.png){kind=icon}
.png){kind=icon}
Return
Adaptive Fuzzy Control for Triadic Interaction of Soft Exosuit-Assisted Locomotion
.png){kind=icon}
PRE
AI
L
李
Y
X
H
G
DOI：10.1109/tfuzz.2026.3698396.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Wearable flexible lower-limb exoskeletons, commonly referred to as soft exosuits, have emerged as a promising technology for enhancing mobility and assisting gait rehabilitation. However, most conventional designs provide assistance failing to adapt to variations in walking cadence or gait phase subdivision. This study presents a planning and control framework for a cable-driven exosuit that delivers personalised ankle assistance in the triadic interaction encompassing the human, the robot, and the environment. The framework enables automatic gait phase detection and adaptive assistance, dynamically responding to changes in gait, environmental conditions, and metabolic demands. Inspired by the periodic nature of human walking, a gait cycle is divided into eight phases using a designed transformer classification model that processes foot force and inertial measurement unit (IMU) data. Heart rate is incorporated to provide feedback on metabolic changes. A human-in-the-loop control strategy based on an event-triggered mechanism utilizing fine gait classification and heart rate is proposed to achieve ankle joint assistance adaptation. When the deviations in gait phases or metabolic conditions exceed thresholds, the ankle joint assistive trajectory is replanned online via optimization with safety constraints. An adaptive fuzzy controller ensures stable tracking under uncertain dynamics and external disturbances. The stability of the control system is analytically verified using Lyapunov theory, and experimental results demonstrate the effectiveness of the proposed approach.
Keywords:
Adaptive fuzzy control
event triggering
gait recognition
human-in-the-loop control
soft exosuit
torque assistance
triadic interaction
Journal
IF:
11.9
Papers:
4.9K
Citations:
2.9W
Organization
Cited Papers