最低センサー数を用いた人間の動さ計測と認識とその応用

使用最少数量的传感器进行人体运动测量和识别及其应用

• 批准号: 14F04768
• 负责人: Venture Gentiane
• 金额: $ 1.47万
• 依托单位: Tokyo University of Agriculture and Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for JSPS Fellows
• 财政年份: 2014
• 资助国家: 日本
• 起止时间: 2014-04-25 至 2017-03-31
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-14F04768/
• 关键词: Robotics; inverse kinematics; inverse dynamics; Extended Kalman filter; motion analysis; 人間力学

Dr. Vincent Bonnet has succeeded in fulfilling his research plan and providing even more results than expected. He has implemented successfully our identification procedure usually used for robots and high precision sensors, on the human with low cost equipment. He has also successfully implemented the quadratic optimization to estimate not only the subset of dynamics parameters but the entire set of dynamic parameters. The parameters are identified with relative low error and the method has been validated with a robot arm (for measurement accuracy validation) and with several human subjects (for methodology validation).Dynamic parameters of the human body can be obtained in less than 60s, with a simple task, that can be adjusted with the level of motion ability of the subject. The motion can be adapted in term of range, velocity, acceleration from healthy athletes to elderly with pathologies. The system and the theory can thus be used in the clinical field for broader applications. He conducted experiments in collaboration with Dr. Noritaka Kawashima at the National Rehabilitation Center for People with Disabilities to validate the clinical application. His research results lead to new break-through in the use of subject-specific models in the clinical and rehabilitation fields, in new diagnosis and quantification procedures.

Vincent Bonnet 博士成功地完成了他的研究计划，并取得了超出预期的成果。他使用低成本设备成功地在人类身上实施了我们通常用于机器人和高精度传感器的识别程序。他还成功地实现了二次优化，不仅可以估计动态参数的子集，还可以估计整个动态参数集。参数识别误差相对较低，并且该方法已通过机器人手臂（用于测量精度验证）和多个人体受试者（用于方法验证）进行了验证。可以在不到 60 秒的时间内获得人体的动态参数，任务简单，可以根据受试者的运动能力水平进行调整。该运动可以在范围、速度、加速度方面进行调整，从健康运动员到患有疾病的老年人。该系统和理论因此可以在临床领域得到更广泛的应用。他与国家残疾人康复中心的Noritaka Kawashima博士合作进行了实验，以验证临床应用。他的研究成果在临床和康复领域、新的诊断和量化程序中使用特定主题模型方面取得了新的突破。

项目成果

Constrained Dynamic Parameter Estimation using the Extended Kalman Filter

使用扩展卡尔曼滤波器进行约束动态参数估计

• 发表时间: 2015
• 期刊: Proc. of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems
• 作者: V. Joukov;V. Bonnet;G. Venture;D. Kulic
• 通讯作者: D. Kulic

University of Montpellier(France)

蒙彼利埃大学（法国）

Human Motion Segmentation using Cost Weights Recovered from Inverse Optimal Control

使用从逆最优控制恢复的成本权重进行人体运动分割

• 发表时间: 2016
• 作者: J. F.-S. Lin;V. Bonnet;A. M. Panchea;N. Ramdani;G. Venture;and D. Kulic
• 通讯作者: and D. Kulic

Comparison of Kinematic and Dynamic Sensor Modalities and Derived Features for Human Motion Segmentation

人体运动分割的运动学和动态传感器模式及派生特征的比较

• 发表时间: 2016
• 作者: J. F.-S. Lin;V. Bonnet;V. Joukov;G. Venture;D. Kulic
• 通讯作者: D. Kulic

Fast Determination of the Planar Body Segment Inertial Parameters Using Affordable Sensors

• DOI: 10.1109/tnsre.2015.2405087
• 发表时间: 2015-07-01
• 期刊: IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
• 影响因子: 4.9
• 作者: Bonnet, Vincent;Venture, Gentiane
• 通讯作者: Venture, Gentiane