Posture Identification of Musicians Using Non-Intrusive Low-Cost Resistive Pressure Sensors

The following paper documents the creation of a prototype of shoe-soles designed to detect various postures of standing musicians using non-intrusive pressure sensors. In order to do so, flexible algorithms were designed with the capacity of working even with an imprecise placement of the sensors. This makes it easy and accessible for all potential users. At least 4 sensors are required: 2 for the front and 2 for the back; this prototype uses 6. The sensors are rather inexpensive, widening the economic availability. For each individual musician, the algorithms are capable of ``per- sonalising'' postures in order to create consistent evaluations; the results of which may be, but are not limited to: new musical interfaces, educational analysis of technique, or music controllers. In building a prototype for the algorithms, data was acquired by wiring the sensors to a data-logger. The algorithms and tests were implemented using MATLAB. After designing the algorithms, various tests were run in order to prove their reliability. These determined that indeed the algorithms work to a sufficient degree of certainty, allowing for a reliable classification of a musician's posture or position.