摘要
下肢康复机器人训练是现代康复医学的关键手段,能给下肢运动功能障碍患者的康复训练提供高效、精准、个性化的解决办法,有助于提升康复效果与生活质量。当前国内有关智能下肢康复机器人技术研究进展的文献比较少,本文对智能下肢康复机器人技术进行系统性地介绍,概述下肢康复机器人的发展历程与主要类型,汇总深度学习技术在下肢康复机器人研究中的应用,包含卷积神经网络、循环神经网络和强化学习等。针对该领域面临的关键问题,如数据获取与标注存在险阻、模型泛化能力有限以及实时性要求高、控制任务有挑战等,本文探讨多模态数据融合、小样本学习以及和虚拟现实技术结合等未来发展趋势,希望推动下肢康复机器人技术发展。
As a key tool in modern rehabilitation medicine,lower limb rehabilitation robots offer efficient,precise,and personalized training solutions for patients with lower limb motor dysfunction,significantly improving rehabilitation outcomes and quality of life.At present,there are relatively few review articles on intelligent lower limb rehabilitation robot technologies in China;this paper aims to provide researchers with a systematic overview of research in this area.This paper briefl y reviews the development history and main types of lower limb rehabilitation robots and elaborates on the application of deep learning technologies in this field,covering convolutional neural networks,recurrent neural networks,and reinforcement learning.By deeply analyzing key challenges in the field,such as diffi culties in data acquisition and labeling,limited model generalization,and the demands of real-time control tasks,this paper explores future development trends including multimodal data fusion,few-shot learning,and integration with virtual reality technologies,with the aim of advancing lower limb rehabilitation robot technologies and enhancing rehabilitation experiences for patients.
作者
袁小维
韩雪
周卓颖
郭磊
Yuan xiaowei;Han Xue;Zhou Zhuoying;Guo lei(School of Computer Science and Engineering(School of Cyber Security),University of Electronic Science and Technology of China,Chengdu,Sichuan;Department of Physical Education,Wuxi Institute of Technology,Wuxi,Jiangsu)
出处
《运动科学与健康研究》
2025年第4期63-70,76,共9页
Research on Sports Science and Health
关键词
下肢康复机器人
深度学习
步态预测
运动模式识别
姿态估计
Lower limb rehabilitation robots
Deep learning
Gait prediction
Locomotion mode recognition
Posture estimation
