| dc.description.abstract | Historically, the vast majority of prescribed transfemoral prostheses have been passive or modulated passive devices, which is to say that they are from the user’s perspective net energy negative. Such devices have proven to be highly effective for persons with amputation navigating the world around them. They perform a large subset of ambulatory tasks well, particularly those involving “ballistic” movements, wherein a person can achieve appropriate knee motion without any energy addition at the knee, simply through healthy hip movement and passive double pendulum dynamics. Representative activities include level and down slope walking. However, such devices are fundamentally incapable of allowing many other so called “non-ballistic” movements, which are swing phase motions wherein the desired knee motion cannot be generated via inertial coupling in combination with healthy hip movement. These tasks, which include activities such as stair climbing, upslope walking, and stumble/scuff perturbation recovery, are significantly more difficult for users, if not impossible, on energetically passive devices.
Over the last several decades, the field of powered prosthetic devices has flourished as a research space to address limitations in energetically passive knee prostheses. These devices use a mechatronic actuation system to provide both dissipation and power addition at the knee joint. However, because these devices are designed with a mechanical system meant to handle both additive and resistive torques (which can be significantly higher), they are required to have large motors and/or large transmissions. As a result, these devices are often high output impedance (relative to requirements of ballistic swing movement) and are difficult to back drive, disabling their users from performing the aforementioned ballistic tasks through natural inertial dynamics. This produces not only a less smooth feeling of movement, but a system where the user is less in charge of the device.
The focus of this dissertation is a reevaluation of the approach of powered prosthetic knees – namely the idea that the best solution is a single, mechatronic system which can handle all the torques required of human movement. Instead, a system is proposed, designed, and tested which uses a small, easily back drivable mechatronic system for active power in parallel with a controllable hydraulic dissipation system which can provide the maximum dissipation required at the knee when desired, but also allow quasi-free movement when necessary. A prototype of this description is designed, fabricated, and experimentally evaluated with regard to the prospective value of this “Stance-Controlled Swing-Assisted” approach to ambulation activities, including assessments of multi-speed walking, multi-slope walking, stair ascent, stair descent, scuff recovery, and stumble recovery. | |
