Related Experiment Videos
Biomechanical modeling and optimal control of human posture
Luciano Luporini Menegaldo1, Agenor de Toledo Fleury, Hans Ingo Weber
1São Paulo State Institute for Technological Research, Control System Group/Mechanical and Electrical Engineering Division, and Department of Mechanical Engineering, Polytechnic School, University of São Paulo, Brazil. lmeneg@ipt.br
Journal of Biomechanics
|October 3, 2003
Summary
This study models human posture in the sagittal plane using optimal control for rising movements. It details a 10-musculotendon actuator model and advanced algorithms for generating controlled motion from a squat.
Area of Science:
- Biomechanics
- Robotics
- Control Theory
Background:
- Human postural control is complex, involving intricate muscle coordination.
- Simulating and controlling dynamic movements like rising from a squat is challenging due to non-linear dynamics.
Purpose of the Study:
- To develop a biomechanical model of human sagittal plane postural mechanics.
- To utilize optimal control techniques to generate open-loop raising movements from a squatting position.
Main Methods:
- A 10-musculotendon actuator model based on 40 muscles and a three-link system (shank, thigh, HAT) was employed.
- Optimal control solutions were derived using Consistent Approximations Theory with Runge-Kutta integration and spline-based control signals.
- Non-linear programming problems were solved using a sequential quadratic programming (SQP) method.
Main Results:
- The study presents control signals (muscular excitations) and angular trajectories for simulated raising movements.
- Specific control strategies were developed to address numerical convergence challenges in non-linear and unstable posture dynamics.
Conclusions:
- The developed biomechanical model and optimal control approach can generate realistic human raising movements.
- The findings highlight the importance of specialized strategies for achieving convergence in complex dynamic systems.