Angular momentum: Insights into walking and its control

The importance of the organization of angular momenta during walking has been suggested by the efforts of researchers to use it to control and stabilize walking robots. However, there has been little attention to the use of angular momenta as a metric of human walking or to gain insights into the control of walking. This paper analyzes the angular momenta of the whole body (WBAM) and body segments of during walking. The normalized angular momenta about the body center of mass (CoM) of the body segments were computed about all three coordinate axes. The normalized angular momenta were small (<0.03) and well controlled in all groups. Principal component analyses (PCA) of the angular momenta suggested that the angular momenta of the body segments were not independent and more than 95% of the variability of the whole body angular momentum could be accounted for with only three principal components (PCs) for all groups. In addition, the PCs of each plane were found to be similar between individuals and independent of walking velocity and subject morphology. The framework of the uncontrolled manifold (UCM) hypothesis was used to get further insights into the control mechanisms of the WBAM. In particular we observed that in the sagittal plane, the central neural system (CNS) takes advantage of the segment redundancy to regulate the time profile of the WBAM from stride to stride. We found that during the stance phase, this regulation makes the WBAM repeatable from stride to stride while during the double support phase it was used to adjust the WBAM from stride to stride. Based on these results, it is shown that the WBAM is a variable whose value is regulated by the CNS during walking activities. Future work in robotics may wish to further investigate the use of similar control schemes that simulate human angular momentum organization.