Data processing: vehicles – navigation – and relative location – Vehicle control – guidance – operation – or indication – Aeronautical vehicle
Reexamination Certificate
2001-06-04
2003-04-29
Nguyen, Tan Q. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Vehicle control, guidance, operation, or indication
Aeronautical vehicle
C701S004000, C244S002000, C342S062000
Reexamination Certificate
active
06556895
ABSTRACT:
FIELD AND BACKGROUND OF THE INVENTION
The present invention relates to in-flight transfer alignment of the inertial navigation system (INS) of a carried vehicle which is launched from a carrier vehicle and, in particularly to the alignment procedure in the presence of unknown delays in the measurements provided by the INS of the carrier vehicle, with respect to the measurements taken by the INS of the carried vehicle.
Airplanes often carry with them other flying vehicles, such as unmanned airplanes or missiles, hereinafter; carried vehicles, which are to be launched during flight.
Once the carried vehicle is launched, it will use its own internal measurement unit (IMU) to provide data to its autonomous INS. In order to do so correctly, the IMU of the carried vehicle has to be calibrated and aligned prior the launch.
The calibration and alignment is performed during flight, while the two vehicles are still attached. In this procedure, called flight transfer alignment, the velocity of the carried vehicle as computed by its INS according to the data supplied by its IMU, is compared to the velocity supplied by the aircraft's INS, which is regarded to be error free and is therefore used as reference.
The difference between these two velocities which is attributed to the misalignment of the carried vehicle with respect to its carrier and to the offset of the gyroscopes and accelerometers of its IMU, is processed by a computerized statistical procedure known as a Kalman filter which yields the misalignment data of the carried vehicle with regard to its carrier as well as the carried IMU drift and bias data.
This procedure, of aligning during flight of the carried vehicle's INS with the carrier's INS by comparing their velocities as supplied by the measurements of their independent IMUs, is known as transfer alignment (TA).
While the estimation of the level misalignment angles (pitch, &thgr; and roll, &phgr;), illustrated in
FIG. 1
to which reference is now made, is accomplished with no difficulties, the azimuth misalignment (yaw angle, p) requires that the carrier vehicle perform alignment maneuvers which enable the Kalman filter TA to achieve observability of the yaw angle, and to separate between tilt errors and accelerometer biases, which otherwise tend to compensate each other.
Generally, flight trajectory during the TA may include four flight segments: The first flight segment is a straight and level flight for the purpose of initial leveling. The second segment whose purpose is the azimuth alignment, consists of a trajectory which is C shaped in parallel to ground. The third segment, for the purpose of final leveling, is again a straight and level flight at the same direction as in the first segment, and the last section, sometimes omitted is used to simulate the navigation of the carried vehicle after the transfer alignment.
It has been shown, however, that the use of the information obtained during the second segment of the transfer alignment flight, which is essential for the estimation of the azimuth misalignment error, introduces a large position error.
The reason for it as explained by Bar-Itzhak and Vitek (I. Y. Bar-Itzhak and Y. Vitek, “
The enigma of false Bias Detection in a Strapdown System During Transfer Alignment
”, J. Guidance, 8, 175 (1985)), is a time delay between the data produced by the cartier's INS and the data of the carried vehicle's INS, which occurs because both systems operate with independent clocks which are not synchronized.
It was further shown by Bar-Itzhak and Vitek that this undefined constant time delay, which was introduced into the measurements, resembles a signal, which only a longitudinal accelerometer bias would have made.
Thus, the Kalman filter algorithm whose model is based on a dynamic system which does not model the influence of the time delay, identifies the signature of the time delay in the measured signal as the signature of the longitudinal accelerometer bias. Consequently the Kalman filter produces an excessive estimate of the accelerometer bias.
It is evident that such synchronization errors arise only during the second segment of the transfer alignment flight, where the data changes due the lateral acceleration in the flight maneuver, where as during the straight and level flight (in segments I and III), where the velocity readings of the INSs are constant, (except for white noise), no time delay effects exist.
Because the trajectory of the second segment is required for the yaw-angle observability, and because the synchronization of the data during this segment requires an extra effort, it is desired to have a method for TA, which overcomes the uncertainties which are introduced by the time delay.
SUMMARY OF THE PRESENT INVENTION
We have comprehended the fact that synchronization errors affect TA only when the velocity measurements of the carrier and of the carried vehicle are compared while the aircraft is accelerating (during the second segment of the alignment flight). Furthermore, we are aware of the prior art, which considered flight in the second segment to be essential for a complete TA and required that velocity errors be measured during flight in this segment.
We have discovered, that the important factor about TA is the fact that there exists a lateral acceleration and thus velocity changes between the first and the third segments, and that this information can be handled later, rather then collecting the data during the transition period.
Accordingly, the present invention is a method of TA, which does not compare the velocities during the acceleration of the aircraft.
The validity of our method was analytically proven and confirmed by simulations and actual flight tests.
The present invention includes the definition of a 12 dimensional state vector, which consist of the estimations of the following carried vehicle's error components: velocities (x3), angles (x3), gyroscope drifts (x3) and accelerometer biases (x3).
This vector which is the output of the Kalman filter, whose input is the measured difference between the carrier velocity and the carried vehicle velocity, (the velocity error vector), is continuously updated during the TA flight.
The values of the terms of the state vector prior to the launch provides the pre-launch initial condition for the carried vehicle navigation during its mission; The conditions are: its initial velocities, its initial attitude with respect to the LLLN (Local Level Local North) coordinate system shown in
FIG. 1
, and its initial instrumental offsets.
The present invention includes a new schedule of the TA flight and a different routine for data processing, which are based on the possibility of separating the state vector entirely into its components, during all stages of the alignment flight.
It is therefore an object of the present invention to provide a method for determining the pre-launch inertial condition for the INS of a carrier based launched vehicle.
It is another object of the present invention to provide a new method of TA.
It is another object of the present invention to provide such a method, which is not subjected to errors, which are caused by time delays in the information supplied by the carrier vehicle.
It is another object of the present invention to provide such a method, in which the alignment maneuvers of the carrier airplane are minimized.
It is another object of the present invention to provide such a method in cost effective manner.
Other objects of the invention will become apparent upon reading the following description taken in conjunction with the accompanying drawings.
It should also be understood that this invention is not limited to aircraft, and that it is applicable to boarding platforms of other kinds including space naval or under water vehicles.
REFERENCES:
patent: 4288049 (1981-09-01), Alpers
patent: 5150856 (1992-09-01), Gaide
patent: 5245909 (1993-09-01), Corrigan et al.
patent: 5260709 (1993-11-01), Nowakowski
patent: 5470033 (1995-11-01), Tsai et al.
patent: 5672872 (1997-09-01), Wu et al.
patent: 609
Ben-Jaacov Joseph
Reiner Jacob
Friedman Mark M.
Nguyen Tan Q.
Rafael-Armament Development Authority Ltd.
Tran Dalena
LandOfFree
Method for transfer alignment of an inertial measurement... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for transfer alignment of an inertial measurement..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for transfer alignment of an inertial measurement... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3037476