Map updated system

Abstract

Receive drive information from a vehicle that traversed a road segment 2710. Provide a representation of the drive information to a trained neural network 2720 and output an identification of an updat

Claims

WHAT IS CLAIMED IS:

1. A system for updating a map for use in navigating a host vehicle relative to a road segment, the system comprising:at least one processor comprising circuitry and a memory, wherein the memory includes instructions that when executed by the circuitry cause the at least one processor to:receive drive information from a vehicle that traversed the road segment, wherein the drive information includes at least one indicator of: a detected road edge location, a detected lane mark location, and a type and corresponding location of a detected road sign;provide a representation of the drive information to a trained model, wherein the trained model is configured to receive the representation of the drive information as input and provide, based on analysis of the representation of the drive information, an output that includes identification of at least one update to make to at least one road topography feature representation stored in the map;generate an updated road topography feature representation in response to the identification of the at least one update to make to the at least one road topography feature representation stored in the map;store the updated road topography feature representation in the map to provide an updated map; anddistribute the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated road topography feature representation.

2. The system of claim 1, wherein the representation of the drive information constitutes the drive information as received from the vehicle.

3. The system of claim 1, wherein the representation of the drive information constitutes an image representation of the drive information.

4. The system of claim 3, wherein the image representation is associated with a vantage point of a camera mounted on the vehicle that traversed the road segment.

5. The system of claim 3, wherein the image representation provides a top view of road topography features associated with the road segment.

6. The system of claim 1, wherein the trained model includes a graph neural network.

7. The system of claim 1, wherein the updated road topography feature includes a host vehicle drivable path, and wherein the updated road topography feature representation includes a path change relative to a previously stored host vehicle drivable path.

8. The system of claim 8, wherein the host vehicle drivable path is represented in the map as a 3D spline.

9. The system of claim 1, wherein the updated road topography feature includes a road edge, and wherein the updated road topography feature representation includes a road edge location change relative to a previously stored road edge representation.

10. The system of claim 1, wherein the updated road topography feature includes a lane marking, and wherein the updated road topography feature representation includes a lane marking location change relative to a previously stored lane marking representation.

11. The system of claim 1, wherein the updated road topography feature includes a road sign, and wherein the updated road topography feature representation includes a road sign type or road sign location change relative to road sign information previously stored in the map.

12. The system of claim 1, wherein the updated road topography feature includes a traffic light, and wherein the updated road topography feature representation includes a traffic light type or traffic light location change relative to traffic light information previously stored in the map.

13. The system of claim 1, wherein the updated road topography feature includes a road barrier, and wherein the updated road topography feature representation includes a road barrier type or road barrier location change relative to road barrier information previously stored in the map.

14. The system of claim 1, wherein the representation of the drive information provided to the trained model as input includes both non-aggregated and aggregated drive information received from a plurality of vehicles that previously traversed the road segment.

15. The system of claim 14, wherein the non-aggregated and aggregated drive information corresponds to a common region of the map.

16. The system of claim 1, wherein the at least one map update includes a region where a road topography change was detected and one or more boundary regions where the received drive information is consistent with potential topography impacts associated with the detected road topography change.

17. The system of claim 16, wherein the region where the road topography change was detected is associated with a first confidence level and wherein the one or more boundary regions are associated with a second confidence level, the second confidence level being different from the first confidence level.

18. The system of claim 17, wherein the first confidence level is associated with a higher stringency than the second confidence level.

19. The system of claim 18, wherein the stringency corresponds to a threshold distance relative to a stored map feature.

20. A non-transitory computer-readable medium storing instructions executable by at least one processor for updating a map for use in navigating a host vehicle relative to a road segment according to a method, the method comprising:receiving drive information from a vehicle that traversed the road segment, wherein the drive information includes at least one indicator of: a detected road edge location, a detected lane mark location, and a type and corresponding location of a detected road sign;providing a representation of the drive information to a trained model, wherein the trained model is configured to receive the representation of the drive information as input and provide, based on analysis of the representation of the drive information, an output that includes identification of at least one update to make to at least one road topography feature representation stored in the map;generating an updated road topography feature representation in response to the identification of the at least one update to make to the at least one road topography feature representation stored in the map;storing the updated road topography feature representation in the map to provide an updated map; anddistributing the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated road topography feature representation.

21. The non-transitory computer-readable medium of claim 20, wherein the trained model includes a graph neural network.

22. The non-transitory computer-readable medium of claim 20, wherein the updated road topography feature includes a host vehicle drivable path, and wherein the updated road topography feature representation includes a path change relative to a previously stored host vehicle drivable path.

23. The non-transitory computer-readable medium of claim 22, wherein the host vehicle drivable path is represented in the map as a 3D spline.

