###
DOI:
飞控与探测:2024,(2):28-35
本文二维码信息
码上扫一扫!
基于惯性传感器的免对准动作的人体上肢运动捕捉方法
(1.南京邮电大学 自动化学院、人工智能学院;2.南京邮电大学 机器人信息感知与控制研究所)
Method of Motion Capture of Human Upper Limbs without Alignment Based on Inertial Sensors
(1.College of Automation and College of Artificial Intelligence, Nanjing University of Posts and Telecommunications;2.Robotics Information Sensing and Control Research Institute, Nanjing University of Posts and Telecommunications)
摘要
相似文献
参考文献
本文已被:浏览 820次   下载 450
    
中文摘要: 基于惯性传感器的人体运动捕捉系统,通常在进行运动捕捉前需要执行特定的标准动作来完成对准,以获得IMU 坐标系相对肢体坐标系的姿态。运动受限用户(如偏瘫患者) 无法完成一些标准动作,而运动正常用户也不能保证均能正确执行标准动作,从而导致人体运动捕捉误差。为了克服现有方法的不足,研究了一种基于惯性传感器的免对准动作的人体上肢运动捕捉方法,该方法无需执行特定对准动作,佩戴IMU 后即可直接进行运动捕捉。通过与“T-Pose”静态对准方案比较,验证了免对准动作方法的有效性,并且基于OpenGL的虚拟人体模型进行了上肢运动捕捉实验。实验表明,所提出的方法简便有效,在无须执行标准对准动作的情况下仍具有较好的人体运动捕捉效果,适合偏瘫患者等行动受限的用户使用。
Abstract:The human motion capture system based on inertial sensors usually needs to perform specific standard actions to complete the alignment before performing motion capture to obtain the attitude of the inertial measurement unit (IMU) coordinate system relative to the limb coordinate system. Users with limited motion (such as hemiplegic patients) cannot complete some standard actions. For users with normal motion, there is no guarantee that they can perform standard actions correctly, resulting in human motion capture errors. In order to overcome the shortcomings of the existing methods, this paper studies a method of human upper limb motion capture based on inertial sensors without alignment action. The method does not need to perform specific alignment action and can directly perform motion capture after wearing IMU. By comparing with the ‘T-Pose’ static alignment scheme, the effectiveness of the alignment-free action method is verified, and the upper limb motion capture experiment is carried out based on the OpenGL virtual human model. Experiments indicate that the proposed method is simple and effective, and it still has a good human motion capture effect without performing standard alignment actions, which is suitable for users with limited movement such as hemiplegic patients
文章编号:20240203     中图分类号:TP391    文献标志码:A
基金项目:江苏省高等学校自然科学研究面上项目(20KJB510023);江苏省自然科学基金(BK20210599)
引用文本:
陶幸,俞帆山,宋越杰,谭彩铭,高翔.基于惯性传感器的免对准动作的人体上肢运动捕捉方法[J].飞控与探测,2024,(2):28-35.