Extended object tracking using RADAR
Techniques are provided for extended object tracking using RADAR return points only. The techniques include receiving the return points from at least one RADAR sensor of the vehicle. One or more clusters based on or from the return points are generated. An estimated position and velocity of each of the one or more clusters is computed. A Recursive Least Squares (RLS) based algorithm is proposed to estimate the instantaneous velocity of a cluster in real-time that allows for accurate track-cluster association, removes the need to perform computationally expensive non-linear state updates, and allows for the estimation of the true velocity even in frames with a large amount of clutter. If it is determined that the one or more clusters are associated with an existing object track, the existing object track is updated using at least the respective positions of the one or more clusters associated with the existing object track.
1. A method comprising:
receiving, using one or more processors of a vehicle, return points from at least one RADAR sensor of the vehicle;
generating, using the one or more processors, one or more clusters based on the return points;
computing, using the one or more processors, an estimated position and a velocity of each of the one or more clusters;
computing, using the one or more processors, a length, a width, and a center of each cluster, wherein return points belonging to a respective cluster are concentrated at the center of the respective cluster, and wherein the center of the respective cluster is a geometric center of a rectangular bounding box drawn through extreme return points in the respective cluster;
determining, using the one or more processors, if the one or more clusters are associated with an existing object track; and
in accordance with the one or more clusters being associated with the existing object track, updating the existing object track using at least the respective positions of the one or more clusters associated with the existing object track.
2. The method of claim 1 , wherein generating the one or more clusters further comprises:
filtering, using the one or more processors, return points to remove return points that have range rates less than a threshold range rate.
3. The method of claim 2 , further comprising:
accumulating, using the one or more processors, the filtered return points for multiple RADAR time-steps or frames.
4. The method of claim 1 , wherein generating the one or more clusters of the return points comprises clustering the return points using density-based, spatial clustering.
5. The method of claim 4 , wherein the density-based, spatial clustering uses an elliptical region around the return points to expand the one or more clusters, and the elliptical region has a major axis that is parallel to traffic flow.
6. The method of claim 1 , further comprising:
computing, using the one or more processors, a heading for each of the one or more clusters using the respective velocities of the each of the one or more clusters.
7. The method of claim 6 , wherein the length is defined as a maximum distance between the return points in a direction of the heading.
8. The method of claim 6 , wherein the width is defined as a maximum distance between the return points that are perpendicular to the heading.
9. The method of claim 1 , wherein the returns are aligned to an axis of a global reference frame prior to computing the length, width and center.
10. The method of claim 1 , wherein determining if the each one or more clusters are associated with the existing object track using the respective positions and velocities of the each of the one or more cluster further comprises:
computing differences between the respective estimated positions and velocities of the one or more clusters and a predicted position and velocity of the existing object track; and
determining if the differences are below respective threshold values for the position and velocity.
11. An apparatus comprising:
one or more processors; and
memory storing instructions that when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, using one or more processors of a vehicle, return points from at least one RADAR sensor of the vehicle;
generating, using the one or more processors, one or more clusters based on the return points;
computing, using the one or more processors, an estimated position and a velocity of each of the one or more clusters;
computing, using the one or more processors, a length, a width, and a center of each cluster, wherein return points belonging to a respective cluster are concentrated at the center of the respective cluster, and wherein the center of the respective cluster is a geometric center of a rectangular bounding box drawn through extreme return points in the respective cluster;
determining, using the one or more processors, if the one or more clusters are associated with an existing object track; and
in accordance with the one or more clusters being associated with the existing object track, updating the existing object track using at least the respective positions of the one or more clusters associated with the existing object track.
12. The apparatus of claim 11 , wherein generating the one or more clusters further comprises:
filtering, using the one or more processors, return points to remove return points that have range rates less than a threshold range rate.
13. The apparatus of claim 12 , the operations further comprising:
accumulating, using the one or more processors, the filtered return points for multiple RADAR time-steps or frames.
14. The apparatus of claim 11 , wherein generating the one or more clusters of the return points comprises clustering the return points using density-based, spatial clustering.
15. The apparatus of claim 14 , wherein the density-based, spatial clustering uses an elliptical region around the return points to expand the one or more clusters, and the elliptical region has a major axis that is parallel to traffic flow.
16. The apparatus of claim 11 , the operations further comprising:
computing, using the one or more processors, a heading for each of the one or more clusters using the respective velocities of the each of the one or more clusters.
17. The apparatus of claim 16 , wherein the length is defined as a maximum distance between the return points in a direction of the heading.
18. The apparatus of claim 16 , wherein the width is defined as a maximum distance between the return points that are perpendicular to the heading.
19. The apparatus of claim 11 , wherein the returns are aligned to an axis of a global reference frame prior to computing the length, width and center.
20. The apparatus of claim 11 , wherein determining if the each one or more clusters are associated with the existing object track using the respective positions and velocities of the each of the one or more cluster further comprises:
computing differences between the respective estimated positions and velocities of the one or more clusters and a predicted position and velocity of the existing object track; and
determining if the differences are below respective threshold values for the position and velocity.