Cable-Driven Upper Body Exosuit (CUBE)
Overview
A novel bilateral upper body augmentation and rehabilitation system incorporating cable-driven mechanics and myoelectric control.
Key Innovations
- Four-degrees-of-freedom bilateral design with minimally rigid architecture
- Real-time myoelectric control system using surface EMG sensors
- Comprehensive validation through human subject testing (30+ participants)
Detailed Description
[Detailed project description to be added]
Methods
[Research methods to be added]
Results
[Key findings and results to be added]
Publications
[Related publications to be added]
Deep Reinforcement Learning-based Auto-tuner
Overview
An innovative approach to auto-tune the internal model of the human central nervous system using deep reinforcement learning.
Key Achievements
- Developed neuromusculoskeletal simulation in Mujoco physics engine
- Implemented DDPG algorithm for internal model tuning
- Achieved superior accuracy and response time compared to conventional methods
Detailed Description
[Detailed project description to be added]
Methods
[Research methods to be added]
Results
[Key findings and results to be added]
Publications
[Related publications to be added]
Cable-driven Ankle Perturbation System
Overview
A sophisticated system designed to study fall mechanics through controlled slip and trip simulations.
Technical Features
- FEA-validated protective frame design
- Dual-mode perturbation system (slip and trip simulation)
- Real-time gait segmentation using plantar pressure measurement
Detailed Description
[Detailed project description to be added]
Methods
[Research methods to be added]
Results
[Key findings and results to be added]
Publications
[Related publications to be added]