System and method for managing a yard

1 . A method of maintaining trailer asset data in a database of a yard, the method comprising:

comparing an update time difference for one or more database entries to a predetermined data refresh threshold, wherein the update time difference for a database entry is a difference between a current time and a time the database entry was last updated;

commanding an ego vehicle to move a first asset within the yard from a first location to a second location;

wherein when the update time difference for a first database entry exceeds the predetermined data refresh threshold, and a first associated location associated with the first database entry is outside of a shortest path between the first location and the second location, causing the ego vehicle to move the first asset from the first location to the second location via a route that passes by the first associated location; and

updating the first database entry with newer data obtained from at least one sensor located on the ego vehicle when the ego vehicle passes by the first associated location.

2 . The method of claim 1 wherein the at least one sensor includes a video camera.

3 . The method of claim 1 wherein

each time an ego vehicle passes by an asset location where an asset may be located, the at least one sensor scans a bounded region at the asset location,

a controller in communication with the at least one sensor determines, based on data provided by the at least one sensor, whether the asset is located within the bounded region,

if the asset is determined to be at least partially within the bounded region, the controller determines an identity of the asset relative to the bounded region and compares the identity to a stored identity for a database entry associated with the bounded region.

4 . The method of claim 3 , wherein if the identity matches the stored identity, the data stored in the database associated with the bounded region is not altered, and a time of a last data update for this database entry is reset to the current time.

5 . The method of claim 3 , wherein if the identity does not match the stored identity, the identity is recorded in the database entry associated with the bounded region, and a time of a last data update for the database entry associated with the bounded region is reset to the current time.

6 . The method of claim 3 , wherein a first image is provided to an asset ID module if the asset located within an unprocessed version the first image occupies at least 50% of the image.

7 . The method of claim 6 , wherein the identity of the asset is determined by

locating an asset identifier within the first image of the asset captured by the ego vehicle,

extracting the asset identifier from the first image by the asset ID module, and

providing the asset identifier to a rectification module to determine if the asset identifier matches a known asset identifier.

8 . The method of claim 7 , wherein the rectification module determines a probability that the asset identifier is erroneous.

9 . The method of claim 8 , wherein if the probability that the asset identifier is erroneous is above a predetermined threshold probability, an ego vehicle is routed to pass by the location of the bounded region to scan any asset present within the bounded region.

10 . The method of claim 8 , wherein if the probability that the asset identifier is erroneous is above a predetermined threshold probability, a second image of the asset is obtained, a second asset identifier is located within the second image of the asset, and

the asset identifier is extracted from the second image by the asset ID module.

11 . The method of claim 10 , wherein the asset identifier extracted from the second image is provided to the rectification module to determine if the asset identifier matches a known asset identifier.

12 . The method of claim 7 , wherein the rectification module uses state data associated with the asset to assist in determining the identity of the asset.

13 . The method of claim 1 , wherein a probability that data in a second database entry is correct is determined to be below a predetermined threshold probability, the ego vehicle is routed to pass by a second associated location associated with the second database entry to re-scan the identity.

14 . The method of claim 13 , wherein prior to being routed past the second associated location, the ego vehicle was idle.

15 . The method of claim 1 , wherein the database maintains state data associated with assets located within the yard, wherein state data comprises data describing physical features of the assets.

16 . The method of claim 1 , wherein the database maintains state data associated with assets located within the yard, wherein state data comprises data describing poses of the assets within the yard.

17 . The method of claim 1 , wherein the database maintains state data associated with assets located within the yard, wherein state data comprises data describing one or more brands of assets within the yard.

18 . The method of claim 1 , wherein if a subset of data in the database entry does not match the newer data obtained from the at least one sensor, a probability measure that data in the database is correct is reduced.

19 . The method of claim 1 , wherein a history of database entries is maintained for at least a predetermined number of passes of an ego vehicle past the first associated location, to allow current data to be compared with the history of database entries.

20 . A method of maintaining trailer asset data in a database of a yard, the method comprising:

determining one or more bounded regions in a yard;

associating a first bounded region with a first database entry;

determining a time difference between a current time and a first time associated with the first database entry, wherein the first time corresponds to when data in the first database entry was entered into the database;

comparing the time difference to a predetermined time difference threshold;

wherein if the time difference is larger than the predetermined time difference threshold, routing an ego vehicle to pass by the first bounded region to scan any assets present within the first bounded region, with sensors located on the ego vehicle, to generate new scanned data;

overwriting data in the first database entry with the new scanned data; and

updating the first time associated with the first database entry to a new time corresponding to when the data in the first database entry was overwritten with the new scanned data.