LIDAR-based shipboard tracking and state estimation for autonomous landing
A system and method for state estimation of a surface of a platform at sea, includes receiving sensor signals indicative of LIDAR data for the platform; applying a Bayesian filter to the LIDAR data for a plurality of hypotheses; determining vertical planes representing azimuth and elevation angles for the LIDAR data; applying a robust linear estimation algorithm to the vertical planes; and determining candidate points in response to the applying of the robust linear estimation algorithm.
1. A method for state estimation of a surface of a platform at sea, comprising:
receiving, with a processor, sensor signals indicative of LIDAR data for the platform;
applying, with the processor, a Bayesian filter to the LIDAR data for a plurality of hypotheses;
determining, with the processor, vertical planes representing azimuth and elevation angles for the LIDAR data;
applying, with the processor, a robust linear estimation algorithm to the vertical planes;
determining, with the processor, candidate points in response to the applying of the robust linear estimation algorithm; and
landing an aircraft on the surface of the platform that is estimated by the candidate points.
2. The method of claim 1 , further comprising initializing the Bayesian filter with an initial state estimate for the platform.
3. The method of claim 1 , further comprising clustering the candidate points into three-dimensional segments.
4. The method of claim 3 , further comprising determining dense planar regions in the candidate points.
5. The method of claim 2 , further comprising determining a refined state estimate of the platform from a robust planar estimation.
6. The method of claim 5 , further comprising determining the state estimation of the platform by applying the Bayesian filter using the refined state estimate.
7. The method of claim 5 , further comprising updating the initial state estimate with the refined state estimate.
8. The method of claim 1 , wherein the applying of the Bayesian filter further comprises applying the Bayesian filter to a plurality of sea-states.
9. A system for state estimation of a surface of a platform at sea, comprising:
a sensor system of an aircraft;
a processor; and
memory having instructions stored thereon that, when executed by the processor, cause the system to:
receive sensor signals indicative of LIDAR data for the platform;
apply a Bayesian filter to the LIDAR data for a plurality of hypotheses;
determine vertical planes representing azimuth and elevation angles for the LIDAR data;
apply a robust linear estimation algorithm to the vertical planes; and
determine candidate points in response to the applying of the robust linear estimation algorithm.
10. The system of claim 9 , wherein the processor is configured to initialize the Bayesian filter with an initial state estimate for the platform.
11. The system of claim 9 , wherein the processor is configured to cluster the candidate points into three-dimensional segments.
12. The system of claim 11 , wherein the processor is configured to determine dense planar regions in the candidate points.
13. The system of claim 10 , wherein the processor is configured to determine a refined state estimate of the platform from a robust planar estimation.
14. The system of claim 13 , wherein the processor is configured to determine the state estimation of the platform by applying the Bayesian filter using the refined state estimate.
15. The system of claim 13 , wherein the processor is configured to update the initial state estimate with the refined state estimate.