24. A computer-implemented method for updating a map for use in navigating a host vehicle relative to a road segment, the method comprising:receiving drive information from a vehicle that traversed the road segment, wherein the drive information includes at least one indicator of: a detected road edge location, a detected lane mark location, and a type and corresponding location of a detected road sign;providing a representation of the drive information to a trained model, wherein the trained model is configured to receive the representation of the drive information as input and provide, based on analysis of the representation of the drive information, an output that includes identification of at least one update to make to at least one road topography feature representation stored in the map;generating an updated road topography feature representation in response to the identification of the at least one update to make to the at least one road topography feature representation stored in the map;storing the updated road topography feature representation in the map to provide an updated map; anddistributing the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated road topography feature representation.

25. The method of claim 24, wherein the representation of the drive information provided to the trained model as input includes both non-aggregated and aggregated drive information received from a plurality of vehicles that previously traversed the road segment.

26. The method of claim 25, wherein the non-aggregated and aggregated drive information corresponds to a common region of the map.

27. The method of claim 24, wherein the at least one map update includes a region where a road topography change was detected and one or more boundary regions where the received drive information is consistent with potential topography impacts associated with the detected road topography change.

28. A system for updating a map for use in navigating a host vehicle relative to a road segment, the system comprising:at least one processor comprising circuitry and a memory, wherein the memory includes instructions that when executed by the circuitry cause the at least one processor to:receive drive information from a vehicle that traversed the road segment, wherein the drive information includes a representation of a road topography feature;determine, based on distribution information associated with the road topography feature, whether to update a stored representation of the road topography feature in the map, the distribution information being generated based on a plurality of representationsof the road topography feature collected by one or more vehicles that previously traversed the road segment;generate an updated representation of the road topography feature in response to a determination to update the stored representation of the road topography feature in the map;store the updated representation of the road topography feature in the map to provide an updated map; anddistribute the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated representation of the road topography feature.

29. The system of claim 28, wherein a position of the stored representation of the road topography feature in the map is determined based on at least some of the plurality of representations of the road topography feature.

30. The system of claim 28, wherein the determination whether to update the stored representation of the road topography feature in the map is based on a comparison of a location of the representation of a road topography feature included in the drive information to the distribution information.

31. The system of claim 30, wherein the comparison includes providing a representation of the drive information to a trained model.

32. The system of claim 31, wherein the representation of the drive information provided to the trained model as input includes both non-aggregated and aggregated drive information.

33. The system of claim 28, wherein the distribution information is based on distances from the plurality of representations of the road topography feature to the stored representation of the road topography feature.

34. The system of claim 28, wherein the distribution information is based on a boundary defined by the plurality of representations of the road topography feature.

35. The system of claim 33, wherein the boundary includes a buffer distance relative to the plurality of representations of the road topography feature.

36. The system of claim 28, wherein the at least one map update includes a region where a road topography change was detected and one or more boundary regions where the received drive information is consistent with potential topography impacts associated with the detected road topography change.

37. The system of claim 28, wherein the at least one map update includes a region where a road topography change was detected and one or more boundary regions where the receiveddrive information is consistent with potential topography impacts associated with the detected road topography change, providing a representation of the drive information to a trained model 38. A non-transitory computer-readable medium storing instructions executable by at least one processor for updating a map for use in navigating a host vehicle relative to a road segment according to a method, the method comprising:receiving drive information from a vehicle that traversed the road segment, wherein the drive information includes a representation of a road topography feature;determining, based on distribution information associated with the road topography feature, whether to update a stored representation of the road topography feature in the map, the distribution information being generated based on a plurality of representations of the road topography feature collected by one or more vehicles that previously traversed the road segment;generating an updated representation of the road topography feature in response to a determination to update the stored representation of the road topography feature in the map;storing the updated representation of the road topography feature in the map to provide an updated map; anddistributing the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated representation of the road topography feature.

39. The non-transitory computer-readable medium of claim 38, wherein a position of the stored representation of the road topography feature in the map is determined based on the plurality of representations of the road topography feature.

40. The non-transitory computer-readable medium of claim 38, wherein the determination whether to update the stored representation of the road topography feature in the map is based on a comparison of a location of the representation of a road topography feature included in the drive information to the distribution information.

41. The non-transitory computer-readable medium of claim 38, wherein the distribution information is based on distances from the plurality of representations of the road topography feature to the stored representation of the road topography feature.

42. A computer-implemented method for updating a map for use in navigating a host vehicle relative to a road segment, the method comprising:receiving drive information from a vehicle that traversed the road segment, wherein the drive information includes a representation of a road topography feature;determining, based on distribution information associated with the road topography feature, whether to update a stored representation of the road topography feature in the map, the distribution information being generated based on a plurality of representations of the road topography feature collected by one or more vehicles that previously traversed the road segment;generating an updated representation of the road topography feature in response to a determination to update the stored representation of the road topography feature in the map;storing the updated representation of the road topography feature in the map to provide an updated map; anddistributing the updated map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the updated representation of the road topography feature.

43. The method of claim 42, wherein the comparison includes providing a representation of the drive information to a trained model.

44. The method of claim 43, wherein the representation of the drive information provided to the trained model as input includes both non-aggregated and aggregated drive information.

45. The method of claim 42, wherein the at least one map update includes a region where a road topography change was detected and one or more boundary regions where the received drive information is consistent with potential topography impacts associated with the detected road topography change.

Images