO robot platform is utilized to perform a task involving pointing to specific fields on a 3x8 grid. An HRI experiment pilot is conducted, where respondents are asked to predict the final state of the robot's arm after the movement is stopped at one point. Half of the respondents witness the robot's movements with its head involved, while the other half only observe the arm movements with the head locked in one position. The experiment is followed by a Godspeed questionnaire to gather feedback and optimize the experimental design. The hypothesis posits that movements involving the head will enhance legibility and make the final states easier to predict. The results of this study will contribute to understanding the role of movement information in human perception of robot intentions, aiding the development of effective human-robot communication systems.

References

[1] A. D. Dragan, K. C. T. Lee, and S. S. Srinivasa, "Legibility and predictability of Robot Motion," 8th ACM/IEEE International Conference on Human-Robot Interaction (HRI), 2013. doi:10.1109/hri.2013.6483603

[2] M. Kerzel et al., "What’s on your mind, Nico?," KI - Künstliche Intelligenz, vol. 36, no. 3–4, pp. 237–254, 2022. doi:10.1007/s13218-022-00772-8

[3] L. Sartori, C. Becchio, and U. Castiello, "Cues to intention: The role of movement information," Cognition, vol. 119, no. 2, pp. 242–252, 2011. doi:10.1016/j.cognition.2011.01.014