Related Experiment Videos
Simulation of human gait using computed torque control.
N F Unver1, S T Tümer, M K Ozgören
1Informatics Institute, Middle East Technical University, Ankara, Turkey.
Summary
This study models human gait, including single and double support phases, using geometric constraints and computed torque control. The model can resume normal gait after perturbations but requires further refinement for realistic joint torques.
Area of Science:
- Robotics and Biomechanics
- Mathematical Modeling
- Control Systems
Background:
- Human gait is a complex dynamic process involving distinct single and double support phases.
- Accurate mathematical modeling is crucial for understanding and replicating human locomotion.
Purpose of the Study:
- To develop a mathematical model for simulating both single and double support phases of human gait.
- To implement a feedback control strategy for achieving normal gait trajectories.
Main Methods:
- Modeling the human body as a linkage system in a floating state.
- Imposing foot-ground interaction using geometric constraints for different gait phases.
- Applying computed torque control with weighted least squares optimization for redundant torques.
Main Results:
- The model successfully simulated imposed hip and ankle trajectories with minor deviations.
- The control strategy demonstrated corrective actions to resume gait patterns after perturbations.
- Significant deviations in control torque magnitudes were observed under perturbation.
Conclusions:
- The proposed planar kinematic model and unbounded joint torque assumption limit the realism of control torques.
- Further development is needed to enhance the model's accuracy and the control strategy's practicality for human gait simulation.