Design and experimental performance assessment of a hip flexion assistive device

More and more, stroke is becoming a leading cause of adult disability and, incidentally, a cause of death, especially in developed countries. A stroke is a life-changing accident : the consequences are dramatic on the victim's daily life, as it is common to develop a weakness or paralysis on one side of the body (hemiparesis or hemiplegia). Movement coordination is heavily affected, for example, and the road to rehabilitation is long and winding. An area that is particularly affected is walking. Recovering a healthy, symmetric gait after a stroke is extremely difficult, and represents a lot of work from both the patient and the physiotherapists. It is at this stage that robotics step in : some devices are able to perfect rehabilitation movements and reproduce them endlessly, without efforts from the therapist, and others are able to help patients at later stages of recovery, directly when they walk. This is where our project positions itself : aiming to restore symmetry in the gait. For this purpose, a certain number of devices have been developed. Some are passive, such as some ankle orthoses that simply block the ankle in a certain position to avoid a "drop foot" gait. Others are active, and actively contribute to the lower limb mobilization, such as active knee orthoses. Due to the high mobility of the hip joint, diverse patient morphology, as well as the lack of features in the area to which a device could be anchored to, the number of actuated hip orthoses available on the market is extremely low. This proves to be a problem, as hip mobility plays an important role during walking, and promotes mobility in the more distal joints. Based on this observation, we set off to design such an orthosis, that would be able to assist hip flexion amounting to 50% of the biomechanical needs. This first prototype focuses on the core specifications of this project, hip flexion assistance and control, but leaves the door open for more peripheral (at this stage) specifications, such as securing the device on a corset, or allowing abduction and adduction movements. Our design is lightweight, featuring new generation 3D-printed materials (total weight of the prototype, including motor : 3.27 [kg]). Its energetic contents have been optimized using a parallel spring, with characteristics determined by a genetic optimization algorithm. On the control side, an instrumented shoe sole prototype has been made to help time the motor assistance to the swing phase of the gait. Experimental torque control with relatively basic control schemes shows promising results.