초록 ▼

AROTAN provides for autonomous terrain aided navigation that is fully independent of the position uncertainty of the vehicle during flight. AROTAN aligns a grid to the search space and periodically updates the search space during the measurement history to account for growth in position uncertainty.

AROTAN provides for autonomous terrain aided navigation that is fully independent of the position uncertainty of the vehicle during flight. AROTAN aligns a grid to the search space and periodically updates the search space during the measurement history to account for growth in position uncertainty. AROTAN computes terrain sufficiency statistics for a moving history to provide robust criteria for when to perform the correlation. A post-correlation refinement provides an additional correction of the horizontal position error and a correction of the altitude. AROTAN can quickly provide the vehicle's location based on the correlation from a single measurement history.

대표청구항 ▼

1. A method for autonomous terrain aided navigation of an airborne vehicle, comprising the steps of: a) loading a terrain height database comprising heights of terrain at given locations from a terrain reference onto the vehicle, said database formatted independently of a flight path of the airborne

1. A method for autonomous terrain aided navigation of an airborne vehicle, comprising the steps of: a) loading a terrain height database comprising heights of terrain at given locations from a terrain reference onto the vehicle, said database formatted independently of a flight path of the airborne vehicle over the terrain;b) initializing a two-dimensional search space about a horizontal position estimate of the airborne vehicle based on an uncertainty of the horizontal position estimate; and during flight of the airborne vehicle, c) using a radar altimeter and an inertial measurement unit (IMU) to gather a history of range-only measurements from the vehicle to the terrain and for each range measurement calculating range predictions from an updated altitude estimate and from terrain heights in the database for positions of the vehicle on an indexed grid aligned to the search space about an updated horizontal position estimate, and updating said search space based on the uncertainty of the updated horizontal position estimate;d) performing a correlation between the history of range-only measurements and the range predictions to determine a correlation offset on the grid, the orientation of the grid and spacing of the grid indices being fixed over the history for each said correlation;e) using the correlation offset to generate a position estimate correction to correct the horizontal position estimate of the vehicle and to correct its uncertainty; andf) using the horizontal position estimate to provide autonomous terrain aided navigation corrections to the flight of the airborne vehicle as the vehicle flies over the terrain. 2. The method of claim 1, wherein the range predications are calculated directly from the terrain heights in the database without transformation of the database to the altitude or heading of the flight path. 3. The method of claim 1, further comprising: one or more times over the history, computing a covariance matrix for the horizontal position estimate as its uncertainty to autonomously update the search space;without changing the orientation or spacing of the grid, adding indices to the grid to cover the search space; andcalculating range predictions for any additional indices in the grid for earlier horizontal position estimates in the history. 4. The method of claim 3, wherein the grid is aligned to the search space for each said history by, computing the covariance matrix at or just prior to starting to gather the history;computing first and second eigenvectors of the covariance matrix to define first and second principal axes of the search space for the horizontal position estimate;computing the first and second eigenvalues of the covariance matrix and multiplying them by a scale factor to determine an initial size of the search space;aligning axes of the grid to the first and second principal axes of the search space; anddefining the spacing between indices along the axes of the grid. 5. The method of claim 1, further comprising: computing a terrain sufficiency statistic for the history from the terrain heights at the updated horizontal position estimate or from a correlation metric; andperforming the correlation when the terrain sufficiency statistic satisfies a terrain sufficiency criteria. 6. The method of claim 5, wherein the terrain sufficiency statistic is computed for a moving window of the history that includes the last N range-only measurements where N is not greater than a chosen maximum length Nmax of the moving history, wherein the correlation is performed between only the last N range-only measurements and the last N range predictions. 7. The method of claim 1, further comprising: using the correlation offset to select a correlation history of range predictions and corresponding three-dimensional range prediction vector;computing a difference vector between the range measurements and the selected range predictions;computing line-of-sight unit vectors from the three-dimensional range prediction vector and correlation history of range predictions;using the difference vector and the line-of-sight unit vectors to compute a three-dimensional offset from the correlation offset that minimizes the squares of the differences over the history; andadding the three-dimensional offset to the correlation offset to provide a correction to the horizontal position estimate and the altitude estimate of the vehicle. 8. The method of claim 1, further comprising calculating the range predictions and performing the correlation iteratively by updating the search space with the corrected uncertainty, repositioning the search space about the correlation offset and reducing the spacing between grid indices. 9. The method of claim 1, further comprising: prior to flight, programming the vehicle with a flight path; andwithout reformatting for the flight path, loading the terrain height database onto the vehicle. 10. The method of claim 1, wherein the terrain height database is formatted by latitude and longitude. 11. The method of claim 1, further comprising: updating the horizontal position estimate at a radar altimeter clock rate;updating the correction and uncertainty of the horizontal position estimate at a filter clock rate less than or equal to the radar altimeter clock rate, andperforming the correlation upon satisfaction of a terrain sufficiency criteria synchronized to the filter clock rate. 12. The method of claim 1, wherein the range-only measurements provided by the radar altimeter represent a first return, wherein the range predictions are computed to model the first return. 