Scholarlink

Preventive interventions for age-related frailty and sarcopenia have gained significant attention. Lower limb function is particularly vulnerable to the effects of aging, often leading to functional decline. To compensate for this loss, individuals may rely on mobility devices, but such reliance can inadvertently contribute to further deterioration in physical abilities. To address these challenges, we proposed and developed PeriPa, a hands-free mobility device designed to preserve and enhance walking ability. This study introduces an advanced operation system for PeriPa that utilizes multiple sensors to extract individual operating characteristics and adaptively learns turning angular velocity based on these characteristics. The system operates on a cooperative learning model, wherein the system learns the user’s characteristics while the user simultaneously learns to operate the system. To evaluate its effectiveness, a comparative experiment was conducted between the developed system and an existing system that does not account for individual characteristics. In the experiment, Heading Error(HE) and Goal Approach Velocity(GAV) were used as new evaluation indices. Results indicated that the cooperative learning operation system was particularly effective for individuals who struggled with the existing system. Furthermore, prior learning of individual characteristics proved beneficial when operating the system for the first time.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of a Cooperative Learning-Based Operation System for Estimating Operator Intent and Enhancing Hands-Free Mobility

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yoshito Kamon,
  • Chao Lyu,
  • Yuri Nishiwaki,
  • Tao Chen,
  • Takazumi Ono,
  • Yusuke Sugahara,
  • Misato Nihei

摘要

Preventive interventions for age-related frailty and sarcopenia have gained significant attention. Lower limb function is particularly vulnerable to the effects of aging, often leading to functional decline. To compensate for this loss, individuals may rely on mobility devices, but such reliance can inadvertently contribute to further deterioration in physical abilities. To address these challenges, we proposed and developed PeriPa, a hands-free mobility device designed to preserve and enhance walking ability. This study introduces an advanced operation system for PeriPa that utilizes multiple sensors to extract individual operating characteristics and adaptively learns turning angular velocity based on these characteristics. The system operates on a cooperative learning model, wherein the system learns the user’s characteristics while the user simultaneously learns to operate the system. To evaluate its effectiveness, a comparative experiment was conducted between the developed system and an existing system that does not account for individual characteristics. In the experiment, Heading Error(HE) and Goal Approach Velocity(GAV) were used as new evaluation indices. Results indicated that the cooperative learning operation system was particularly effective for individuals who struggled with the existing system. Furthermore, prior learning of individual characteristics proved beneficial when operating the system for the first time.