Modeling of Graceful Human Motions and Generating of Graceful Robot Motions

优美的人体动作建模和优美的机器人动作生成

• 批准号: 23500249
• 负责人: UEDA Etsuko
• 金额: $ 3.24万
• 依托单位: Nara National College of Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2011
• 资助国家: 日本
• 起止时间: 2011 至 2013
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-23500249/
• 关键词: ロボット動作生成; 優美な動作; 動作解析; モーションキャプチャ

This study is focusing on "graceful" motions of humans and aims to quantify it to make robot motions more favorable. In the field of sports kinesiology and aesthetics, "Grace of motion" was defined qualitatively, but it was not defined quantitatively, yet. We have attempted to extract the grace features from the serving motions of waitperson and the classical dance motions, using these qualitative definition.We proposed some algorithms about "S-shape features extraction" and "quantification of motion fluidity". Motion trajectories were mapped to 2 dimensional plane by conducting principal component analysis, and calculated "S-shaped feature parameters" by curve fitting. Moreover we divided motion trajectories using curvature factor and torsion factor and extracted the number of divided trajectories per 1m, as "fluidity of motion". By comparing quantification results with impression evaluation results, we confirmed successful of quantification of "Grace of motion" about some tasks.

这项研究的重点是人类的“优雅”动作，旨在对其进行量化，以使机器人的动作更加有利。在运动运动学和美学领域，“运动之美”有定性的定义，但尚未有定量的定义。我们试图利用这些定性定义从服务员的服务动作和古典舞蹈动作中提取优雅特征。我们提出了一些关于“S形特征提取”和“运动流动性量化”的算法。通过主成分分析将运动轨迹映射到二维平面，并通过曲线拟合计算“S形特征参数”。此外，我们使用曲率因子和扭转因子来划分运动轨迹，并提取每1m划分的轨迹数量，作为“运动的流动性”。通过将量化结果与印象评估结果进行比较，我们确认了对某些任务的“动作优雅”的量化是成功的。

项目成果

Modeling of graceful motions: determining characteristics of graceful motions from handover motion

优美动作建模：从切换动作中确定优美动作的特征

• 作者: Takahiro Tanaka;Takumi Tsuduki;Etsuko Ueda;Kentaro Takemura & Takayuki Nakamura
• 通讯作者: Kentaro Takemura & Takayuki Nakamura

古典舞踊動作の優美さ特徴

古典舞动作的优雅特征

• 作者: Maegawa K.;Shiotani T.;Iwamoto K.;Kasetani M.;Noguchi T.;Lee J.H.;上田 悦子，後藤 毅朗，田中 隆寛，西 諒一郎
• 通讯作者: 上田 悦子，後藤 毅朗，田中 隆寛，西 諒一郎

Extraction of graceful motion features from pick-up task

从拾取任务中提取优美的运动特征

• 发表时间: 2012
• 作者: Takumi Tsuzuki;Etsuko Ueda
• 通讯作者: Etsuko Ueda