13. A method for autonomous terrain aided navigation of an airborne vehicle, comprising the steps of: a) loading a terrain height database comprising heights of terrain at given locations from a terrain reference onto the vehicle, said database formatted independently of a flight path of the airborne vehicle over the terrain;b) initializing a search space based on a covariance of a horizontal position estimate;c) initializing and aligning an indexed grid to the search space;d) using a radar altimeter and an inertial measurement unit (IMU) to gather a moving history of the last N range measurements from the vehicle to the terrain at a radar altimeter clock rate, where N is not greater than a chosen maximum length Nmax of the moving history;e) receiving an updated horizontal position estimate and an updated altitude estimate of the vehicle at the radar altimeter clock rate;f) without transformation of the database to the altitude or heading of the flight path of the vehicle, calculating range predictions at the radar altimeter clock rate from the updated altitude estimate and terrain heights in the database for positions of the vehicle on the indexed grid about the updated horizontal position estimate;g) updating the covariance and the search space at a filter clock rate less than or equal to the radar altimeter clock rate;h) adding indices to the grid to cover the updated search space and calculating range predictions for each measurement in the moving history for the added indices;i) computing a terrain sufficiency statistic for the moving history from the terrain heights or from a correlation metric, if the statistic satisfies terrain sufficiency criteria, performing a correlation between the moving history of range measurements and the last N range predictions on the grid to determine grid indices of a correlation offset, the orientation and spacing of the grid being fixed over the history for each said correlation;j) using the correlation offset to provide a correction to the horizontal position estimate of the vehicle and to correct the covariance;k) resetting the history of range-only measurements with each said correlation; andl) using the horizontal position estimate to provide autonomous terrain aided navigation corrections to the flight of the airborne vehicle. 14. The method of claim 13, wherein the grid is aligned to the search space for each said history by, computing the covariance matrix at or just prior to starting to gather the history;computing first and second eigenvectors of the covariance matrix to define first and second principal axes of the search space for the horizontal position estimate;computing the first and second eigenvalues of the covariance matrix and multiplying them by a scale factor to determine an initial size of the search space;aligning axes of the grid to the first and second principal axes of the search space; anddefining the spacing between indices along the axes of the grid. 15. The method of claim 13, further comprising: using the correlation offset to select a correlation history of range predictions and corresponding three-dimensional range prediction vector;computing a difference vector between the range measurements and the selected range predictions;computing line-of-sight unit vectors from the three-dimensional range prediction vector and correlation history of range predictions;using the difference vector and the line-of-sight unit vectors to compute a three-dimensional offset from the correlation offset that minimizes the differences over the history; andadding the three-dimensional offset to the correlation offset to provide a correction to the horizontal position estimate and the altitude estimate of the vehicle. 16. The method of claim 13, wherein the moving history of measurements comprises only range measurements. 17. A method for autonomous terrain aided navigation of an airborne vehicle, comprising the steps of: a) loading a terrain height database comprising heights of terrain at given locations from a terrain reference onto the vehicle, said database formatted independently of a flight path of the airborne vehicle over the terrain;b) initializing a search space based on an uncertainty of a horizontal position estimate;c) using a radar altimeter and an inertial measurement unit (IMU) to gather a moving history of the last N horizontal position estimates, altitude estimates and range measurements from the vehicle to the terrain at a radar altimeter clock rate and, where N is not greater than a chosen maximum length Nmax of the moving history, without transformation of the database to the altitude or heading of the flight path of the vehicle, calculating range predictions from the altitude estimate and the terrain heights in the database for positions of the vehicle in the search space about each said horizontal position estimate;d) one or more times during the moving history, updating the search space based on a growing uncertainty of the horizontal position estimate and calculating range predictions for each measurement in the moving history for the additional search space;e) computing a terrain sufficiency statistic for the moving history from the terrain heights or from a correlation metric, if the statistic satisfies terrain sufficiency criteria, performing a correlation between the moving history of range measurements and the last N range predictions on the grid to determine a correlation offset;f) using the correlation offset to provide a correction to the horizontal position estimate of the vehicle and to correct its uncertainty;g) resetting the moving history of range measurements with each said correlation; andh) using the horizontal position estimate to provide autonomous terrain aided navigation corrections to the flight of the airborne vehicle. 18. The method of claim 17, further comprising: using the correlation offset to select a correlation history of range predictions and corresponding three-dimensional range prediction vector;computing a difference vector between the range measurements and the selected range predictions;computing line-of-sight unit vectors from the three-dimensional range prediction vector and correlation history of range predictions;using the difference vector and the line-of-sight unit vectors to compute a three-dimensional offset from the correlation offset that minimizes the differences over the history; andadding the three-dimensional offset to the correlation offset to provide a correction to the horizontal position estimate and the altitude estimate of the vehicle. 19. The method of claim 17, wherein the computation of the range predictions and the correlation are performed iteratively about the correlation offset with a decreasing and more finely resolved search space. 20. The method of claim 17, wherein the moving history of measurements comprises only range measurements. 21. An autonomous terrain aided navigation system for an airborne vehicle, comprising: a terrain height database comprising heights of terrain at given locations from a terrain reference onto the vehicle, said database formatted independently of a flight path of the airborne vehicle over the terrain;a radar altimeter and inertial measurement unit (IMU) configured to gather a history of range-only measurements from the vehicle to the terrain; andone or more processors configured to receive an initial two-dimensional search space about a horizontal position estimate of the airborne vehicle based on an uncertainty of the horizontal position estimate and, during flight of the airborne vehicle, for each range measurement to calculate range predictions from an updated altitude estimate and from terrain heights in the database for positions of the vehicle on an indexed grid aligned to the search space about an updated horizontal position estimate, and update said search space based on the uncertainty of the updated horizontal position estimate, perform a correlation between the history of range-only measurements and the range predictions to determine a correlation offset on the grid, the orientation of the grid and spacing of the grid indices being fixed over the history for each said correlation, and to use the horizontal position estimate to provide autonomous terrain aided navigation corrections to the flight of the airborne vehicle as the vehicle flies over the terrain.