Phd in INDUSTRIAL AND MANAGEMENT ENGINEERING

Thesis title: Design and validation of a solution for restoring proprioception in patients with trans-tibial prostheses using elastomeric sensors and vibrotactile feedback

Lower limb amputation is a widespread problem worldwide. Amputee subjects present an asymmetrical gait, greater metabolic consumption and reduced speed. Furthermore, the lack of proprioception of the missing limb causes an increased risk of falling. Thus, it is important to devise new ways to provide sensory feedback to the amputee. Currently, one of the most used methods is vibrotactile feedback. Vibrotactile feedback is simple to implement, easy to control, and inexpensive. Force sensors placed under the sole of the foot are often used to obtain the information to be supplied to the feedback device. However, recently new technologies for biomechanical analysis of the lower limbs are emerging and one of them is elastomeric strain sensors. They are easy to make and can have complex geometries. Furthermore, thanks to their versatility they are an excellent solution for the analysis of deformations and loads. In this thesis the design and the realization of a proprioceptive device for trans-tibial prostheses will be exposed. The qualitative test of the selected vibratory motors will be reported, showing excellent results. The subjects identified with good accuracy values ​​both the different levels of vibration intensity of the motors positioned on the leg, and a walking pattern recreated through vibration. Subsequently, through a finite element analysis, the point of greatest deformation of the cosmetic covering of the prosthesis on the upper part of the ankle was identified with a deformation of about 20 mm/m. With this information, the mechanical characterization of a 3D printed elastomeric sensor filled with liquid metal was performed. The sensor was characterized statically in a strain range of ± 5 mm and dynamically in a range of ± 1 mm for ten frequencies from 1 Hz to 10 Hz. A thermal characterization was also carried out. The sensor showed a linear graduation curve with a good sensitivity (34.35 mV/mm) and a good dynamic response in the analyzed frequency range.