Fine grained control of robotic calligraphy

The existing methods of robotic Chinese calligraphy either apply industrial programming techniques to teach a robot some specific characters on whole, or take brush-strokes as basic writing units to pursue agile calligraphy. However, the former is too stiff while the latter is difficult to reach a great similarity. To achieve both similarity and flexibility, we consider that finer grained control of robotic calligraphy is needed. Motivated by it, we present a new framework and method for controlling robotic calligraphy in fine granularity. It focuses on brushstroke primaries, more basic writing elements, approaching to the problem of low accuracy in local robotic writing. First, we study the effect of writing route, pressing depth and brush lag on brush footprints. Second, the writing route and the corresponding stroke information acquired from a character image are transformed into the writing actions of a robot. Finally, we conduct extensive experiments to validate the presented method and the results show that it can achieve satisfactory effect on robotic calligraphy.