Merging intensities in a PHD filter based on a sensor track ID

• US 10,309,784 B2
• Filed: 07/31/2014
• Issued: 06/04/2019
• Est. Priority Date: 07/31/2014
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Bageshwar, Vibhor L., Elgersma, Michael Ray, Euteneuer, Eric A.
• Primary Examiner(s): Ngon, Ricky
• Application Number:
  US14/448,808
• Publication Number:
  US 20160033276A1
• Time in Patent Office: 1,769 Days
• Field of Search: 702150
• CPC Class Codes: G01B 21/16 for measuring distance of clearance between spaced objects
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/86 Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
• CPC Class Codes: G01S 13/865 Combination of radar systems with lidar systems
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G01S 13/933 of aircraft or spacecraft

Abstract

In one embodiment, a method of tracking multiple objects with a probabilistic hypothesis density filter is provided. The method includes generating a first intensity by combining a first one or more measurements, wherein a first set of track IDs associated with the first intensity includes track IDs corresponding to respective measurements in the first one or more measurements. A second intensity is generated by combining a second one or more measurements, wherein a second set of track IDs associated with the second intensity includes track IDs corresponding to respective measurements in the second one or more measurements. The first set of track IDs is compared to the second set of track IDs, and the first intensity is selectively merged with the second intensity based on whether any track IDs in the first set of track IDs match any track IDs in the second set of track IDs.

Claims

1. A method of tracking multiple objects with a probabilistic hypothesis density filter, the method comprising:

receiving reflected signals in a first one or more sensors and a second one or more sensors from an object in an environment around a vehicle, wherein the first one or more sensors and the second one or more sensors are onboard the vehicle;

detecting the received reflected signals in the first one or more sensors and the second one or more sensors;

computing a first one or more measurements based on the detected reflected signals in the first one or more sensors;

computing a second one or more measurements based on the detected reflected signals in the second one or more sensors;

generating a first intensity by combining the first one or more measurements, wherein a first set of track identifiers (IDs) includes one or more track IDs provided by the first one or more sensors corresponding to a respective measurement in the first one or more measurements, wherein the first intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at a first time;

generating a second intensity by combining the second one or more measurements, wherein a second set of track IDs includes one or more track IDs provided by the second one or more sensors corresponding to a respective measurement in the second one or more measurements, wherein the second intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at the first time;

comparing the first set of track IDs to the second set of track IDs; and

selectively merging the first intensity with the second intensity based on whether any track IDs in the first set of track IDs match any track IDs in the second set of track IDs.

2. The method of claim 1, further comprising:

computing a statistical distance between the first intensity and the second intensity; and

wherein selectively merging includes selectively merging the first intensity with the second intensity based on whether the statistical distance is less than a threshold.

3. The method of claim 2, wherein selectively merging includes merging the first intensity with the second intensity if any track IDs in the first set of track IDs match any track IDs in the second set of track IDs and if the statistical distance is less than a threshold.

4. The method of claim 1, further comprising:

if any track IDs in the first set of track IDs match any track IDs in the second set of track IDs, computing a statistical distance between the first intensity and the second intensity; and

if the statistical distance is less than a threshold, merging the first intensity with the second intensity.

5. The method of claim 4, wherein computing a statistical distance includes computing a Malhanobis distance.

6. The method of claim 1, further comprising:

generating a plurality of intensities each having one or more track IDs associated therewith, wherein the first and second intensity are in the plurality of intensities; and

wherein selectively merging includes merging the plurality of intensities to produce one or more merged intensities, wherein a given intensity is merged with one or more other intensities if any of one or more track IDs associated with the given intensity match any of one or more track IDs associated with the one or more other intensities and if a statistical distance between the given intensity and the one or more other intensities having a matching track ID is less than a threshold.

7. The method of claim 1, wherein each track ID of the first set of track IDs and the second set of track IDs is one of an identifier assigned by a sensor to a set of measurements obtained by that sensor that are correlated across time, or an international civil aviation organization (ICAO) aircraft address received by a sensor and corresponding to a respective measurement for that sensor.

8. The method of claim 1, wherein the first intensity includes one of a predicted intensity, a measurement-to-track intensity, and a new intensity and the second intensity includes one of a predicted intensity, a measurement-to-track intensity, and a new intensity.

