Crowd sourcing data for autonomous vehicle navigation
Systems and methods of processing crowdsourced navigation information for use in autonomous vehicle navigation are disclosed. A method may include processing, by a mapping server, crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the navigation information describes road lanes of a road segment; collecting data about landmarks identified proximate to the road segment, the landmarking including a traffic sign; generating, by the mapping server, an autonomous vehicle map for the road segment, wherein the autonomous vehicle map includes a spline corresponding to a lane in the road segment and the landmarks identified proximate to the road segment; and distributing, by the mapping server, the autonomous vehicle map to an autonomous vehicle for use in autonomous navigation over the road segment.
1. A method of processing crowdsourced navigation information for use in autonomous vehicle navigation, the method comprising:
processing, by a mapping server, crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the crowdsourced navigation information describes road lanes of a road segment;
collecting data about a first landmark identified proximate to the road segment, the first landmark comprising a traffic sign;
generating, by the mapping server, an autonomous vehicle map for the road segment, wherein:
the autonomous vehicle map includes a spline corresponding to a lane in the road segment;
the autonomous vehicle map includes position information of the first landmark identified proximate to the road segment, the position information including at least one of a physical size of the first landmark, a distance between the first landmark and the road segment, a height of the first landmark, or a distance between the first landmark and a second landmark;
the position information of the first landmark is based on the crowdsourced navigation information from the plurality of vehicles; and
the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the spline; and
distributing, by the mapping server, the autonomous vehicle map to the autonomous vehicle for use in autonomous navigation over the road segment.
2. The method of claim 1 , crowdsourced navigation information includes road geometry data.
3. The method of claim 1 , wherein the plurality of vehicles includes autonomous vehicles.
4. The method of claim 1 , wherein sensor data obtained by the sensors includes road geometry data.
5. At least one non-transitory machine-readable medium including instructions for processing crowdsourced navigation information for use in autonomous vehicle navigation, wherein the instructions, when executed by a mapping server, cause the mapping server to perform operations comprising:
processing crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the crowdsourced navigation information describes road lanes of a road segment;
collecting data about a first landmark identified proximate to the road segment, the first landmark comprising a traffic sign;
generating an autonomous vehicle map for the road segment, wherein:
the autonomous vehicle map includes a spline corresponding to a lane in the road segment;
the autonomous vehicle map includes position information of the first landmark identified proximate to the road segment, the position information including at least one of a physical size of the first landmark, a distance between the first landmark and the road segment, a height of the first landmark, or a distance between the first landmark and a second landmark;
the position information of the first landmark is based on the crowdsourced navigation information from the plurality of vehicles; and
the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the spline; and
distributing the autonomous vehicle map to the autonomous vehicle for use in autonomous navigation over the road segment.
6. The at least one non-transitory machine-readable medium of claim 5 , wherein the crowdsourced navigation information includes road geometry data.
7. The at least one non-transitory machine-readable medium of claim 5 , wherein the plurality of vehicles includes autonomous vehicles.
8. The at least one non-transitory machine-readable medium of claim 5 , wherein sensor data obtained by the sensors includes road geometry data.
9. A mapping server for processing crowdsourced navigation information for use in autonomous vehicle navigation, the mapping server comprising:
at least one processor; and
a memory device including instructions, which when executed by the at least one processor, cause the at least one processor to:
process crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the crowdsourced navigation information describes road lanes of a road segment;
collect data about a first landmark identified proximate to the road segment, the first landmark comprising a traffic sign;
generate an autonomous vehicle map for the road segment, wherein:
the autonomous vehicle map includes a spline corresponding to a lane in the road segment;
the autonomous vehicle map includes position information of the first landmark identified proximate to the road segment, the position information including at least one of a physical size of the first landmark, a distance between the first landmark and the road segment, a height of the first landmark, or a distance between the first landmark and a second landmark;
the position information of the first landmark is based on the crowdsourced navigation information from the plurality of vehicles; and
the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the spline; and
distribute the autonomous vehicle map to the autonomous vehicle for use in autonomous navigation over the road segment.
