Systems and methods for particle filter tracking

1. A method for operating a mobile platform, comprising:

obtaining, by a computing device, a LiDAR point cloud;

using, by the computing device, the LiDAR point cloud to generate a track for a given object by generating states of a given object over time using a particle filter algorithm, each said state having a score indicating a likelihood that a cuboid would be created given an acceleration value and an angular velocity value; and

using, by the computing device, the track to train a machine learning algorithm to detect and classify objects based on sensor data.

2. The method according to claim 1 , further comprising causing the machine learning algorithm to be used for controlling movement of the mobile platform.

3. The method according to claim 1 , wherein each said state is defined by a position, a velocity and a heading for the given object at a particular time.

4. The method according to claim 1 , wherein the acceleration value and the angular velocity value comprise random numbers.

5. The method according to claim 1 , further comprising generating the score by:

setting a score value for the cuboid equal to zero;

generating a first adjusted score by adding to the score value a likelihood of seeing the acceleration value and the angular velocity value in a context; and

generating a second adjusted score value be adding to the first adjusted score value a negative squared distance from each data point of said LiDAR point cloud to a closest edge of the cuboid.

6. The method according to claim 1 , wherein the track is generated by further generating an initial cuboid encompassing at least some data points in the LiDAR point cloud.

7. The method according to claim 6 , wherein the track is generated further by randomly selecting different sets of acceleration and angular velocity values.

8. The method according to claim 7 , wherein the track is further generated by creating a set of cuboids using the initial cuboid and the different sets of acceleration and angular velocity values.

9. The method according to claim 8 , wherein the track is generated by further determining a score for each said cuboid of the set of cuboids that indicates a likelihood that the cuboid would be created given the respective one of the different sets of acceleration and angular velocity values.

10. The method according to claim 9 , wherein the track is generated further by: identifying scores of said scores that are less than a maximum score minus a threshold value; and removing said cuboids from the set of cuboids that are associated with the scores which were identified.

11. A system, comprising:

a processor;

a non-transitory computer-readable storage medium comprising programming instructions that are configured to cause the processor to implement a method for operating a mobile platform, wherein the programming instructions comprise instructions to:

obtain a LiDAR point cloud;

use the LiDAR point cloud to generate a track for a given object by generating states of a given object over time using a particle filter algorithm, each said state having a score indicating a likelihood that a cuboid would be created given an acceleration value and an angular velocity value; and

use the track to train a machine learning algorithm to detect and classify objects based on sensor data.

12. The system according to claim 11 , wherein the programming instructions further comprise instructions to cause the machine learning algorithm to be used for controlling movement of the mobile platform.

13. The system according to claim 11 , wherein each said state is defined by a position, a velocity and a heading for the given object at a particular time.

14. The system according to claim 11 , wherein the programming instructions further comprise instructions to generate the score by:

setting a score value for the cuboid equal to zero;

generating a first adjusted score by adding to the score value a likelihood of seeing the acceleration value and the angular velocity value in a context; and

generating a second adjusted score value be adding to the first adjusted score value a negative squared distance from each data point of said LiDAR point cloud to a closest edge of the cuboid.

15. The system according to claim 11 , wherein the track is generated further by generating an initial cuboid encompassing at least some data points in the LiDAR point cloud.

16. The system according to claim 15 , wherein the track is generated further by randomly selecting different sets of acceleration and angular velocity values.

17. The system according to claim 16 , wherein the track is generated further by creating a set of cuboids using the initial cuboid and the different sets of acceleration and angular velocity values.

18. The system according to claim 17 , wherein the track is generated by further determining a score for each said cuboid of the set of cuboids that indicates a likelihood that the cuboid would be created given the respective one of the different sets of acceleration and angular velocity values.

19. The system according to claim 18 , wherein the track is generated further by: identifying scores of said scores that are less than a maximum score minus a threshold value; and removing said cuboids from the set of cuboids that are associated with the scores which were identified.

20. A computer program product comprising a non-transitory memory and programming instructions that are configured to cause a processor to:

obtain a LiDAR point cloud;

use the LiDAR point cloud to generate a track for a given object by generating states of a given object over time using a particle filter algorithm, each said state having a score indicating a likelihood that a cuboid would be created given an acceleration value and an angular velocity value; and

use the track to train a machine learning algorithm to detect and classify objects based on sensor data.