Modelling and Analysis of Coupling Dynamics of a Lower Limb Exoskeleton

Abstract Exoskeleton is a promising technology to enhance the mobility of aged and disable people. With comfortable human-exoskeleton interaction, mechanical and control designers aim at accurately tracking a desired trajectory, while saving the wearer's energy expenditure is the preference from the viewpoint of biomechanics. Given all of these effects, we propose a new model, consisting of the swing dynamics of both human and robot's lower extremities coupled by damped springs representing elastic and viscous properties of band and human tissue. With the coupling coefficients identified with an experimental platform, the analytical model yields consistent results compared with an experimental exoskeleton, especially in the prediction of the interactive forces. Further analyses are then performed based on this validated model, revealing the influences of desired trajectory, mass ratio, misalignment, coupling points, health condition and band tightness on the human-exoskeleton coupling dynamics. It is found that introducing gravity compensation and tuning the feedback gain improve the tracking accuracy, but hardly change the interactive force. The most comfortable interaction requires a healthy wearer coupled with a light-weight exoskeleton without any misalignment, but properly changing the trajectory, coupling points and tightness can partly reduce the interactive forces if the ideal condition is unachievable.