面向患者意图理解与辅助技能学习的康复机器人人机交互协同方法研究
结题报告
批准号:
61973065
项目类别:
面上项目
资助金额:
63.0 万元
负责人:
丁其川
依托单位:
学科分类:
机器人学与智能系统
结题年份:
2023
批准年份:
2019
项目状态:
已结题
项目参与者:
丁其川
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
客服二维码
微信扫码咨询
中文摘要
针对当前康复机器人存在的患者意图理解能力、人机融合环境适应能力以及作业技巧性严重不足的问题,本项目聚焦基于神经解码的意图理解、面向现实场景的技能学习、人机融合行为规划与协同控制三方面科学问题。通过脑肌电同源与相干性分析,建立基于脑肌神经活跃度互补融合的运动意图识别模型;提出通过视频学习康复辅助技能的新思路,研究基于视觉序列的协同行为识别及语义理解方法,提出基于协同行为语义引导的机器人辅助技能知识生成方法;研究人机融合环境下患者意图、技能知识、人机安全等因素的多约束表达,提出多约束人机协同行为优化及力-运动协调控制方法。在以上研究基础上,初步建立面向运动康复机器人的人机交互协同技术理论体系,研发上肢多关节康复机器人实验平台,设计针对典型康复训练的实验范式,系统验证提出的方法及技术的先进性与有效性。本项目研究为未来研发高智能运动康复机器人,满足广大肢体损伤患者的康复需求提供新的方法和技术支撑。
英文摘要
With respect to the problem that the understanding ability for patient intention, the adaptability for the human-robot integration environment, and operational skills are seriously insufficient in current rehabilitation robots, this project focuses on three scientific issues: the intention understanding based on neural decoding, the skill learning from real-life scenarios, and the behavior planning and collaborative control in human-robot collaboration. Based on the homology and coherence analysis of the electroencephalography (EEG) and the electromyography (sEMG), a model for recognizing the motion intention is built based on the complementary fusion of the cranial nerve activity and the muscle nerve activity. The project proposes a new idea for learning rehabilitation skills through videos, and studies the methods of the collaborative behavior recognition and semantic understanding through visual sequences. Then, a method of generating the aided skills for robots is proposed based on the guidance of the collaborative behavior semantic. The project also studies to extract the multi-constrained expression of factors, including patient intention, skill knowledge and human/robot safety in the human-robot integration environment, and then proposes the method of the multi-constrained behavior optimization and the force-motion coordinated control for the human-robot collaboration. Based on the above researches, the project preliminarily establishes the theoretical system of the human-robot collaboration technology for movement rehabilitation robots. An upper limb multi-joint rehabilitation robot is developed. Through designing the experimental paradigms for typical assisted rehabilitations, the project systematically verifies the advancement and effectiveness of the proposed methods and technologies. This project provides the method and technical supports for developing high-intelligent movement rehabilitation robots, and meeting the rehabilitation needs of the majority of patients with limb injuries.
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
DOI:10.1063/5.0008434
发表时间:2020-07
期刊:The Review of scientific instruments
影响因子:--
作者:Weiwei Zhang;Fei Wang;Shichao Wu;Zongfeng Xu;Jingyu Ping;Yang Jiang
通讯作者:Weiwei Zhang;Fei Wang;Shichao Wu;Zongfeng Xu;Jingyu Ping;Yang Jiang
DOI:10.1007/s00521-023-08873-7
发表时间:2023-08
期刊:Neural Computing and Applications
影响因子:6
作者:Fei Wang;Libo Zhang;Hao Yan;Shuai Han
通讯作者:Fei Wang;Libo Zhang;Hao Yan;Shuai Han
DOI:10.1016/j.neucom.2022.07.080
发表时间:2022-07
期刊:Neurocomputing
影响因子:6
作者:Zhongyu Bai;Qichuan Ding;Hongli Xu;Jianning Chi;Xiang-Sun Zhang;Tiansheng Sun
通讯作者:Zhongyu Bai;Qichuan Ding;Hongli Xu;Jianning Chi;Xiang-Sun Zhang;Tiansheng Sun
DOI:10.1007/s13042-021-01301-z
发表时间:2021-03
期刊:International Journal of Machine Learning and Cybernetics
影响因子:5.6
作者:Fei Wang;Guorui Wang;Yuxuan Du;Zhenquan He;Yong Jiang
通讯作者:Fei Wang;Guorui Wang;Yuxuan Du;Zhenquan He;Yong Jiang
DOI:--
发表时间:2023
期刊:Neurocomputing
影响因子:--
作者:Wang Fei;Zhang Xing;Chen Tianyue;Shen Ze;Liu Shangdong;He Zhenquan
通讯作者:He Zhenquan
融合本体感应能力的肌电假肢功能提升机理研究
  • 批准号:
    62373086
  • 项目类别:
    面上项目
  • 资助金额:
    50万元
  • 批准年份:
    2023
  • 负责人:
    丁其川
  • 依托单位:
国内基金
海外基金