| 研究実績の概要 |
In FY2019 we have completed the implementation of the control architecture to carry out the experiment with external participants. We recruited 23 external participants who took part of the experiment, during which they have been equipped with the markers for the motion capture, eye tracking, GSR sensor, and heart beat monitor. All the participants have received proper information and given an informed consent to participate in the study. Before the experiment, they have been asked to fill in a questionnaire to evaluate their personality, then after the experiment they have been asked to fill in two more questionnaires, the Godspeed Series Questionnaire (GSQ) and the CH-33, to evaluate their perceptions regarding the robot. The data collected during the experiments have been elaborated to identify the “interaction factors”, mainly based on statistical analysis. This analysis helped us to identify a few interesting factors, for example relationship between personality and distance to the robot, however not in a satisfactory way. Considering that this work was a first attempt in combining physical and social human-robot interaction in order to identify measurable interaction factors, we deemed the results to be relevant for the design of a second experiment. In the new experiment, the action of the robot will be not be a disturbance, but it will be a specific intended action that we let the user to understand. We believe that with the new experiment, more consistent factors will emerge.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
The current status is on schedule with respect to the revised plan for FY2020. With the experiment we designed and carried out in FY2019, we have identified a few relevant interaction factors. Based on this we will proceed with the design of the new experiment, in order to further investigate factors identification, before proceeding with the development of the control framework.
The results of the experiments carried out in FY2019 have been submitted to Robotics and Automation Letters (RA-L) as journal paper with the IROS2020 conference option, it is now under review.
|
| 今後の研究の推進方策 |
In FY2020 we will complete the implementation of the new experiment using the Sawyer robot. The sensors setup will be similar to the previous one, but a new and more precise eye-tracker will be used to ensure higher quality data. Given that the overall setup will be similar to the previous experiment, we expect to be able to elaborate the results in a shorter time, allowing to extract further interaction factors. The experiment will be performed with a higher number of participants with respect to FY2019 to ensure consistent data. We expect to be able to submit another paper based on the new experiment.
|
