Parameterizing Human Locomotion Across Quasi-Random Treadmill Perturbations and Inclines.

Area of Science:

• Biomechanics
• Human locomotion analysis
• Robotics and control

Background:

• Mechanical phase variables can track human gait cycles but are sensitive to body segment behavior.
• Task-dependent and disturbance-induced changes in body segment behavior limit traditional phase variable accuracy.

Purpose of the Study:

• Compare time, thigh phase angle, and tibia phase angle for quantifying gait cycle progression.
• Evaluate parameterization method robustness under split-belt treadmill perturbations and varying inclines.

Main Methods:

• Motion capture data from ten able-bodied subjects.
• Walking on a split-belt instrumented treadmill with quasi-random phase-shifting perturbations.
• Statistical comparison of time, thigh phase angle, and tibia phase angle parameterizations.

Main Results:

• Both phase parameterization methods (thigh and tibia) significantly outperformed time parameterization.
• Thigh phase angle demonstrated superior performance over tibia phase angle in most comparisons.
• Robustness of phase variables confirmed under perturbed walking conditions.

Conclusions:

• Phase-based gait cycle parameterization offers greater accuracy and robustness than time-based methods.
• Thigh phase angle is a preferred kinematic measure for gait analysis, particularly in dynamic environments.
• Findings support advanced control strategies for prosthetics and exoskeletons.