| 摘要 |
PROBLEM TO BE SOLVED: To realize a state estimation device that acquires multiple pieces of observation data about an object to be traced and uses likelihood obtained from each of the acquired observation data and reliability of the observation data to estimate an internal state of the object.SOLUTION: A first observation acquisition part acquires first observation data, and a second observation acquisition part acquires second observation data. A first likelihood acquisition part acquires first likelihood on the basis of the first observation data. A second likelihood acquisition part acquires second likelihood on the basis of the second observation data. A likelihood synthesis part acquires synthetic likelihood on the basis of the first likelihood, the second likelihood, first reliability data showing reliability of the first observation data and second reliability data showing reliability of the second observation data. A posterior probability distribution acquisition part acquires posterior probability distribution data being a probability distribution of an internal state of an observation target at current time t, from the synthetic likelihood and predictive probability distribution data.SELECTED DRAWING: Figure 1 |
