SYSTEMS AND METHODS OF MAINTAINING MAP FOR AUTONOMOUS DRIVING

1 . A method of maintaining a map, comprising:

receiving a sensor dataset acquired by a sensor subsystem, wherein:

the sensor dataset includes information about a road,

the sensor subsystem comprises multiple different types of sensors including at least one of a camera, a light detection and ranging (LiDAR) sensor, a positioning sensor, a radar sensor, or a mapping sensor, and

the sensor dataset has a first spatial accuracy level;

determining, by at least one processor, a confidence level by comparing the sensor dataset and the map that includes prior information about the road, wherein the map has a second spatial accuracy level;

in response to determining that the confidence level exceeds a confidence threshold, processing the map by the at least one processor; and

storing the processed map as an electronic file, wherein the processed map is configured to guide an autonomous vehicle to operate on the road.

2 . The method of claim 1 , further comprising:

in response to determining that the confidence level exceeds the confidence threshold,

causing a notification to be transmitted to an operator; and

receiving an input from the operator indicating at least one of: maintaining the map, or updating the map based on the sensor dataset; and

processing the map comprises processing the map according to the input.

3 . The method of claim 1 , wherein processing the map comprises updating the map based on the sensor dataset.

4 . The method of claim 1 , wherein:

the road comprises a plurality of road units;

the sensor dataset comprises a set of data frames, each of the set of data frames corresponding to a section of the road represented in the data frame; and

each of the plurality of road units corresponds to multiple data frames of the sensor dataset.

5 . The method of claim 4 , wherein comparing the sensor dataset and the map comprises: for each of the plurality of road units, determining a unit confidence level.

6 . The method of claim 5 , wherein:

determining that the confidence level exceeds a threshold comprises determining that at least one unit confidence level of the plurality of road units exceeds the confidence threshold, and

processing the map comprises: for each of the plurality of road units that has a corresponding unit confidence level exceeding the confidence threshold, updating, based on multiple data frames of the sensor dataset of the road unit, a portion of the map that corresponds to the road unit.

7 . The method of claim 5 , wherein for a first road unit from the plurality of road units, determining the unit confidence level comprises:

obtaining multiple frame confidence levels for multiple data frames corresponding to the first road unit by comparing the multiple data frames with corresponding portions of the map, respectively; and

determining the unit confidence level of the first road unit based on the multiple frame confidence levels.

8 . The method of claim 5 , further comprising:

identifying at least two neighboring road units along the road that satisfy a merger condition, and

obtaining a road segment by merging the at least two neighboring road units.

9 . The method of claim 8 , further comprising:

determining a segment confidence level of the road segment based on unit confidence levels of the at least two neighboring road units.

10 . The method of claim 5 , further comprising:

obtaining a plurality of road segments, wherein each of the plurality of road segments is obtained by merging at least two neighboring road units that satisfy a merger condition.

11 . The method of claim 10 , further comprising:

causing a road representation of the road to be output to a display, wherein the road representation comprises a plurality of segment representations each of which corresponds to a road segment of the plurality of road segments and relates to a segment confidence level of the road segment.

12 . The method of claim 10 , wherein a difference between segment confidence levels of any two neighboring road segments along the road fails to satisfy the merger condition.

13 . The method of claim 1 , further comprising:

obtaining trajectories of a plurality of candidate users;

identifying, based on the trajectories, a target user from the plurality of candidate users; and

transmitting the processed map or a notification regarding the processed map to the target user before the target user reaches the road.

14 . The method of claim 13 , wherein the plurality of candidate users comprise the autonomous vehicle.

15 . The method of claim 13 , wherein:

the processed map comprises an updated map that is generated based on the sensor dataset, and

the notification comprises a prompt inviting an acceptance of the processed map.

16 . A system for maintaining a map, comprising:

at least one processor configured to execute instructions that cause the at least one processor to perform operations comprising:

receiving a sensor dataset acquired by a sensor subsystem, wherein:

the sensor dataset includes information about a road,

the sensor subsystem comprises multiple different types of sensors including at least one of a camera, a light detection and ranging (LiDAR) sensor, a positioning sensor, a radar sensor, or a mapping sensor, and

the sensor dataset has a first spatial accuracy level;

determining, by the at least one processor, a confidence level by comparing the sensor dataset and the map that includes prior information about the road, wherein the map has a second spatial accuracy level;

in response to determining that the confidence level exceeds a confidence threshold, processing the map by the at least one processor; and

storing the processed map as an electronic file, wherein the processed map is configured to guide an autonomous vehicle to operate on the road.

17 . The system of claim 16 , further comprising a transmitter configured to transmit a processed map to an autonomous vehicle.

18 . The system of claim 17 , wherein the at least one processor is located outside the autonomous vehicle.

19 . The system of claim 16 , wherein the at least one processor is configured to receive a sensor dataset of a road represented in the map, the sensor dataset being acquired by a sensor subsystem that is located at a different location than at least one of the at least one processor.

20 . A non-transitory computer-readable media having instructions stored thereon, the instructions, when executed on one or more processors, cause the one or more processors to implement a method comprising:

receiving a sensor dataset acquired by a sensor subsystem, wherein:

the sensor dataset includes information about a road,

the sensor subsystem comprises multiple different types of sensors including at least one of a camera, a light detection and ranging (LiDAR) sensor, a positioning sensor, a radar sensor, or a mapping sensor, and

the sensor dataset has a first spatial accuracy level;

determining, by at least one processor, a confidence level by comparing the sensor dataset and a map that includes prior information about the road, wherein the map has a second spatial accuracy level;

in response to determining that the confidence level exceeds a confidence threshold, processing the map by the at least one processor; and

storing the processed map as an electronic file, wherein the processed map is configured to guide an autonomous vehicle to operate on the road.