UC Berkeley

Over 265,000 individuals in the United States have a spinal cord injury. The wheelchair is currently the most prescribed mobility option for this population. Robotic exoskeleton devices are a promising alternative means of mobility that enable a paraplegic user to regain the physical and emotional health benefits of standing and walking. This dissertation will discuss novel gait strategies that were designed for a minimally-actuated medical exoskeleton device.

Two novel stance control strategies will be presented that are designed to propel the exoskeleton and pilot forward out of double-stance. The first stance strategy relies on the movement of the pilot's torso as a means to dynamically propel the system forward through a step. Second, a more advanced strategy that propels the user forward while stabilizing their torso will also be proposed. Through the manipulation of lower extremity kinematics, the user can be propelled forward out of double-stance with an exoskeleton that only has actuation at the hips. Sequential methods of trajectory generation will be described for both strategies. Experimental results will be presented that support the efficacy for these stance strategies.

In addition, this dissertation proposes an original sequential swing phase model. This model is used to produce a gait trajectory for the swing leg that dynamically controls an unactuated swing knee. A method for swing phase hip trajectory generation will be presented. Experimental results will be shown that validate the efficacy of the swing phase trajectory at producing a natural human-like swing phase from a swing leg that has no actuation at the knee.

Extensive pilot testing of the medical exoskeleton in both structured and unstructured environments was completed. The results from these tests support the efficacy of the control algorithms and show great promise in the minimally actuated exoskeleton paradigm. This work enables a reduction in exoskeleton hardware complexity, reduced weight, and anticipated lower costs for end users so that more spinal cord injury patients may have greater access to this technology.