Communications: directive radio wave systems and devices (e.g. – Radar ew
Reexamination Certificate
2003-04-21
2004-10-05
Sotomayor, John B. (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Radar ew
C342S096000, C342S097000, C342S108000, C342S113000, C342S133000, C342S139000, C342S146000, C342S147000, C342S417000, C342S450000, C342S451000, C342S465000
Reexamination Certificate
active
06801152
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to radars, and more particularly, to bearings-only passive emitter tracking. Even more particularly, the present invention is related to a method of and apparatus for determining the optimum observer heading change in bearings-only passive emitter tracking.
A moving observer, e.g. an aircraft, able to measure bearings to an emitter, generates the range, speed and heading of the emitter based on the measured bearings. This bearings-only passive emitter tracking is called target motion analysis (TMA). Passive TMA is useful because emitters can be detected and tracked at much longer ranges than possible using active radar.
However, passive emitter tracking or ranging is intrinsically less accurate than active radar tracking and requires many bearing measurements, and hence, additional time to converge to a solution. Also bearings-only TMA requires the observer to maneuver at some point during the bearing measurement period. The fact that the observer does not know the emitter's location or velocity until after the observer has maneuvered introduces a large element of risk when aircraft use TMA for target tracking.
To combine bearing measurements with the requisite maneuver, the observer typically flies a doglegged course. Constant velocity tracks with frequent heading changes allow the estimator, i.e. the observer, to uniquely determine range and velocity based on the bearing measurements if the emitter is flying a constant velocity track. When a unique solution exists, the target location, speed and heading is said to be “observable.” The general concept of estimator observability is presented in Anderson and Moore,
Optimal Filtering
, Prentice-Hall, New Jersey 1979. Fogel and Gavish, in “N
th
-Order Dynamics Target Observability from Angle Measurements”,
IEEE Transactions on Aerospace and Electronic Systems
, AES-24, 3 (May 1988), describe the observability problem specifically for bearings-only passive emitter tracking. In particular, they demonstrate that heading change is sufficient to provide convergence to a unique solution when tracking a constant velocity target.
Prior methods of bearings-only target tracking, such as the method described by U.S. Pat. No. 5,877,998 (the '998 patent) to Aidala, et at. in “Recursive Method for Target Motion Analysis” emphasize such observer motion. The '998 patent refers to observer tracks as data collection legs. In the '998 patent, bearing measurements from the first and second leg, are filtered to produce a smoothed bearing estimate, a bearing-velocity estimate, and a bearing-acceleration estimate. These estimates are used to generate target range and velocity. Further, the '998 patent finds it desirable to incorporate at least a third, and possibly more, measurement legs to reduce estimation error.
The method disclosed in the '998 patent and other current techniques for doing bearings-only TMA, fail to exploit the results presented by B. J. McCabe in “Accuracy and Tactical Implications of Bearings-Only Ranging Algorithms”,
Operations Research
, Vol. 33, No. 1, 1985. McCabe showed that there is a preferred method in performing data collection over the first two legs. McCabe defines the tracking observer, or tracker, as leading the emitter if the velocity vectors
100
,
101
(
FIG. 1
a
) are on the same side of the line-of-sight (LOS) vector
105
when the first data collection leg begins. The tracker lags if tracker velocity vector
102
(
FIG. 1
b
) and the target velocity vector
103
are on opposite sides of the LOS vector
106
.
Hence, a two leg maneuver, as required by Aidala, may be either a lead followed by a lag, or vice-versa. A lead-lag observer maneuver occurs when both emitter and observer velocity vectors are initially on the same side of the line-of-sight, then after performing a maneuver the vectors are on opposite sides. A lag-lead observer maneuver occurs when the observer velocity vector is initially on the opposite side of the line-of-sight to the emitter's velocity vector, then the velocity vectors are on the same side after performing a maneuver. The two leg maneuver could also be a lead-lead or a lag-lag maneuver. McCabe describes that among all possible two leg maneuvers the lead-lag (
FIG. 1
a
) is much preferred in conventional TMA. For instance, the estimated range error at the end of the lead-lag maneuver can, theoretically, be only 20% of the lag-lead (
FIG. 1
b
) error. Thus, if the lead-lag maneuver is performed, a third leg
104
(
FIG. 1
b
) to reduce estimation error would not be required in a significant number of cases. Furthermore, by a straightforward extension of McCabe's work it can be shown that both lead-lag and lag-lead maneuvers are generally superior to the other dogleg maneuver combinations.
As McCabe noted, conventional TMA implementations, such as that described by Aidala, are unable to take advantage of the above facts. At the start of TMA, the tracker does not know whether it is leading or lagging the emitter because the emitter's velocity vector is not known. Because of the observability constraint discussed by Fogel and Gavish, the velocity vector is not obtained until the second leg.
This has potentially dire tactical consequences when the observer is a high-speed aircraft , such as an air-intercept (AI) jet, engaging a high speed threat. Not only can McCabe's results not be exploited, but if the Al aircraft is initially leading the target, inadvertently performing a suboptimal lead-lead maneuver may put the aircraft in a vulnerable position. But executing the optimal lead-lag maneuver, i.e. turning away on the second leg, may cause the aircraft to fail to intercept the target. Thus, it is vitally important for tactical aircraft performing TMA to know where they will be relative to the emitter at the end of the data collection legs, both for self-protection and the potential for enhanced performance from data collection maneuvers. However, the tracker must obtain the emitter heading prior to executing the first turn in order to determine the target's relative position. In current TMA implementations, the observer cannot determine the target's relative position prior to executing the first turn based on the above-described observability constraint.
One way to avoid both the “require the track to best estimate the track” paradox and the TMA observer-maneuver vulnerability problem is using the technique described in the present inventor's patent disclosure entitled, “A Method for Passively Estimating an Emitter's Position and Velocity Using Bearings-Only Without Requiring Observer Acceleration,” Ser. No. 10/419193, hereinafter referred to as “inventor's co-pending application” filed on even date herewith and incorporated by reference by its entirety into the instant specification. The method described in the aforementioned patent application avoids the observability constraint. That is, obtaining emitter range, speed and heading does not require the observer to fly a dogleg course, or otherwise accelerate. The described method obtains emitter range by estimating speed in two ways: (1) using platform and mission identification to estimate a discrete target speed; and (2) obtaining speed from bearing rates-of-change, but as a function of unknown range. Equating the functionality speed with the discrete speed determines emitter range.
The above-described approach performs well for many critical emitters. However, a large data base covering the performance of all platforms encountered is required in order to accurately estimate emitter speed. Usually, associating signal pulse parameters with target kinematics requires an artificial intelligence or expert system implementation. And even with an extensive data base and sophisticated logic, the track generated for a subset of emitters can be ambiguous because several discrete speeds are equally likely. This speed-ambiguity arises from a one-to-several radar-to-platform mapping, and also a one-to-several r
Lowe Hauptman & Gilman & Berner LLP
Northrop Grumman Corporation
Sotomayor John B.
LandOfFree
Method for determining the optimum observer heading change... 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 determining the optimum observer heading change..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining the optimum observer heading change... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3275122