Map updates based on data captured by an autonomous vehicle

A computing system comprising:

• at least one non-transitory computer readable medium comprising instructions stored thereon, wherein the instructions are effective to cause the computing system to:
• receive data indicating an inconsistency between stored data describing a feature of a location, and current data describing the feature of the location, wherein the stored data comprises semantic labels identifying the feature, and wherein the current data is captured by at least one sensor of an autonomous vehicle;
• automatically check for an availability of pre-existing updates to the semantic labels associated with the stored data based on the inconsistency;
• generate revised data based on the current data if no pre-existing updates to the semantic labels are available;
• and
• publish the revised data.

A non-transitory computer readable medium comprising instructions stored thereon, wherein the instructions are effective to cause an autonomous vehicle to:

• receive data indicating an inconsistency between stored data describing a feature of a location, and current data describing the feature of the location, wherein the stored data comprises semantic labels identifying the feature, and wherein the current data is captured by at least one sensor of an autonomous vehicle;
• automatically check for an availability of pre-existing updates to the semantic labels associated with the stored data based on the inconsistency;
• generate revised data based on the current data if no pre-existing updates to the semantic labels are available;
• and
• publish the revised data.