研究実績の概要 |
This year, object handover experiments, which is one of human-human collaboration tasks, were first conducted. From the experiments, we investigated whether humans have a model of their partner’s handover behavior. Handovers were executed by several subjects in the absence of visual feedback and the results suggested that humans can predict the partner's handover behavior and there are some important variables including physical and social features of the pairs, which effect human motion modification. Namely, the existence of partner has a significant influence on human motion in collaboration tasks as we expected, and the variables will provide crucial guidelines for design of robot behavior. In addition to the human-human experiments, we developed a robot system to examine the human perception of the robot behavior for the next step. In parallel we have started to work on understanding human collaboration in co-worker scenarios. For this, with a PhD and Master's student, we have developed the first experimental setup to monitor human arm movements during a co-worker task. We have done the first preliminary experiments with this setup and are now analyzing the first human hand movement data.
|
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 data collection, which was the target for FY2017, is still ongoing, but we were able to finish the first phase of data analysis for handover tasks. Thus, the project seems to be going well.
|
今後の研究の推進方策 |
As we submitted for FY2018 research plan, we will continue the experiments of human-human collaboration to accumulate a certain amount of data for data analysis and computational model of the peri-limb space. In parallel, experiments with the human-embodied-robot scenario will be started to examine the human perception of the robot behavior. By the end of FY2018, the first rules to define the peri-limb space are expected to be clear. We added a 研究分担者 from FY2018 to accelerate the progress.
|