Updating maps based on traffic object detection

1. A computer-implemented method comprising:

receiving traffic light data from a plurality of autonomous vehicles, the traffic light data including a geographic location of a traffic light;

comparing the traffic light data of each of the plurality of autonomous vehicles with known traffic light data;

determining discrepancy values between the traffic light data of each of the plurality of autonomous vehicles and the known traffic light data, wherein the discrepancy values indicate an extent of change of the traffic light data from the known traffic light data;

grouping the traffic light data of each of the plurality of autonomous vehicles into a first group of traffic light data and a second group of traffic light data based on the determining of the discrepancy values, wherein the first group of traffic light data includes data from autonomous vehicles indicating a first degree of change in traffic light data from the known traffic light data, and wherein and the second group of traffic light data includes data from autonomous vehicles indicating a second degree of change from the known traffic light data;

comparing the first group of traffic light data to the second group of traffic light data to determine that the first group of traffic light data is more accurate;

determining whether the first group of traffic light data exceeds a threshold; and

updating a traffic light map in response to determining that the first group of traffic light data exceeds the threshold.

2. The computer-implemented method of claim 1 , further comprising providing a traffic light map update to the plurality of autonomous vehicles based on the updated traffic light map.

3. The computer-implemented method of claim 1 , wherein the traffic light data further includes at least one of traffic light lane association, angle of the traffic light, position of the traffic light, type of the traffic light, duration of the traffic light, and color of the traffic light.

4. The computer-implemented method of claim 3 , wherein the type of the traffic light includes arrow-type traffic lights and arrowless-type traffic lights.

5. The computer-implemented method of claim 1 , further comprising preparing traffic light data tables that include the traffic light data for each of the plurality of autonomous vehicles.

6. The computer-implemented method of claim 5 , further comprising comparing the traffic light data tables of each of the plurality of autonomous vehicles with a known traffic light data table.

7. The computer-implemented method of claim 6 , further comprising grouping the traffic light data tables based on a discrepancy level, the discrepancy level being the difference between the traffic light data tables of each of the plurality of autonomous vehicles and the known traffic light data table.

8. The computer-implemented method of claim 1 , wherein the discrepancy between the traffic light data of each of the plurality of autonomous vehicles and the known traffic light data is a value that corresponds to the difference between the received traffic light data of the plurality of autonomous vehicles and the known traffic light data.

9. The computer-implemented method of claim 1 , wherein the first group of the traffic light data is based on a percentage of difference between the traffic light data from the plurality of autonomous vehicles and the known traffic light data.

10. The computer-implemented method of claim 1 , further comprising replacing the known traffic light data with the traffic light data of the group that exceeds the threshold.

11. A system comprising:

one or more processors; and

at least one computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the system to:

receive traffic light data from a plurality of autonomous vehicles, the traffic light data including a geographic location of a traffic light;

compare the traffic light data of each of the plurality of autonomous vehicles with known traffic light data;

determine discrepancy values between the traffic light data of each of the plurality of autonomous vehicles and the known traffic light data, wherein the discrepancy values indicate an extent of change of the traffic light data from the known traffic light data;

group the traffic light data of each of the plurality of autonomous vehicles into a first group of traffic light data and a second group of traffic light data based on the determination of the discrepancy values, wherein the first group of traffic light data includes data from autonomous vehicles indicating a first degree of change in traffic light data from the known traffic light data, and wherein the second group of traffic light data includes data from autonomous vehicles indicating a second degree of change from the known traffic light data;

compare the first group of traffic light data to the second group of traffic light data to determine that the first group of traffic light data is more accurate;

determine whether the first group of traffic light data exceeds a threshold; and

update a traffic light map in response to determining that the first group of traffic light data exceeds the threshold.

12. The system of claim 11 , wherein the instructions which, when executed by the one or more processors, further cause the system to provide a traffic light map update to the plurality of autonomous vehicles based on the updated traffic light map.

13. The system of claim 11 , wherein the traffic light data further includes at least one of traffic light lane association, angle of the traffic light, position of the traffic light, type of the traffic light, duration of the traffic light, and color of the traffic light.

14. The system of claim 13 , wherein the type of the traffic light includes arrow-type traffic lights and arrowless-type traffic lights.

15. The system of claim 11 , wherein the instructions which, when executed by the one or more processors, further cause the system to prepare traffic light data tables that include the traffic light data for each of the plurality of autonomous vehicles.

16. The system of claim 15 , wherein the instructions which, when executed by the one or more processors, further cause the system to compare the traffic light data tables of each of the plurality of autonomous vehicles with a known traffic light data table.

17. The system of claim 16 , wherein the instructions which, when executed by the one or more processors, further cause the system to group the traffic light data tables based on a discrepancy level, the discrepancy level being the difference between the traffic light data tables of each of the plurality of autonomous vehicles and the known traffic light data table.

18. The system of claim 11 , wherein the discrepancy between the traffic light data of each of the plurality of autonomous vehicles and the known traffic light data is a value that corresponds to the difference between the received traffic light data of the plurality of autonomous vehicles and the known traffic light data.

19. The system of claim 11 , wherein the groups of the traffic light data is based on a percentage of difference between the traffic light data from the plurality of autonomous vehicles and the known traffic light data.

20. The system of claim 11 , wherein the instructions which, when executed by the one or more processors, further cause the system to replace the known traffic light data with the traffic light data of the group that exceeds the threshold.

21. A non-transitory computer-readable storage medium comprising:

instructions stored on the non-transitory computer-readable storage medium, the instructions, when executed by one more processors, cause the one or more processors to:

receive traffic light data from a plurality of autonomous vehicles, the traffic light data including a geographic location of a traffic light;

compare the traffic light data of each of the plurality of autonomous vehicles with known traffic light data;

determine discrepancy values between the traffic light data of each of the plurality of autonomous vehicles and the known traffic light data, wherein the discrepancy values indicate an extent of change of the traffic light data from the known traffic light data;

group the traffic light data of each of the plurality of autonomous vehicles into a first group of traffic light data and a second group of traffic light data based on the determination of the discrepancy values, wherein the first group of traffic light data includes data from autonomous vehicles indicating a first degree of change in traffic light data from the known traffic light data, and wherein the second group of traffic light data includes data from autonomous vehicles indicating a second degree of change from the known traffic light data;

compare the first group of traffic light data to the second group of traffic light data to determine that the first group of traffic light data is more accurate;

determine whether the first group of traffic light data exceeds a threshold; and

update a traffic light map in response to determining that the first group of traffic light data exceeds the threshold.

22. The non-transitory computer-readable storage medium of claim 21 , wherein the traffic light data further includes at least one of traffic light lane association, angle of the traffic light, position of the traffic light, type of the traffic light, duration of the traffic light, and color of the traffic light.