10. The mapping server of claim 9 , wherein the crowdsourced navigation information includes road geometry data.
11. The mapping server of claim 9 , wherein the plurality of vehicles includes autonomous vehicles.
12. The mapping server of claim 9 , wherein sensor data obtained by the sensors includes road geometry data.
13. A vehicle system for assisting autonomous vehicle navigation using crowdsourced navigation information, the vehicle system comprising:
at least one processor; and
a memory device including instructions, which when executed by the at least one processor, cause the at least one processor to:
access navigation information at the vehicle system, the navigation information derived from sensor data obtained by a plurality of sensors coupled to the vehicle system;
provide the navigation information to a mapping server, the navigation information processed by the mapping server including crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the crowdsourced navigation information describes road lanes of a road segment; and
receive an autonomous vehicle map, wherein:
the autonomous vehicle map includes a spline corresponding to a lane in the road segment;
the autonomous vehicle map includes position information of a first landmark identified proximate to the road segment, the position information including at least one of a physical size of the first landmark, a distance between the first landmark and the road segment, a height of the first landmark, or a distance between the first landmark and a second landmark;
the position information of the first landmark is based on the crowdsourced navigation information from the plurality of vehicles; and
the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the spline.
14. The vehicle system of claim 13 , wherein the crowdsourced navigation information includes road geometry data.
15. The vehicle system of claim 13 , wherein the plurality of vehicles includes autonomous vehicles.
16. The vehicle system of claim 13 , wherein sensor data obtained by the sensors includes road geometry data.
17. At least one non-transitory machine-readable device including instructions for assisting autonomous vehicle navigation using crowdsourced navigation information, wherein the instructions, when executed by a vehicle system, cause the vehicle system to perform operations comprising:
accessing navigation information at the vehicle system, the navigation information derived from sensor data obtained by a plurality of sensors coupled to the vehicle system;
providing the navigation information to a mapping server, the navigation information processed by the mapping server including crowdsourced navigation information from a plurality of vehicles obtained by sensors coupled to the plurality of vehicles, wherein the crowdsourced navigation information describes road lanes of a road segment; and
receiving an autonomous vehicle map, wherein:
the autonomous vehicle map includes a spline corresponding to a lane in the road segment;
the autonomous vehicle map includes position information of a first landmark identified proximate to the road segment, the position information including at least one of a physical size of the first landmark, a distance between the first landmark and the road segment, a height of the first landmark, or a distance between the first landmark and a second landmark;
the position information of the first landmark is based on the crowdsourced navigation information from the plurality of vehicles; and
the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the spline.
18. The at least one non-transitory machine-readable device of claim 17 , wherein the crowdsourced navigation information includes road geometry data.
19. The at least one non-transitory machine-readable device of claim 17 , wherein the plurality of vehicles includes autonomous vehicles.
20. The at least non-transitory one machine-readable device of claim 17 , wherein the sensor data obtained by the sensors includes road geometry data.
21. The method of claim 1 , wherein the position information of the first landmark identified proximate to the road segment is associated with a signature of the first landmark in the autonomous vehicle map.
22. The method of claim 1 , wherein generating the autonomous vehicle map includes combining different parts of the crowdsourced navigation information, the different parts being associated with different ones of the plurality of vehicles.
23. The method of claim 22 , wherein combining the different parts of the crowdsourced navigation information includes averaging the different parts of the crowdsourced navigation information.
24. The method of claim 1 , wherein generating the autonomous vehicle map includes matching, using the position information of the first landmark, a first curve based on first navigation information associated with a first vehicle and a second curve based on second navigation information associated with a second vehicle.
25. The method of claim 1 , wherein the autonomous vehicle map further includes the distance between the first landmark and a second landmark, and the autonomous vehicle is configured to use the distance for vehicle localization.
26. The method of claim 1 , wherein the autonomous vehicle map is configured for an autonomous vehicle to use the position information of the first landmark for vehicle localization along the spline.
27. The method of claim 26 , wherein the spline corresponds to a target trajectory for the autonomous vehicle.
28. The method of claim 1 , wherein the spline corresponds to a target trajectory in the lane and the autonomous vehicle map is configured for the autonomous vehicle to use the position information of the first landmark for vehicle localization relative to the target trajectory.
29. The method of claim 1 , wherein generating the autonomous vehicle map comprises matching, using two or more landmarks, curves determined by different vehicles during different drives.