Use of smell in human-robot interaction

Research Project

Project/Area Number	23K11271
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Review Section	Basic Section 61050:Intelligent robotics-related
Research Institution	Kyoto University
Principal Investigator	Brscic Drazen 京都大学, 情報学研究科, 准教授 (50605011)
Project Period (FY)	2023-04-01 – 2026-03-31
Project Status	Granted (Fiscal Year 2023)
Budget Amount *help	¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000) Fiscal Year 2025: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000) Fiscal Year 2024: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000) Fiscal Year 2023: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Keywords	Smell / Human-robot interaction / Nonverbal communication / huamn-robot interaction / smell
Outline of Research at the Start	The project aims to study the robot's use of smell during the interaction of humans. The plan is elucidate the appropriate ways to integrate smell within the interaction and the
Outline of Annual Research Achievements	During FY2023 we focused on integrating active smell delivery in robots to enhance interactions with humans. Since humans basically never use smells on purpose as a means of interaction, the ways how to appropriatelly use them in interactions cannot be understood by simply copying humans, like for other nonverbal communication. Instead, we approached the addressing of this problem by using a series of exploratory studies. This included interviews, human-human interactions, and human-robot interactions. Within these studies we actively included smell into the interactions by using a smell emitting device. By rep@eating such investigations we were able to gradually identify what are the key design factors for integrating smells with robots and how to use them in interactions. The findings from the studies revealed that three main factors need to be considered: (1) the mounting position of the device and how to position it appropriately for smell delivery, (2) the effect that the additional motions for positioning the smell-emitting device have on the interactions, and (3) the appropriate timing of smell emission and coordination with speech and movements. For each of these factors we also identified the specific ways how they can be accounted for in the case of a humanoid social robot. We believe such considerations will be important for any future use of smell for interactions on robots. The findings were presented in a poster session at an refereed international conference.
Current Status of Research Progress	Current Status of Research Progress 2: Research has progressed on the whole more than it was originally planned. Reason The research in 2023 has pioneered the use of smells on robots and resulted in a preliminary, but unique and novel contribution to the knowledge on olfactory human-robot interaction. This follows closely the originally planned work for FY202. We have finished the Phase 1 of the project "Natural delivery of smell" as planned, and we believe that we fully achieved the results that we hoped for. We were also able to publish the results of that phase internationally and present a poster at a conference. We also successfully started working on Phase 2. We thus judge the progress to be smooth.
Strategy for Future Research Activity	In FY2024 we will focus more on the concrete use of smell for interactions, as was originally planned for the Phase 2 of the project. One of the unexpected results from the experiments in the previous year was that the accurate timing of smell and the synchronization with dialogue is difficult to achieve. We therefore plan to alter the research direction slightly from what was planned originally. Instead of focusing solely on fully autonomous robots, we are going to include the use of smell by people through teleoperated robots. We believe that this is actually an even better direction, which will allow us to gain more insight on how to integrate smell within the robot’s dialogue. We plan to evaluate such use of smells in experiments with participants, in which one of the participants will be teleoperating a robot that has the capability to use smells.