Two step pruning in a PHD filter

Patent metadata

Field	Value
Publication number	US-10605607-B2
Application number	US-201414448819-A
Country	US
Kind code	B2
Filing date	Jul 31, 2014
Priority date	Jul 31, 2014
Publication date	Mar 31, 2020
Grant date	Mar 31, 2020

Abstract

In one embodiment, a method of tracking multiple objects with a probabilistic hypothesis density filter is provided. The method includes providing a plurality of intensities. The plurality of intensities are pruned based on their respective weight to remove lower weighted intensities and produce a first set of intensities. Intensities in the first set of intensities that are identified as corresponding to the same object are then merged to produce a second set of intensities. The second set of intensities is then pruned again to produce a final set of intensities, wherein pruning the second set of intensities includes in the final set of intensities, up to a threshold number of intensities having the largest weights in the second set of intensities and excludes from the final set of intensities any remaining intensities in the second set of intensities.

First claim

What is claimed is:

1. A tracking method comprising:
tracking multiple objects in an environment around a vehicle with a probabilistic hypothesis density filter, with an aid of one or more sensors onboard the vehicle and one or more processing devices operatively coupled to the one or more sensors, which are configured to perform a method comprising:
detecting a plurality of reflected signals in the one or more sensors from the multiple objects in the environment around the vehicle;
computing a plurality of measurements based on the detected reflected signals in the one or more sensors;
generating a plurality of intensities based on the computed measurements, each intensity being associated to a track of one of the multiple objects and including a weight, a state mean vector, and a state covariance matrix of statistics of the track, at a first time;
first pruning the plurality of intensities based on their respective weight to remove lower weighted intensities than a certain weighted intensity and produce a first set of intensities;
merging intensities in the first set of intensities that are identified as corresponding to a same object to produce a second set of intensities; and
second pruning the second set of intensities to produce a final set of intensities, wherein second pruning includes in the final set of intensities, up to a threshold number of intensities having the largest weights in the second set of intensities and excludes from the final set of intensities, any remaining intensities in the second set of intensities;

wherein in the final set of intensities, each intensity corresponds to an object being tracked.

2. The method of claim 1 , wherein first pruning includes in the first set of intensities, any intensity in the plurality of intensities having a weight above a respective threshold weight for that intensity and excludes from the first set of intensities, any intensity having a weight below the respective threshold weight for that intensity.

3. The method of claim 2 , further comprising:
selecting each respective threshold weight based on which one or more of the sensors have measurements that correspond to the intensity corresponding to the respective threshold weight.

4. The method of claim 1 , wherein the plurality of intensities include one or more of a predicted intensity, a measurement-to-track intensity, and a new intensity.

5. The method of claim 1 , wherein first pruning deletes any intensity in the plurality of intensities not included in the first set of intensities.

6. The method of claim 1 , wherein second pruning deletes any intensity in the second set of intensities not included in the final set of intensities.

7. A tracking system comprising:
one or more processing devices onboard a vehicle;
one or more sensors onboard the vehicle and operatively coupled to the one or more processing devices; and
one or more data storage devices including instructions which, when executed by the one or more processing devices, cause the one or more processing devices to:
track multiple objects in an environment around the vehicle with a probabilistic hypothesis density filter with an aid of the one or more sensors onboard the vehicle;
detect a plurality of reflected signals received in the one or more sensors from the multiple objects in the environment around the vehicle;
compute a plurality of measurements based on the detected reflected signals in the one or more sensors;
generate a plurality of intensities based on the computed measurements, each intensity being associated to a track of one of the multiple objects and including a weight, a state mean vector, and a state covariance matrix of statistics of the track, at a first time;
first prune the plurality of intensities based on their respective weight to remove lower weighted intensities than a certain weighted intensity and produce a first set of intensities;
merge intensities in the first set of intensities that are identified as corresponding to a same object to produce a second set of intensities; and
second prune the second set of intensities to produce a final set of intensities, wherein second prune includes in the final set of intensities, up to a threshold number of intensities having the largest weights in the second set of intensities and excludes from the final set of intensities, any remaining intensities in the second set of intensities;

wherein in the final set of intensities, each intensity corresponds to an object being tracked.

8. The tracking system of claim 7 , wherein first prune includes in the first set of intensities, any intensity in the plurality of intensities having a weight above a respective threshold weight for that intensity and excludes from the first set of intensities, any intensity having a weight below the respective threshold weight for that intensity.

9. The tracking system of claim 8 , wherein the instructions cause the one or more processing devices to:
select each respective threshold weight based on which one or more of the sensors have measurements that correspond to the intensity corresponding to the respective threshold weight.

10. The tracking system of claim 7 , wherein the plurality of intensities include one or more of a predicted intensity, a measurement-to-track intensity, and a new intensity.

11. The tracking system of claim 7 , wherein first prune deletes any intensity in the plurality of intensities not included in the first set of intensities.

12. The tracking system of claim 7 , wherein second prune deletes any intensity in the second set of intensities not included in the final set of intensities.

13. The tracking system of claim 7 , wherein:
the vehicle comprises an aircraft; and
the plurality of sensors comprise one or more of a radar, a traffic collision avoidance system (TCAS) sensor, an automatic dependent surveillance-broadcast (ADS-B) sensor, an optical camera, or a LiDAR.

14. A non-transitory computer readable medium including instructions which, when executed by one or more processing devices, cause the one or more processing devices to:
track multiple objects in an environment around a vehicle with a probabilistic hypothesis density filter with an aid of one or more sensors onboard the vehicle;
detect a plurality of reflected signals received in the one or more sensors from the multiple objects in the environment around the vehicle;
compute a plurality of measurements based on the detected reflected signals in the one or more sensors;
generate a plurality of intensities based on the computed measurements, each intensity being associated to a track of one of the multiple objects and including a weight, a state mean vector, and a state covariance matrix of statistics of the track, at a first time;
first prune the plurality of intensities based on their respective weight to remove lower weighted intensities than a certain weighted intensity and produce a first set of intensities;
merge intensities in the first set of intensities that are identified as corresponding to a same object to produce a second set of intensities; and
second prune the second set of intensities to produce a final set of intensities, wherein second prune includes in the final set of intensities, up to a threshold number of intensities having the largest weights in the second set of intensities and excludes from the final set of intensities, any remaining intensities in the second set of intensities;
wherein in the final set of intensities, each intensity corresponds to an object being tracked.

15. The non-transitory computer readable medium of claim 14 , wherein first

Assignees

• Honeywell Int Inc

Inventors

• Bageshwar Vibhor L
• Elgersma Michael Ray
• Euteneuer Eric A

Classifications

• G01S13/865

  Combination of radar systems with lidar systems · CPC title

• G01S13/86

  Combinations of radar systems with non-radar systems, e.g. sonar, direction finder · CPC title

• G01S13/933

  of aircraft or spacecraft · CPC title

• G01C21/20 Primary

  Instruments for performing navigational calculations (G01C21/24, G01C21/26 take precedence) · CPC title

• G01S13/867

  Combination of radar systems with cameras · CPC title