Communications: directive radio wave systems and devices (e.g. – Return signal controls radar system – Receiver
Reexamination Certificate
1998-09-25
2001-05-29
Gregory, Bernarr E. (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Return signal controls radar system
Receiver
C342S089000, C342S090000, C342S094000, C342S095000, C342S097000, C342S195000
Reexamination Certificate
active
06239739
ABSTRACT:
This application claims priority under 35 U.S.C. §§119 and/or 365 to 9703470-6 filed in Sweden on Sep. 26, 1997; the entire content of which is hereby incorporated by reference.
BACKGROUND
The present invention has its main application within the field of sensors and relates to a method for association between tracks of moving measurement objects, in particular when a pause has occurred between the target tracks between which association is to be performed.
Certain types of sensor, for example search radar systems or surveillance radar systems, do not measure the position of one and the same measurement object constantly but instead measure at discrete times the positions of a plurality of measurement objects within a given area or volume. In connection with this type of sensor, it may be desirable to obtain information about where a measurement object will be located at a point in time before the next measurement time, in other words it is desirable to be able to predict the position of the measurement object, the target, before the next measurement time.
Such prediction is often referred to as target tracking and is performed on the basis of a number of consecutive measurement times which are hereinafter referred to as detections. Target tracking provides information about, for example, the speed, course and acceleration of the target. This information is obtained with the aid of calculations which are performed on the basis of a number of consecutive detections of the target concerned. A problem may arise in the calculations if more than one target is located within the area or volume scanned by the sensor. The problem lies in the uncertainty about which new detections are to be considered to belong to which existing target tracks.
This problem can be solved by so-called association, which is often performed in two stages, correlation and allocation. Correlation usually consists of a method for determining the degree of probability that a new detection belongs to an existing target track. On the basis of the result of the correlation, the new detection can then be allocated to a given existing target track.
A conventional method for association is to predict at the correlation stage where the targets that are being tracked will be located the next time they are detected, after which a tracked target is allocated the new detection that lies closest to the predicted position of the tracked target.
There are a great many methods for obtaining improved reliability in association on the basis of this fundamental principle. One way is, with reference to the last detected position of a tracked target, to create an imaginary area or volume within which it is considered that it will be possible for the tracked target to be located at the next detection. A new detection which is carried out within this area or volume is allocated to the tracked target. Another method is, on the basis of a tracked target, to initiate a number of alternative tracks, so-called hypothetical tracks, in order at a later stage to investigate which of the hypothetical tracks corresponds most closely to a given new detection.
The problem with association between new detections and existing target tracks is rendered even more difficult if the sensor used has made a pause in its scanning of a given area or volume. It has not been possible for the target tracks that were located within this area or volume during the pause to be updated with new detections during the pause. The method described above for calculating possible positions for targets after the pause then leads to unreasonably large areas for the possible positions. In the same way, the method using hypothetical tracks leads to altogether too many allocation possibilities.
European Patent Application EP 607 015 describes a method for association between a new detection and an existing target track with the aid of the first of the methods described above, in other words information about the previous movement of the target is used to predict a volume within which the new position of the target may be.
European Patent Application EP 660 134 describes a method for tracking a number of targets simultaneously. The method includes the formation of hypothetical tracks, in principle according to the second of the two methods described above.
SUMMARY
The problem that is solved with the aid of the present invention is that of obtaining improved reliability in association in connection with target tracking with the aid of sensors, in particular association carried out after a pause in the scanning by a sensor of a given area or volume.
According to the invention, this problem is solved by, in contrast to previously known methods, not attempting to perform association between new detections and existing target tracks. In the method according to the invention, new target tracks are instead initiated after a pause, which new target tracks are each compared with the target tracks that existed before the pause for the purpose of being able to perform correct allocations.
When, in the association method according to the invention, a new target track is compared with one of the target tracks that existed before the pause, use is made of the fact that certain parameters of the target tracks compared are known, such as points in time, positions, courses and speeds.
On the basis of certain assumptions, calculations are performed, in each comparison, relating to movement between two positions, one in each of the two target tracks compared. The assumptions that are made according to the invention are that a given maximum number of changes in course and changes in speed with given maximum values can have taken place during movement between the two positions. In the calculations, one of the known parameters for the two positions is allowed to vary, while other known parameters are kept constant. The calculations that are performed relate to a maximum and a minimum value which the parameter that is allowed to vary may assume in order for movement to be possible between the two positions.
If the known value of the parameter that is allowed to vary lies between the calculated maximum and minimum values, the new target track is allocated to the target track with which it is compared.
If, in accordance with the principle of the invention described above, it is possible to perform more than one allocation, the allocation is made that is considered to indicate the smallest change with regard to the speed and/or course during the movement.
Every sensor has a certain measurement inaccuracy, due to which the parameters that according to the invention are kept constant are nevertheless allowed to vary to an extent that is defined by the measurement inaccuracy of the sensor.
In a preferred embodiment of the invention, the movement time, in other words the time difference between the two positions compared, is the known parameter that is allowed to vary. The maximum and minimum values for the parameter that is allowed to vary are then the longest and, respectively, the shortest time that the movement can have taken between the positions compared. If the movement time lies between this shortest and, respectively, longest time, the new target track is allocated to the target track with which it is compared.
As target tracks are built up on the basis of a number of detections, the method according to the intention means that it will be possible for attempts at comparison after a pause to be made only when a number of detections for one and the same target have been carried out after the pause. This delay can nevertheless be considered to be compensated by the increased reliability that is obtained in the association, and also by the fact that the delay is in most cases negligible in relation to the duration of the pause.
REFERENCES:
patent: 4992795 (1991-02-01), Lassallette et al.
patent: 5138321 (1992-08-01), Hammer
patent: 5325098 (1994-06-01), Blair et al.
patent: 5414643 (1995-05-01), Blackman et al.
patent: 5491645 (1996-02-01), Kennedy et al.
patent: 5798942 (1998-08-01), Danchick e
Bentland Benny
Kronhamn Thomas
Thomson David
Burns Doane Swecker & Mathis L.L.P.
Gregory Bernarr E.
Telefonaktiebolaget LM Ericsson (publ)
LandOfFree
Method for association 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 association, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for association will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2453768