Ridge regression signal processing for position-fix navigation systems

• US 4,876,550 A
• Filed: 10/08/1987
• Issued: 10/24/1989
• Est. Priority Date: 10/08/1987
• Status: Expired due to Fees

• Current Assignee: Alliedsignal Inc. (Honeywell International Inc.)
• Original Assignee: Alliedsignal Inc. (Honeywell International Inc.)
• Inventors: Kelly, Robert J.
• Primary Examiner(s): Tarcza, Thomas H.
• Assistant Examiner(s): CAIN, DAVID C
• Application Number:
  US07/107,430
• Time in Patent Office: 747 Days
• Field of Search: 342/461, 342/450, 342/451, 364/443, 364/446, 364/449
• US Class Current: 342/451
• CPC Class Codes: G01S 13/876 Combination of several spaced transponders or reflectors of known location for determining the position of a receiver G01S13/874 takes precedence
• CPC Class Codes: G01S 2205/03 Airborne
• CPC Class Codes: G01S 5/02 using radio waves using satellite radio beacon systems for determining position G01S19/00

Abstract

A position estimator for determining the position and velocity of a moving platform in cooperation with radio navigation aids is described incorporating an unbiased estimator, such as a least means square estimator, a biased estimator for determining the angle of inner section of the lines of position from the radio naviation aids for determining the liklihood of geometric dilution of precision (GDOP) and a switch for selecting the estimate of position and velocity from said biased estimator at first times and the unbiased estimator at second times. The invention overcomes the problem of accuracy degredation associated with a nearly collinear measurement geometry which causes the variance of the position estimates to be higly inflated.

Claims

1. Apparatus for determining the position and velocity of a moving platform comprising:

a radio navigation aid having at least three portions, two of which are distant from said platform and having a known position for providing to said platform a plurality of samples indicative of the position of said platform at respective times,

a position estimator for receiving said plurality of samples for generating an estimate of position and velocity both with an unbiased estimator wherein the mean square error of the estimate is the variance and with a biased estimator and selecting said estimate of position and velocity from said biased estimator at times said geometry of said moving platform is substantially co-linear with said two distant portions of said radio navigation aid.

2. The apparatus of claim 1 wherein said biased estimator is a Ridge estimator.

3. The apparatus of claim 1 wherein said position estimator includes a switch coupled to the outputs of said unbiased estimator and said estimator biased for coupling one of said outputs to an output terminal of said position estimator in response to a control signal.

4. The apparatus of claim 3 wherein said position estimator further includes a geometric collinear estimator for determining the presence of collinearity having an output coupled to the control input of said switch.

5. The apparatus of claim 4 wherein said geometric collinear estimator estimates the angle between two lines of position and for providing an output signal to the control input of said switch at times said angle is less than a first predetermined value and at times said angle is greater than a second predetermined value.

6. The apparatus of claim 1 wherein said position estimator includes a means for calculating the Ridge parameter K having a input coupled to the output of an unbiased estimator and wherein the output of said means for calculating is coupled to an input of said bias estimator.

7. The apparatus of claim 1 further including an adder coupled to the output of said position estimator having a second input adapted for coupling to position commands, the output of said adder coupled to the input of an autopilot, said autopilot adapted for coupling to said moving platform to direct said moving platform along a disired flight path.

8. Apparatus for determining the position and velocity of a moving platform in cooperation with at least two distant radio navigation aids spaced apart in distance by a predetermined amount comprising:

means for receiving a plurality of samples from said radio navigation aids indicative of the position of said platform at respective times, means for determining the lines of position to each navigation aid and the angle of intersection of said lines of position at said platform,

means for generating an estimate of position and velocity from said plurality of samples with an unbiased estimator wherein the mean square error of the estimate is the variance,

means for generating an estimate of position and velocity with a biased estimator, and

means for selecting said estimate of position and velocity from said biased estimator at times said angle of intersection of said lines of position are less than a first predetermined value or greater than a second predetermined value and for selecting said estimate of position and velocity from said unbiased estimator at other times.

9. The apparatus of claim 8 further including means for calculating a Ridge parameter K as a function of the output of said unbiased estimator, said Ridge parameter K coupled to an input of said biased estimator.

10. The apparatus of claim 8 wherein said biased estimator is a Ridge estimator.

11. A method for determining the position and velocity of a moving platform in cooperation with at least two distant radio navigation aids spaced apart in distance by a predetermined amount comprising the steps of:

receiving a plurality of samples from said radio navigation aids indicative of the position of said platform at respective times, determining the lines of position to each navigation aid and the angle of intersection of said lines of position at said platform,

generating an estimate of position and velocity from said plurality of samples with an unbiased estimator wherein the mean square error of the estimate is the variance,

generating an estimate of position and velocity with a biased estimator, and

selecting said estimate of position and velocity from siad biased estimator at times said angle of intersection of said lines of position are less than a first predetermined value or greater than a second predetermined value and for selecting said estimate of position and velocity from said unbiased estimator at other times.

12. The method of claim 11 including calculating a Ridge parameter K as a function of the output of said unbiased estimator, said Ridge parameter K coupled to an input of said biased estimator.

13. Apparatus for determining the position and velocity of a moving platform in a position-fix navigation system having known first and second transmitter locations at times the geometry of said transmitter locations with respect to the estimated position of the receiver on a moving platform is nearly collinear comprising:

first means for receiving and storing a plurality of n samples indicative of range each spaced apart by an intersample time interval Δ

T_S and occurring within time interval Δ

T_M from each of said first and second transmitters having a known distance therebetween,

second means for determining the initial position X_O, Y_O of said platform from said plurality of n samples from each of said first and second transmitters where ##EQU50## where R₀₁ is the distance to the first transmitter at time T₁, R₀₂ is the distance to the second transmitter at time T₂,

and X_C1 is the position of said first transmitter,

third means for determining ##EQU51## +10 where ##EQU52## fourth means for determining H^T_H where ##EQU53## +10 where A=cos² 74 ₁ +cos² θ

₂

B=cosθ

₁ sinθ

₁ +cosθ

₂ sinθ

₂

C=sin² θ

₁ +sin² θ

₂

fifth means for determining H^T H^-1,

sixth means for determining H^T δ

R ##EQU54## seventh means for determining β

_OLS where β

_OLS =(H^T H)^-1 H^T δ

R,

eighth means for determining β

_OLS^T,

ninth means for selecting K_OP where ##EQU55## +10where P is the number of variables and S² is the sample standard deviation where ##EQU56## +10and tenth means for determining β

_R where β

_R =(H^T H+K_OP I)^-1 H^T δ

R to provide [δ

X, δ

Y, V_x, V_y ].