Skip to content
DraconiticResearch
Search Sources API
203,948 studies31 sourcesUpdated 59 min ago
Peer-reviewedCC BYBiosensors

Integrated Multi-Terrain Locomotion Recognition and Knee-Exoskeleton Assistance for Load Carrying: A Feasibility Study.

  • Jun Liu1
  • Yichen Lin2,3,4
  • Yinuo Yao3
  • Yu Zhu3,4
  • Haoran Zhang3
  • Xiangyang Wang3
  • Jianquan Sun3
  • Bo Yuan5
  • Yue Ma3
5 affiliations
  1. 1National Earthquake Response Support Service, Beijing 100049, China.
  2. 2College of Engineering, Southern University of Science and Technology, Shenzhen 518055, China.
  3. 3Guangdong Provincial Key Laboratory of Robotics and Intelligent Systems, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China.
  4. 4Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology, Shenzhen 518107, China.
  5. 5Niudi Tech. Co., Ltd., No. 9 Jiubin Road, Jiulongpo District, Chongqing 400050, China.
Read full textPDFDOICite

Abstract

Multi-terrain load carriage changes knee motion and muscular demand, requiring assistance that adapts to the current locomotion task. This feasibility study evaluated an integrated pipeline in which five inertial measurement units (IMUs) and a quaternion-based dynamic temporal pooling Transformer (QDTPformer) selected task-specific dynamic movement primitive (DMP) references for powered knee-exoskeleton assistance. Six participants took part across the offline recognition and human-assistance evaluations. For level walking (Flat), ramp ascent (Ramp), and stair ascent (Stairs), QDTPformer achieved 95.74% offline test accuracy and 89.54%±1.07% mean window-level accuracy in leave-one-subject-out evaluation. Online recognition selected the corresponding DMP reference and assistance parameters. Phase-aligned DMP-knee correlations were 0.967, 0.944, and 0.777 for Flat, Ramp, and Stairs, respectively. The equally weighted mean surface electromyography (sEMG) root mean square (RMS) reduction across two muscles and three tasks was 8.42%, with task- and wearer-dependent responses. The main contribution is an end-to-end recognition-to-assistance pipeline evaluated across three terrains using both kinematic and muscle-activity outcomes.

Related

  1. A Gait-Conditioned Human-in-the-Loop Optimization Framework Based on Gait–Assistance–Response Relationship Model for Exoskeleton Assistance
    Jiaqi Han, Liping Wang, Jianquan Ding +5
    Bioengineering (Basel, Switzerland)Peer-reviewedCC BYFull textEurope PMC
  2. Sit-to-Stand Assistance for Patients with Parkinson’s Disease Using a Lower-Limb Exoskeleton Robot
    Yu-Cheng Wu, Lo-Ping Yu, Si-Huei Lee +1
    Bioengineering (Basel, Switzerland)Peer-reviewedCC BYFull textEurope PMC
  3. Passive ankle exoskeleton assistance selectively modulates triceps surae neural control during balance.
    Raphaël Hamard, François Hug, Taylor J M Dick
    Journal of the Royal Society, InterfacePeer-reviewedCC BYFull textEurope PMC
  4. In-view vital-sign alarm recognition using a smart glass during simulated bronchoscopy: a phantom feasibility study
    Shotaro Okachi, Takuma Ina, Aki Ikeda +6
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textSurgery & anaesthesiaCrossref
  5. Feasibility and preliminary validation of an artificial intelligence-based facial emotion recognition application for real-time assessment of pediatric dental anxiety
    Amit Khatri, Rishi Tyagi, Padma Yangdol +3
    Journal of dental anesthesia and pain medicinePeer-reviewedCC BY-NCFull textEurope PMC

Cite

BibTeXRIS
BibTeX
@article{liu2026integrated,
  title = {{Integrated Multi-Terrain Locomotion Recognition and Knee-Exoskeleton Assistance for Load Carrying: A Feasibility Study.}},
  author = {Jun Liu and Yichen Lin and Yinuo Yao and Yu Zhu and Haoran Zhang and Xiangyang Wang and Jianquan Sun and Bo Yuan and Yue Ma},
  journal = {Biosensors},
  year = {2026},
  month = oct,
  doi = {10.3390/bios16100570},
  url = {https://europepmc.org/article/MED/42930117},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Integrated Multi-Terrain Locomotion Recognition and Knee-Exoskeleton Assistance for Load Carrying: A Feasibility Study.
AU  - Jun Liu
AU  - Yichen Lin
AU  - Yinuo Yao
AU  - Yu Zhu
AU  - Haoran Zhang
AU  - Xiangyang Wang
AU  - Jianquan Sun
AU  - Bo Yuan
AU  - Yue Ma
PY  - 2026
DA  - 2026/10/10
JO  - Biosensors
DO  - 10.3390/bios16100570
UR  - https://europepmc.org/article/MED/42930117
AB  - Multi-terrain load carriage changes knee motion and muscular demand, requiring assistance that adapts to the current locomotion task. This feasibility study evaluated an integrated pipeline in which five inertial measurement units (IMUs) and a quaternion-based dynamic temporal pooling Transformer (QDTPformer) selected task-specific dynamic movement primitive (DMP) references for powered knee-exoskeleton assistance. Six participants took part across the offline recognition and human-assistance evaluations. For level walking (Flat), ramp ascent (Ramp), and stair ascent (Stairs), QDTPformer achieved 95.74% offline test accuracy and 89.54%±1.07% mean window-level accuracy in leave-one-subject-out evaluation. Online recognition selected the corresponding DMP reference and assistance parameters. Phase-aligned DMP-knee correlations were 0.967, 0.944, and 0.777 for Flat, Ramp, and Stairs, respectively. The equally weighted mean surface electromyography (sEMG) root mean square (RMS) reduction across two muscles and three tasks was 8.42%, with task- and wearer-dependent responses. The main contribution is an end-to-end recognition-to-assistance pipeline evaluated across three terrains using both kinematic and muscle-activity outcomes.
SN  - 2079-6374
ER  -