A sliding mode control model for perturbed upright stance in healthy subjects

Human upright stance and balance maintenance in quiet conditions have been extensively evaluated throughout the years. However, relatively less information is available on how the central nervous system (CNS) acts to maintain balance after sudden perturbations of stance. Here, a sliding mode control (SMC) model for the characterization of balance maintenance after external perturbations is proposed. Human stance was modeled as an inverted pendulum (IP), which describes kinematics in the sagittal plane; the choice of a SMC allowed to avoid model linearization, commonly employed when using a single-link IP for bipedal stance modeling, thus providing a more accurate description of the human-stance system dynamics. Model was applied on experimental data obtained from perturbed stance trials consisting of a series of disruptions of the same magnitude. This experimental condition was able to elicit a well-known feature called “habituation rate”, which refers to the subject capacity to self-adapt his/her responses to identical perturbations. SMC parameters were identified through a robust optimization procedure. Results showed limited tracking errors for center of mass displacement. One of the SMC parameters exhibited a clear trend from the first to the last trial, appearing able to quantify the habituation rate effect. The application of such a control model to the non-quiet stance can provide additional information in understanding how the CNS tailors balance responses in different conditions.