Thevbow: an expressive musical controller haptic human-computer interface

This dissertation describes the development of the vBow, a virtual violin bow controller. This interface was designed to accurately sense the component physical motions of a violinist's bowing gesture, while providing the performer with both the auditory feedback of the sound synthesis and the tactile sensations of the haptic feedback, produced by the system. Both the sound synthesis and haptic feedback are generated by software which uses the sensor readings, from the encoders on the vBow, as parameter data for a bowed string physical model, a friction model, and simulations of detents, elasticity, and barriers, produced by the motors on the vBow. The vBow was designed around the component physical motions of the bowing gesture, with each encoder sensing the movement, and each servomotor producing the tactile feedback, that corresponds to the four main trajectories, or degrees of freedom, of the bowing gesture. On an acoustic violin, the manipulation of each of these four main trajectories of bowing contributes to the expression inflected in the tone of the violin. Similarly, with the vBow, maneuvering each of the four degrees of freedom generates data from each of the four encoders, which affect the parameters of the sound synthesis software, producing expressive variations in the timbre. Just as the four component physical motions of bowing contribute to variations in the timbre produced by both an acoustic violin and the vBow, each degree of freedom also contributes to the tactile feedback of both systems. For an acoustic violin, each trajectory of motion produces the tactile feedback of vibration, friction, detents, or elasticity. On the vBow, each encoder is attached to a servomotor, which generates the same haptic feedback that corresponds to the bowing motion producing the encoder reading. The encoders and servomotors used by the vBow were selected for their high resolution of sensing and wide range of force, to maximize the expressive potential of the instrument. Similarly, the sound synthesis and haptic feedback models were chosen because of their responsiveness and flexibility.