間欠的確定観測情報に基くカルマンフィルタ状態推定法

抄録

In the inertial navigation system, the accuracy decreases as the measuring time becomes longer because of the accumulation of errors. Kalman filter is usually used to prevent the error accumulation by combining the acceleration with some additional information such as the velocity of the system measured by, for example, a vehicle motion sensor. However, extra sensors make the system larger, heavier and more expensive. Fortunately, it is possible to know the velocity exactly without any velocity sensor if the system is at rest. Based on this obvious fact, we propose here an inertial navigation system that utilizes the velocity but has no velocity sensor, and describe the experiments performed to evaluate the usefulness of the proposed system.