Sensor calibration and time system for ground truth static scene sparse flow generation

1. A system for motion planning, the system comprising:

a memory;

a set of sensors including a light detection and ranging (LiDAR) sensor; and

at least one processor in communication with the set of sensors and the memory and configured to:

perform a LiDAR scan resulting in a pair of images, each image of the pair of images containing observed points and each image of the pair of images performed at a different time;

generate a LiDAR static-scene point cloud by extracting static points using a transformed reference scan; and

generate an image point correspondence for the pair of images using the LiDAR static-scene point cloud,

wherein the at least one processor is further configured to i) identify, for the pair of images, the static points belonging to static objects based on a comparison between a point from the LiDAR scan and a point in the transformed reference scan and ii) align the observed points based on an identification of the static points.

2. The system according to claim 1 , wherein the set of sensors further includes a camera and a GNSS-inertia system and wherein the LiDAR scan is performed in a swing form.

3. The system according to claim 1 , wherein the at least one processor is further configured to filter out image point correspondences related to invalid pixels.

4. The system according to claim 1 , wherein the at least one processor is further configured to:

detect a GNSS-inertial pose estimate near an acquisition time of the LiDAR scan; and

align the GNSS-inertial pose estimate with the observed points of the LiDAR scan.

5. The system according to claim 4 , wherein the at least one processor is further configured to:

determine a LiDAR pose based on interpolation; and

transform the LiDAR scan into an undistorted LiDAR scan.

6. The system according to claim 5 , wherein determining the LiDAR pose includes interpolating relative positions from a GNSS-inertial system.

7. The system according to claim 5 , wherein the

static points are identified using interpolating of relative positions from a GNSS-inertial system, and

the observed points are aligned based on GNSS-inertial estimates.

8. The system according to claim 7 , wherein the transformed reference scan is temporally proximate to the LiDAR scan and the transformed reference scan is aligned with the LiDAR scan.

9. The system according to claim 8 , wherein the at least one processor is further configured to:

determine that a distance between a static point from the LiDAR scan and a point from the transformed reference scan is smaller than a threshold; and

extract static point from the LiDAR scan.

10. The system according to claim 1 , wherein the at least one processor is further configured to:

correspond, for the pair of images, a first image of the pair of images to the LiDAR static-scene point cloud.

11. The system according to claim 10 , wherein the at least one processor is further configured to:

generate, for a 3D point in the LiDAR static-scene point cloud, two image points in sub-pixel accuracy by projecting the 3D point onto the pair of images; and

establish a correspondence between the two image points.

12. The system according to claim 1 , wherein the at least one processor is further configured to:

compare local image patches near a set of image point correspondences; and

filter a subset of image point correspondences having a significant difference in the local image patches.

13. The system according to claim 12 , wherein the at least one processor is further configured to:

store filtered image point correspondences as a ground truth for a static-scene dataset motion planning.

14. A non-transitory computer readable storage medium including instructions, which when executed by one or more processors of a computing device, cause the computing device to implement a method comprising:

performing a LiDAR scan resulting in a pair of images, each image of the pair of images containing observed points and each image of the pair of images performed at a different time;

generating a LiDAR static-scene point cloud by extracting static points using a transformed reference scan; and

generating an image point correspondence for the pair of images using the LiDAR static-scene point cloud,

wherein the at least one processor is further configured to i) identify, for the pair of images, the static points belonging to static objects based on a comparison between a point in one of the pair of images and a point in the transformed reference scan and ii) align the observed points based on an identification of the static points.

15. The non-transitory computer readable medium of claim 14 , further comprising:

filtering an image point correspondence related to invalid pixels.

16. The non-transitory computer readable medium of claim 14 , further comprising:

detecting a GNSS-inertial pose estimate near an acquisition time of the LiDAR scan; and

aligning the GNSS-inertial pose estimate with the LiDAR scan.

17. The non-transitory computer readable medium of claim 16 , further comprising:

determining a LiDAR pose based on interpolation; and

transforming the LiDAR scan into an undistorted LiDAR scan.

18. The non-transitory computer readable medium of claim 17 ,

wherein the static points are identified using interpolating of relative positions from a GNSS-inertial system, and

the observed points are aligned based on GNSS-inertial estimates.

19. The non-transitory computer readable medium of claim 18 wherein the:

transformed reference scan is temporally proximate to the LiDAR scan and the transformed reference scan is aligned with the LiDAR scan.

20. The non-transitory computer readable medium of claim 19 , further comprising:

determining that a distance between a static point from the LiDAR scan and a point from the transformed reference scan is smaller than a threshold; and

extracting the static point from the LiDAR scan.