Does lower-limb angular velocities scale linearly with walking speeds?

During human locomotion, lower limb segments form two kinematic chains linked at the plevis and oscillates back and forth repeatatively. The angular motion of each segment (foot, shank and thigh) is inherently constrained by the adjacent segments. Therefore, the movement for each segment is not totally independent and there exists a high degree of redundancy in the kinematic data for each segment due to the kinematic constraints as well as the 180° phase shift between the left and right leg. Previous work has demonstrated the kinematic co-variance between the elevation angles of lower limb segments [1]. Through principle component analysis (PCA), invariant features have been discovered that are independent of the walking speeds. However, Little is known about the relationship between waveforms at different speeds, or one step further, whether there exists a linear relationship between the human gait waveforms at one walking speed compared to another. If a linear relationship does in fact exist, the brain or central pattern generator (CPG) could use this simple linear model to regulate the limb kinematics to produce walking at a predefined speed. Therefore, the objective of this paper is to determine whether the lower limb angular velocities scale linearly with walking speeds.