Identification of Postural Controllers in Human Standing Balance

Area of Science:

• Biomechanics
• Neuroscience
• Systems Biology

Background:

• Human standing balance is crucial but not fully explained by current control laws.
• Existing system identification methods for the central nervous system (CNS) are limited by short motion durations or linearity assumptions.
• Comprehensive control laws require analysis of complex, long-duration balance tasks with nonlinear models.

Purpose of the Study:

• To develop and validate a method for identifying complex CNS control laws for human standing balance.
• To investigate the necessity of nonlinearities and time delays in balance control models.
• To extract comprehensive control laws from extensive human standing balance data.

Main Methods:

• Utilized trajectory optimization with the direct collocation method for system identification.
• Validated the method using synthetic motion data to ensure accurate parameter extraction.
• Identified six controller types, ranging from linear to nonlinear, using extensive motion data from perturbed standing.

Main Results:

• The trajectory optimization method successfully identified complex CNS models for standing balance.
• Results indicated that multiple time-delay paths are essential for accurate modeling.
• Nonlinear properties of the feedback control system were found to be necessary for explaining human balance.

Conclusions:

• Trajectory optimization with direct collocation is effective for identifying sophisticated CNS control laws.
• Human standing balance relies on complex feedback mechanisms involving time delays and nonlinearities.
• This approach advances the understanding of motor control in human balance.