EAGER: Multi-modal human gait experimentation and analysis on unconstrained terrains

EAGER：无约束地形上的多模式人类步态实验和分析

• 批准号: 1451292
• 负责人: Dimitris Metaxas
• 金额: $ 15万
• 依托单位: Rutgers University New Brunswick
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1451292&HistoricalAwards=false
• 关键词: EAGER; Multi; modal; human; gait

This project explores a new direction to analyze human gait. The project obtains high accurate gait data, movement and high resolution ground reaction forces on different terrains, which has not been possible to measure and acquire previously. The project enables to produce the first ever detailed and multimodal gait database that can be used for research and education. It can assist computer scientists and biomedical researchers in addressing complex questions such as understanding, analyzing and simulating the kinematics and dynamics of walking in natural settings. The resulting research and datasets can be incorporated in the curriculum of many fields such as computer science, biomedical engineering and medicine and will result in significantly better educated future researchers, professors, and employees in gait analytics. The project focus on collecting novel, accurate, multi-modal gait data on unconstrained terrains and developing a novel approach to analyze the gait data. The research team develops predictive models to create realistic simulations of human gait, which is important for many industries such as healthcare, sports, footwear, and entertainment. This requires accurate human movement data and ground reaction forces on different terrains. Currently the most common method to collect the gait data is to put surface markers on human subjects. Due to the skin movement, surface marker data collected by this method contain a significant error. Gait kinematics estimated from the surface markers are not accurate and may be quite off from the real human movement. Moreover, the most common methods to collect the ground reaction forces use a simple force plate. However, a simple force plate can only measure the ground reaction forces on one simple type of terrain and cannot measure the ground force close to real daily life. The ground reaction forces on the terrains similar to real life are unknown. This exploration project enables many new studies and applications.

本课题为人体步态分析探索了一个新的方向。该项目获得了高精度的步态数据、不同地形上的运动和高分辨率的地面反作用力，这是以前无法测量和获取的。该项目能够产生有史以来第一个详细的、可用于研究和教育的多模式步态数据库。它可以帮助计算机科学家和生物医学研究人员解决复杂的问题，如理解、分析和模拟在自然环境中行走的运动学和动力学。由此产生的研究和数据集可以被纳入许多领域的课程，如计算机科学、生物医学工程和医学，并将导致受过更好教育的未来步态分析方面的研究人员、教授和员工。该项目致力于在不受限制的地形上收集新颖、准确、多模式的步态数据，并开发一种新的方法来分析步态数据。该研究团队开发预测模型来创建真实的人体步态模拟，这对医疗保健、体育、鞋类和娱乐等许多行业都很重要。这需要准确的人体运动数据和不同地形上的地面反作用力。目前，采集步态数据最常用的方法是在受试者身上放置表面标记。由于皮肤的运动，这种方法采集的表面标记数据包含很大的误差。从表面标记估计的步态运动学是不准确的，可能与真实的人类运动相当偏离。此外，收集地面反力最常见的方法是使用简单的力板。然而，简单的测力板只能测量一种简单地形上的地面反力，无法测量接近实际生活的地面反力。类似于现实生活的地形上的地面反作用力是未知的。这一勘探项目使许多新的研究和应用成为可能。

项目成果

Predicting 3-D Lower Back Joint Load in Lifting: A Deep Pose Estimation Approach

• DOI: 10.1109/thms.2018.2884811
• 发表时间: 2019-02-01
• 期刊: IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS
• 影响因子: 3.6
• 作者: Mehrizi, Rahil;Peng, Xi;Li, Kang
• 通讯作者: Li, Kang