Modeling human hand movement with motor signature based on velocity segment

Recent experiments demonstrate that rhythmic movements of human hand exhibit individual, idiosyncratic motion characteristics in the solo condition. It is undoubtedly promising to computationally produce the individualized hand movement for its great potential in motor rehabilitation. In this paper, we develop a computational approach that is able to spontaneously generate the hand movement of a prescribed human participant in the paradigm of mirror game. The profile of velocity segment is regarded as a marker of motor signature, which allows us to distinguish solo movements produced by different human participants. The base segment of velocity is thus constructed to describe motor signature and give birth to actual velocity segments. In particular, a feedback-based strategy is designed to assemble the actual velocity segments and produce the continuous, smooth trajectory while preserving individual motor signature. The proposed algorithm has been validated by matching experimental data. This work opens a new venue to investigating motor coordination among human participants.