Book Description

From improving our technique in sports to providing feedback for rehabilitation therapy, wearable motion capture sensors have the potential to greatly enhance and improve our lives. While traditional camera-based systems have limited usage outdoors, wearable sensors have the ability to follow us from our workplace to our homes, and continuously provide feedback on how we’re moving, anywhere and at any time. One promising candidate for a wearable sensor is the dielectric elastomer (DE), a soft, flexible and highly stretchable polymer. In order to use DE sensors for motion capture, we need to be able to measure their capacitance, both accurately and efficiently. However, the large majority of low cost DEs have non-ideal electrode properties that cause problems for traditional capacitance sensing methods. Existing DE sensing methods were mainly developed for high voltage applications and lack the efficiency, safety and scalability to be implemented on a large scale, such as in a sensing suit. This thesis addresses these sensing challenges. First, we present a new low voltage hardware design that can measure the capacitance of multiple DEs at the same time. Then, in order to reduce computational power, we discuss the design of an efficient capacitance circuit that can increase the monitoring period of the sensor by an order of magnitude. We also quantify, for the first time, how the sensing frequency can affect the accuracy of the capacitance measurement. This new model presents a new design guide on how to select a suitable frequency for any sensor design. Afterwards, we demonstrate the ability to sense local capacitance changes within the sensor. This breakthrough significantly helps to reduce the amount of wires, connectors and sensing electronics for large sensor systems, a key step for increasing the scalability of these systems. Finally, we extend our method for strain mapping into two dimensions, producing a soft touch keyboard. The new low voltage capacitance sensing methods developed in this thesis are efficient, accurate and highly scalable. These improvements are key enablers that will allow DEs to be used as wearable motion capture sensors.