Formulating Social Cues for Collaborative Robots

Date

2024-9-4

Author

Bolat, Burak

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

32
views

0
downloads

Cite This

Collaborative robotic manipulators (a.k.a cobots) are rapidly being deployed on factory floors to operate alongside human workers as their mates to reduce their physical/cognitive load. For human-robot interaction (HRI) research, these cobots, which are designed as open-chain manipulators to carry out manipulation tasks 24/7 at high reliability/precision, provide both an opportunity (of using cobots as a reliable and affordable platform to conduct HRI research) and a challenge (of imbuing cobots with HRI skills to improve their interaction with workers). In this thesis, we propose two non-verbal HRI filters, namely ``gazing'' and ``breathing", that would transform a UR5 robot arm equipped with a two-finger gripper into a research platform for HRI research. Specifically, the cobot ``gazed" by pointing its gripper towards the point of attention using its last three joints, and ``breathed" by performing an oscillatory movement with respect to a reference posture using the first three joints close to its base. In the first user study, we conducted and reported a human subjects experiment with 57 participants, evaluating the efficacy of the two HRI filters (namely gazing and breathing) towards improving the HRI quality of the cobot's interaction with a human in a collaborative assembly scenario. Our analysis based on post-experiment questionnaires in a repeated measures experiment shows that both filters significantly improve the HRI quality measured through pre-validated constructs including Social Presence, Anthropomorphism, and Animacy. Interestingly, improvements in the primary outcomes using any single filter or using both filters simultaneously were statistically indistinguishable. In the second user study, we conducted and reported a human subjects experiment with 22 participants testing different parameters of ``gazing'' and ``breathing". We used the same evaluation with the first user study and quantitative distance measurements. Our analysis shows a few improvements in HRI quality, yet no significant impact on cobot-participant distance.

Subject Keywords

Non-verbal communication, Collaborative Robots, Human Robot Interaction, Social Cues, Collaborative Assembly, Gazing, Breathing

URI

https://hdl.handle.net/11511/111539

Collections

Graduate School of Natural and Applied Sciences, Thesis