Looking at handwriting generation from a velocity control perspective.

This paper presents a model that explains the origin of the asymmetric bell-shaped velocity profiles generally observed in handwriting and other rapid movements. Applying the central limit theorem to describe the converging behavior of a sequence of dependent neural and muscular networks, it is shown that velocity profiles can be described by log-normal curves. An analysis-by-synthesis experiment is reported to support the model and to specify its mathematical implementation. Practical implications of this approach are discussed at the end of the paper to provide an analytical definition of a stroke, to clarify the concept of fluency and to suggest a powerful method for segmenting complex movements, particularly cursive script.