The idea to use a cost-effective pneumatic padding for sensing of physical interaction\nbetween a user and wearable rehabilitation robots is not new, but until now there has not been\nany practical relevant realization. In this paper, we present a novel method to estimate physical\nhuman-robot interaction using a pneumatic padding based on artificial neural networks (ANNs).\nThis estimation can serve as rough indicator of applied forces/torques by the user and can\nbe applied for visual feedback about the user�s participation or as additional information for\ninteraction controllers. Unlike common mostly very expensive 6-axis force/torque sensors (FTS), the\nproposed sensor system can be easily integrated in the design of physical human-robot interfaces of\nrehabilitation robots and adapts itself to the shape of the individual patient�s extremity by pressure\nchanging in pneumatic chambers, in order to provide a safe physical interaction with high user�s\ncomfort. This paper describes a concept of using ANNs for estimation of interaction forces/torques\nbased on pressure variations of eight customized air-pad chambers. The ANNs were trained one-time\noffline using signals of a high precision FTS which is also used as reference sensor for experimental\nvalidation. Experiments with three different subjects confirm the functionality of the concept and the\nestimation algorithm.
Loading....