原文传递 CELESTIAL AIDED INERTIAL NAVIGATION BY TRACKING HIGH ALTITUDE VEHICLES.
题名: CELESTIAL AIDED INERTIAL NAVIGATION BY TRACKING HIGH ALTITUDE VEHICLES.
作者: Kim, M. S.
关键词: Navigation systems, Celestial bodies
摘要: Celestial and inertial navigation systems have been used symbiotically within the area of alternative navigation solutions to Global Positioning System. Celestial systems normally provide attitude updates only by tracking known stars from a catalog, measuring their angular position with respect to the horizon and determining a deviation from the estimated vehicle attitude. However, by imaging reference objects with known positions and velocities against a background #12;of stars, the celestial system can triangulate its own position and velocity. With the ubiquitous use of aircraft, cooperative vehicles can be a ready source of information from which to make a reference. With a cooperative aircraft, it is assumed that the observer canget continuous updates about the reference vehicle via secure communication links. This thesis attempts to determine the navigation accuracy of a remotely piloted aircraft using celestial and inertial sensors and a barometric altimeter to track a second aircraft as the reference object. Simulations were performed of the navigating vehicle with #12;xed star trackers pointed directly up taking observations of the stars and Research Laboratory that simulates a navigating vehicle's flight for which inertial data is generated to provide a dead-reckoning estimate of the vehicle pose. The barometric altimeter provides frequent vertical dimension updates, with additional measurements given by the star trackers. The sensor measurements that are fed into the simulation must be generated externally, which is accomplished as part of this research. The script includes an extended Kalman #12;lter that propagates and updates the navigation state estimates based on the sensor measurements. The results show that stellar observations alone show a slight improvement in the position, velocity, and attitude estimates, but because the position and velocity are not directly updated, the estimates are still subject to large drift. In the second scenario, observations of the reference vehicle reduce the position error by over 99.9% from stellar observations only with a mean accuracy of 5.93 m. In the third scenario, the combination of both processes provides minor improvement in averageposition error, down to 5.67 m, but also improves the attitude error uncertainty by 87%. Additionally, the results show that the frequency of measurement updates has little impact on the navigation accuracy due to the use of a navigation grade inertial unit. No signi#12;cant changes in the navigation estimates were seen from varying the observation frequency from 1 s to 50 s. However at frequencies of 60 s and longer, the measurements are so sparse that other factors interfere with the ability to track the reference aircraft. the other aircraft to aid the inertial measurements, which is known to drift. Three different navigating scenarios were studied: 1) using a star sensor to get attitude updates only with just stellar observations, 2) using a star sensor to get position and velocity updates with just reference aircraft observations, and 3) using the the star sensors to get a complete navigation solution with both stellar and reference aircraft observations. The position, velocity, and attitude accuracy are compared between the three scenarios. Additionally, the frequency of celestial navigation measurements is varied between scenarios to determine its effect on navigation accuracy. The study makes use of an adaptive MATLAB script created by the Air Force
报告类型: 科技报告
检索历史
应用推荐