ROAD SEGMENT SPATIAL EMBEDDING

Abstract

The subject disclosure relates to techniques for enabling autonomous vehicles to reason about similarities of features between drivable areas. A process of the disclosed technology can include receiving first sensor data corresponding with a first roadway segment, receiving second sensor data corresponding with a second roadway segment, encoding the first sensor data into a first vector, wherein the first vector represents a first roadway characteristic associated with the first roadway segment, encoding the second sensor data into a second vector, wherein the second vector represents a second roadway characteristic associated with the second roadway segment, and determining a similarity between the first roadway segment and the second roadway segment based on the first vector and the second vector.

Description

Claims

1. A computer-implemented method comprising: receiving first sensor data corresponding with a first roadway segment;receiving second sensor data corresponding with a second roadway segment;encoding the first sensor data into a first vector, wherein the first vector represents a first roadway characteristic associated with the first roadway segment;encoding the second sensor data into a second vector, wherein the second vector represents a second roadway characteristic associated with the second roadway segment; anddetermining a similarity between the first roadway segment and the second roadway segment based on the first vector and the second vector.

2. The computer-implemented method of claim 1, wherein the first sensor data and the second sensor data are perceived by sensors of an autonomous vehicle.

3. The computer-implemented method of claim 1, wherein the first sensor data is perceived by sensors of a first autonomous vehicle and the second sensor data is perceived by sensors of a second autonomous vehicle.

4. The computer-implemented method of claim 1, wherein the first roadway characteristic includes at least one of velocity of vehicles on the first roadway segment, density of vehicles on the first roadway segment, velocity of pedestrians on the first roadway segment, and density of vehicles on the first road segment.

5. The computer-implemented method of claim 1, wherein the first roadway characteristic is associated with timestamp information.

6. The computer-implemented method of claim 1, wherein the first vector includes timestamp information.

7. The computer-implemented method of claim 1, wherein determining the similarity between the first roadway segment and the second roadway segment includes clustering the first vector and the second vector.

8. A system comprising: a storage configured to store instructions; anda processor configured to execute the instructions and cause the processor to: receive first sensor data corresponding with a first roadway segment;receive second sensor data corresponding with a second roadway segment;encode the first sensor data into a first vector, wherein the first vector represents a first roadway characteristic associated with the first roadway segment;encode the second sensor data into a second vector, wherein the second vector represents a second roadway characteristic associated with the second roadway segment; anddetermine a similarity between the first roadway segment and the second roadway segment based on the first vector and the second vector.

9. The system of claim 8, wherein the first sensor data and the second sensor data are perceived by sensors of an autonomous vehicle.

10. The system of claim 8, wherein the first sensor data is perceived by sensors of a first autonomous vehicle and the second sensor data is perceived by sensors of a second autonomous vehicle.

11. The system of claim 8, wherein the first roadway characteristic includes at least one of velocity of vehicles on the first roadway segment, density of vehicles on the first roadway segment, velocity of pedestrians on the first roadway segment, and density of vehicles on the first road segment.

12. The system of claim 8, wherein the first roadway characteristic is associated with timestamp information.

13. The system of claim 8, wherein the first vector includes timestamp information.

14. The system of claim 8, wherein determining the similarity between the first roadway segment and the second roadway segment includes clustering the first vector and the second vector.

15. A non-transitory computer readable medium comprising instructions, the instructions, when executed by a computing system, cause the computing system to: receive first sensor data corresponding with a first roadway segment;receive second sensor data corresponding with a second roadway segment;encode the first sensor data into a first vector, wherein the first vector represents a first roadway characteristic associated with the first roadway segment;encode the second sensor data into a second vector, wherein the second vector represents a second roadway characteristic associated with the second roadway segment; anddetermine a similarity between the first roadway segment and the second roadway segment based on the first vector and the second vector.

16. The computer readable medium of claim 15, the first sensor data and the second sensor data are perceived by sensors of an autonomous vehicle.

17. The computer readable medium of claim 15, the first sensor data is perceived by sensors of a first autonomous vehicle and the second sensor data is perceived by sensors of a second autonomous vehicle.

18. The computer readable medium of claim 15, the first roadway characteristic includes at least one of velocity of vehicles on the first roadway segment, density of vehicles on the first roadway segment, velocity of pedestrians on the first roadway segment, and density of vehicles on the first road segment.

19. The computer readable medium of claim 15, the first roadway characteristic is associated with timestamp information.

20. The computer readable medium of claim 15, the first vector includes timestamp information.