DEVICE AND METHOD FOR IDENTIFYING A CYCLIC MOVEMENT AND CORRESPONDING COMPUTER PROGRAM

摘要

A device identifying a cyclic movement type of a physical system from a set of possible cyclic movements types includes: a movement sensor providing an observation data sequence concerning the physical system; a storage storing at least one statistical hidden-state Markov model; and a computer programmed to analyze the observation data sequence and select one cyclic movement type based on the stored statistical model. The storage includes least one statistical hidden-state Markov model for each possible cyclic movement type and the computer is programmed to analyze compatibility of the observation data sequence with each possible cyclic movement type based on the statistical model associated with this cyclic movement type. The computer is further programmed to provide a sequence of hidden states of the statistical model associated with the selected cyclic movement type, based on the observation data sequence and the statistical model associated with the selected cyclic movement type.