Advance Model for Capturing Real Life Human Gait Process

Human gait represents a highly coordinated multi-dimensional and energy efficient process involving complex precision control mechanisms. Several attempts have been made in the literature to capture every minute detail of this process and develop accurate models. Although available state of art neuromuscular models demonstrate higher degrees of accuracy, the extent to which the shoulder muscles actively drive the arms, their effect on stability and economy during gait are not well established till date. Most of these models are sufficiently accurate to replicate the human gait in upright position, but fail to capture the energy efficiency and analysis while in a bent position such as the start-up posture just before a running event. Moreover performance of existing models degrade while capturing motions around a smooth turn. The prime objective of this work is to clearly bring out the effect of arm swing and posture on the energy efficiency of human gait process. This work can be a potential enhancement to performance of existing state of art neuro-musculoskeletal models, thereby reducing energy expenditure by approximately 7.89%. In this work we present a simple and systematic methodology for deriving the control system model of human gait considering the challenges faced in previous models and includes advanced effects encountered in real life. Although the single inverted pendulum is widely accepted as an adequate model of bipedal motion, but creates accuracy as well as stability issues and is less likely to capture advance dynamics of the human gait process. In addition to the motion of ankle joints, human gait often involves the motion of hip and knee joints for improved balancing, increased flexibility in face of the multitude external disturbances and robustness in terms of fail safe. For optimized results, a multi-pendulum model with forward dynamics approach has been considered in this work. In order to achieve real time performance with good controllability, LQR controller with state feedback techniques has been adapted in the model. Typical observations like swinging of hands out of phase with respect to legs, effect of posture prior to a running event are also analyzed and included into the model. We investigate the control and function of arm swing in human gait process to test three competing hypotheses i.e. (1) The arms are actively driven by shoulder muscles, (2) The arms are passively powered by movement of the lower body, (3) During few initial steps of gait arm movement is actively driven by shoulder muscles and consequently by passive dynamic effect of the thorax, inertia and gravity. Effects of removing arm swing that create stability problems during walking and especially running, resulting in greater variability in footfall positions are also analyzed. A comparative analysis between distance covered, maximum velocity achieved, effort on foot for the same input torque at the hip joint, and energy efficiency computations (work done per step per meter) is carried out for the above mentioned cases with and without hand motion during the gait process. This work finds potential application in development of energy efficient automated robots usually employed in industries, biomimetic, prosthetic, neuro-rehabilitation engineering and sports biomechanics where the energy efficiency and performance under varying postures are at priority. It drives gait modelling methodology towards an advanced low constrained multidimensional approach as is required by modern high end systems and compromise between energy efficiency and speed. This model can be cleverly utilized to suggest the best initial posture for different athletes having different body structures to obtain maximum speed efficiently. Strategic approach towards the development of a flexible and an accurate gait model are analyzed and discussed in detail.