Data fusion for a hand prosthesis tactile feedback system

Current myoelectrically controlled hand prostheses normally lack tactile feedback, thus limiting their functionalities and user acceptance. We propose a non-invasive tactile sensory feedback system, consisting miniaturized sensors, wireless communication module and vibrotactile actuators, aiming at providing a natural sense of touch to amputees. We model the tactile feedback chain from tactile sensors to actuators. This model includes a 2D hand model, a sensor data fusion model, and a phantom map model. We implemented the sensor data fusion model by different techniques and compared their performance.