Temporal sequence recognition using uncertain sensor data

The problem addressed in the paper concerns temporal sequence recognition for a dynamic system. Several formal models can be used such as rule based systems, or graphs such as transition graphs or Petri nets in order to describe the sequences to be recognized. Then, according to the inputs obtained from the system's sensors at different times, the goal is to evaluate confidence into the fact that the sequence is in progress. The confidence is modeled by a distribution of mass of evidence proposed in Dempster-Shafer's theory.