Research Projects

Human-gait involves muscle activity and synergetic movements which makes the understanding of the mechanics a complex task. Prosthetics, so far, have limited capacity to restore the function of lost limbs. The semi-active or active prosthesis are mainly controlled by microprocessor based controllers or embedded systems based on sensorial inputs such as force, inertia or surface electromyographic (sEMG) signals. These controllers assist the amputee during the swing and stance phase of the gait cycle. However, these prosthetics limit movement in joints or create gait asymmetry that results in muscle disorders. There are several works related to myoelectric control providing knee joint movements in the prosthetics. Usually, a sEMG is acquired from the residual limb inside the prosthetic socket that is prone to movement artifacts. During sEMG acquisition, the residual limb undergoes dynamic pressure changes during gait cycle while causing electrode contact variation, noise addition and pain creation. The three main sets of muscles that provide movement, balance, and stability at the knee joint are popliteus, quadriceps and hamstrings muscles. Therefore, sEMG acquisition from these muscles particularly including the origin of these muscles (ischial tuberosity muscles) will reduce the signal artifacts. These sEMG signals together with the active components will enhance the user’s effort to actuate the knee joint which can improve the level of walking similar to healthy people. Moreover, the control system for knee and ankle joints are complicated and there are numerous challenges such as actuator redundancy, dynamics of actuators, delay in control stimuli and muscle fatigue. Hence, there is a necessity to develop a sophisticated control system to achieve a highly coordinated gait. It is proposed to develop an embedded system based transfemoral prosthesis that controls multi-axial knee and ankle motions upon a real-time EMG stimuli generated from a machine learning based decision making system. It is anticipated that the decision making system will reduce the computational time in providing a stimulus to knee actuators such that the knee dynamics, ground-foot interaction and body movement are achieved swiftly like normal people. This will make it an appropriate device for the transfemoral amputee. The bone and muscle combination making a body is a Parallel Manipulator equivalent. Hence, a 6-degrees of freedom device is proposed as an ankle support that when integrated with knee motion will enhance stability and control in walking. This research will help in improving mobility and provide a better lifestyle for people with lower limb amputation.