With the widespread development of leg exoskeletons to provide external force-based repetitive training for gait rehabilitation, the prospect of undesired movement adaptation due to applied forces and imposed constraints require adequate investigation. A cable-driven leg exoskeleton, CDLE, presents a lightweight, flexible, and redundantly actuated architecture that enables the possibility of system parameters modulation to alter human–robot interaction while applying the desired forces. In this work, multi-joint stiffness performance of CDLE is formulated to systematically analyze human–CDLE interaction. Further, potential alterations in CDLE architecture are presented to tune human–CDLE interaction that favors the desired human leg movement during a gait rehabilitation paradigm.

In this paper, a parametric analysis of the inverse dynamics of an upright partially unloaded walking is performed. This motion is produced through a gait-training machine emulating the over-ground walking using a body weight support mechanism and a cable-driven robot. The input motion is the kinematics of a normal gait, and the ultimate output result is the required tensions to be generated by the cable robot in order to drive the lower limb. The dynamic analysis is carried out based on the Newton–Euler approach. A Matlab Simscape model is also built to validate the analytical results. The obtained dynamic model is used to investigate the effect of the variation of the gait simulation parameters on the actuation wrench and the cable tensions. The obtained results could be used to determine the optimal design of the gait trainer actuators and they are useful in estimating optimal gait training parameters.