9. A method of tracking multiple objects with a probabilistic hypothesis density filter, the method comprising:

receiving reflected signals in a first one or more sensors and a second one or more sensors from an object in an environment around a vehicle, wherein the first one or more sensors and the second one or more sensors are onboard the vehicle;

detecting the received reflected signals in the first one or more sensors and the second one or more sensors;

computing a first one or more measurements based on the detected reflected signals in the first one or more sensors;

computing a second one or more measurements based on the detected reflected signals in the second one or more sensors;

generating a first intensity by combining the first one or more measurements, wherein a first set of track identifiers (IDs) includes one or more track IDs provided by the first one or more sensors corresponding to a respective measurement in the first one or more measurements, wherein the first intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at a first time;

generating a second intensity by combining the second one or more measurements, wherein a second set of track IDs includes one or more track IDs provided by the second one or more sensors corresponding to a respective measurement in the second one or more measurements, wherein the second intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at a second time;

selectively comparing the first set of track IDs to the second set of track IDs based on whether there are any track IDs in the first set of track IDs and the second set of track IDs that are the same type; and

selectively merging the first intensity with the second intensity based on whether any track IDs in the first set of track IDs match any track IDs in the second set of track IDs;

wherein selectively comparing includes;

if the first set of track IDs includes a first track identifier (ID) that is an international civil aviation organization (ICAO) aircraft address and the second set of track IDs includes a second track ID that is an ICAO aircraft address, then;

comparing the first track ID to the second track ID;

if the first track ID matches the second track ID, then;

computing a statistical distance between the first intensity and the second intensity; and

merging the first intensity and the second intensity if the statistical distance is less than a first threshold;

if the first set of track IDs includes a first track ID from a first sensor that correlates measurements across time and the second set of track IDs includes a second track ID from the first sensor, then;

comparing the first track ID to the second track ID;

if the first track ID matches the second track ID, then;

 

computing a statistical distance between the first intensity and the second intensity; and

 

merging the first intensity and the second intensity if the statistical distance is less than a second threshold;

if either of the first and second set of track IDs do not include a track ID that is an ICAO aircraft address or if either of the first and second set of track IDs do not contain a track ID from the same sensor which correlates measurements across time, then;

computing a statistical distance between the first intensity and the second intensity; and

merging the first intensity and the second intensity if the statistical distance is less than a third threshold;

otherwise, not merging the first intensity and the second intensity.

10. The method of claim 9, wherein the first threshold is set based on which sensor provided the first track ID and which sensor provided the second track ID,

wherein the second threshold is set based on the first sensor,

wherein the third threshold is set based the one or more sensors that provided the first set of track IDs and the second set of track IDs.

11. The method of claim 10, wherein the first threshold, second threshold, and third threshold are set higher for a type of sensor having a poorer accuracy.

12. The method of claim 9, wherein each track ID of the first set of track IDs and the second set of track IDs is one of an identifier assigned by a sensor to a set of measurements obtained by that sensor that are correlated across time, or an ICAO aircraft address received by a sensor and corresponding to a respective measurement for that sensor.

13. The method of claim 9, wherein the first intensity includes one of a predicted intensity, a measurement-to-track intensity, and a new intensity and the second intensity includes one of a predicted intensity, a measurement-to-track intensity, and a new intensity.

14. The method of claim 9, wherein selectively merging includes clustering two or more intensities identified for merging together prior to combining the two or more intensities to create a merged intensity.

15. A tracking system for a vehicle, comprising:

one or more processing devices onboard the vehicle;

a plurality of sensors onboard the vehicle and operatively coupled to the one or more processing devices; and

one or more data storage devices onboard the vehicle and 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, wherein the instructions cause the one or more processing devices to perform a method comprising;

detecting reflected signals received by a first one or more of the plurality of sensors, and a second one or more of the plurality of sensors, from an object in the environment around the vehicle;

computing a first one or more measurements based on the detected reflected signals in the first one or more of the plurality of sensors;

computing a second one or more measurements based on the detected reflected signals in the second one or more of the plurality of sensors;

generating a first intensity by combining the first one or more measurements, wherein a first set of track identifiers (IDs) includes one or more track IDs provided by the first one or more of the plurality of sensors corresponding to a respective measurement in the first one or more measurements, wherein the first intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at a first time;

generating a second intensity by combining the second one or more measurements, wherein a second set of track IDs includes one or more track IDs provided by the second one or more of the plurality of sensors corresponding to a respective measurement in the second one or more measurements, wherein the second intensity includes a weight, a state mean vector, and a state covariance matrix of statistics of a track of the object at the first time;

comparing the first set of track IDs to the second set of track IDs; and

selectively merging the first intensity with the second intensity based on whether any track IDs in the first set of track IDs match any track IDs in the second set of track IDs.

16. The tracking system of claim 15, wherein:

the vehicle comprises an aircraft; and

the plurality of sensors comprise a radar, a traffic collision avoidance system (TCAS) sensor, an automatic dependent surveillance-broadcast (ADS-B) sensor, an optical camera, or a LiDAR.