Methods and apparatus for navigational aiding using celestial object tracking

Patent metadata

Field	Value
Publication number	US-9702702-B1
Application number	US-201615152104-A
Country	US
Kind code	B1
Filing date	May 11, 2016
Priority date	Jun 15, 2015
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

Methods and computer products for establishing at least one of attitude, direction and position of a moving platform. At least one skymark of known ephemeris is imaged at each of a first set of discrete instants by means of an optical sensor coupled to the platform. A measurement is also obtained of the attitude of the platform at each of a second set of discrete instants by means of an inertial navigation system. A recursive of estimation filter is then applied to successive skymark position vectors to update an estimate of platform navigation state, with the measurement of attitude of the platform tightly coupled to the estimate of platform navigation state as updated by the recursive estimation filter. Various techniques of frame stacking and multi-hypothesis tracking may be applied to improve the robustness of navigation solutions.

First claim

We claim:

1. A method for establishing at least one of attitude, direction and position of a platform in motion in an inertial frame, the method comprising:
a. optically imaging at least one skymark of known ephemeris at each of a first set of discrete instants by means of a sensor coupled to the platform to obtain successive skymark position vectors;
b. obtaining a measurement of attitude of the platform at each of a second set of discrete instants by means of an inertial navigation system;
c. combining a first of the successive skymark position vectors and the measurement of attitude of the platform at one of the second set of discrete instants to obtain an estimate of platform navigation state; and
d. applying a recursive estimation filter to successive skymark position vectors to update the estimate of platform navigation state; wherein the measurement of attitude of the platform is tightly coupled to the estimate of platform navigation state as updated by the recursive estimation filter,

wherein the recursive estimation filter employs multi-hypothesis tracking.

2. A method in accordance with claim 1 , wherein the second set of discrete instants substantially overlaps the first set of discrete instants.

3. A method in accordance with claim 1 , wherein the recursive estimation filter is a Kalman filter.

4. A method in accordance with claim 1 , wherein the recursive estimation filter is an unscented Kalman filter.

5. A method in accordance with claim 1 , wherein the at least one skymark comprises at least one satellite.

6. A method in accordance with claim 1 , wherein multi-hypothesis tracking includes retaining multiple hypotheses regarding assignment of bright spots to skymarks during recursive estimation.

7. A method in accordance with claim 1 , further comprising separately summing successive image frames in parallel pipelines, wherein a first pipeline compensates for predicted motion and a second pipeline does not compensate for predicted motion.

8. A method in accordance with claim 7 , wherein the predicted motion includes motion of a skymark.

9. A method in accordance with claim 7 , wherein the predicted motion includes motion of the platform.

10. A method in accordance with claim 1 , wherein applying a recursive estimation filter comprises updating a set of states of a state vector at successive instants.

11. A method in accordance with claim 10 , wherein updating the set of states includes updating a proper subset of the set of states of the state vector.

12. A method in accordance with claim 10 , wherein updating the set of states includes at least one of appending a state to the state vector and deleting a state from the state vector.

13. A method in accordance with claim 12 , wherein appending a state to the state vector corresponds to entry of a satellite into a specified field of view.

14. A method for establishing at least one of attitude, direction and position of a platform in motion in an inertial frame, the method comprising:
a. optically imaging at least one skymark of known ephemeris at each of a first set of discrete instants by means of a sensor coupled to the platform to obtain successive skymark position vectors;
b. obtaining a measurement of attitude of the platform at each of a second set of discrete instants by means of an inertial navigation system;
c. combining a first of the successive skymark position vectors and the measurement of attitude of the platform at one of the second set of discrete instants to obtain an estimate of platform navigation state; and
d. applying a recursive estimation filter to successive skymark position vectors to update the estimate of platform navigation state; wherein the measurement of attitude of the platform is tightly coupled to the estimate of platform navigation state as updated by the recursive estimation filter,
e. identifying at least one bright spot detected by the sensor;
f. attributing the bright spot to a putative celestial object and employing the estimate of platform navigation state to predict a position of the bright spot in a subsequent image frame provided by the sensor;
g. detecting the putative celestial object at a subsequent time; and
h. discriminatingly selecting the bright spot as a skymark for purposes of successive applications of the recursive estimation filter based on congruence of an actual position in the subsequent image frame relative to the predicted position.

15. A computer program product for use on a computer system for establishing at least one of attitude, direction and position of a platform in motion in an inertial frame, the computer program product comprising a non-transitory computer readable medium having computer readable program code fixed thereon, the computer readable program code including:
a. program code for inputting an image of at least one skymark of known ephemeris at each of a first set of discrete instants by means of a sensor coupled to a platform to obtain successive skymark position vectors;
b. program code for inputting a measurement of attitude of the platform at each of a second set of discrete instants; and
c. program code for combining a first of the successive skymark position vectors and the measurement of attitude of the platform at one of the second set of discrete instants to obtain an estimate of platform navigation state; and
d. a recursive estimation filter adapted to update the estimate of platform navigation state,

wherein the recursive estimation filter employs multi-hypothesis tracking.

16. A computer program product in accordance with claim 15 , wherein multi-hypothesis tracking includes retaining multiple hypotheses regarding assignment of bright spots to skymarks during recursive estimation.

17. A computer program product for use on a computer system for establishing at least one of attitude, direction and position of a platform in motion in an inertial frame, the computer program product comprising a non-transitory computer readable medium having computer readable program code fixed thereon, the computer readable program code including:
a. program code for inputting an image of at least one skymark of known ephemeris at each of a first set of discrete instants by means of a sensor coupled to a platform to obtain successive skymark position vectors;
b. program code for inputting a measurement of attitude of the platform at each of a second set of discrete instants; and
c. program code for combining a first of the successive skymark position vectors and the measurement of attitude of the platform at one of the second set of discrete instants to obtain an estimate of platform navigation state; and
d. a recursive estimation filter adapted to update the estimate of platform navigation state,

wherein the computer-readable program code further comprises program code for separately summing successive image frames in parallel pipelines, wherein a first pipeline compensates for predicted motion and a second pipeline does not compensate for predicted motion.

Assignees

• Charles Stark Draper Laboratory Inc

Inventors

• Lane Benjamin F
• Whitacre William W

Classifications

• G06T2207/10032

  Satellite or aerial image; Remote sensing · CPC title

• G06T7/0044

  Physics · mapped topic

• G06T7/204

  Physics · mapped topic

• G01C21/025 Primary

  with the use of startrackers · CPC title

• G06T2207/20024

  Filtering details · CPC title