Localization of a vehicle
A computer includes a processor and a memory, and the memory stores instructions executable by the processor to receive a first top-view image of an area obtained from a sensor external to a vehicle, generate a second top-view image of the area based on data from a camera of the vehicle, and determine an estimated location of the vehicle in the area based on comparing the first top-view image and the second top-view image.
1. A computer comprising a processor and a memory, the memory storing instructions executable by the processor to:
receive a first top-view image of an area obtained from a sensor external to a vehicle;
generate a second top-view image of the area based on data from a camera of the vehicle;
determine pixel coordinates in the first top-view image that correspond to an image location of the vehicle in the second top-view image based on comparing the first top-view image and the second top-view image,
determine an estimated location of the vehicle in the area according to the pixel coordinates in the first top-view image; and
actuate a component of the vehicle based on the estimated location;
wherein comparing the first top-view image and the second top-view image includes incrementally moving a window having a same set of pixel dimensions as the second top-view image across the first top-view image and identifying a region of the first top-view image encompassed by the window having a greatest similarity to the first top-view image, the identified region defining the pixel coordinates in the first top-view image.
2. The computer of claim 1 , wherein the instructions to generate the second top-view image include instructions to generate the second top-view image based on image data from a plurality of cameras of the vehicle including the camera.
3. The computer of claim 2 , wherein the cameras are oriented generally horizontally, and the instructions to generate the second top-view image include instructions to project the image data from the cameras to a common horizontal plane.
4. The computer of claim 1 , wherein the instructions to compare the first top-view image and the second top-view image include instructions to identify a best match region of the first top-view image to the second top-view image.
5. The computer of claim 4 , wherein the instructions to identify the best match region include instructions to compare each of a plurality of comparison regions of the first top-view image to the second top-view image.
6. The computer of claim 5 , wherein the instructions to identify the best match region include instructions to:
determine a score of each comparison region of the first top-view image by comparing the respective comparison region to the second top-view image; and
identify the best match region by selecting the comparison region with the best score.
7. The computer of claim 5 , wherein the instructions further include instructions to determine the plurality of comparison regions based on an identified location of the vehicle.
8. The computer of claim 4 , wherein the instructions further include instructions to determine an estimated heading of the vehicle by rotating the second top-view image and determining the best match region by comparing the second top-view image to the first top-view image after rotating the second top-view image.
9. The computer of claim 1 , wherein the instructions to determine the estimated location include instructions to apply an image location of the vehicle in the second top-view image to the first top-view image.
10. The computer of claim 1 , wherein the instructions further include instructions to determine a location of the vehicle based on the estimated location and on a previous location of the vehicle.
11. The computer of claim 10 , wherein the instructions to determine the location of the vehicle further include instructions to iteratively filter the estimated location and the previous location.
12. The computer of claim 1 , wherein the estimated location includes two horizontal coordinates.
13. The computer of claim 1 , wherein the instructions further include instructions to determine an estimated heading of the vehicle based on comparing the first top-view image and the second top-view image.
14. The computer of claim 1 , wherein the first top-view image lacks metadata of lanes.
15. The computer of claim 1 , wherein the first top-view image is grayscale.
16. The computer of claim 1 , wherein the second top-view image is grayscale.
17. The computer of claim 1 , wherein the first top-view image is a satellite image.
18. The computer of claim 1 , wherein the instructions to actuate the component include instructions to navigate the vehicle based on the estimated location.
19. A method comprising:
receiving a first top-view image of an area obtained from a sensor external to a vehicle;
generating a second top-view image of the area based on data from a camera of the vehicle;
determining pixel coordinates in the first top-view image that correspond to an image location of the vehicle in the second top-view image based on comparing the first top-view image and the second top-view image,
determining an estimated location of the vehicle in the area according to the pixel coordinates in the first top-view image; and
actuating a component of the vehicle based on the estimated location;
wherein comparing the first top-view image and the second top-view image includes incrementally moving a window having a same set of pixel dimensions as the second top-view image across the first top-view image and identifying a region of the first top-view image encompassed by the window having a greatest similarity to the first top-view image, the identified region defining the pixel coordinates in the first top-view image.