SYSTEMS AND METHODS FOR HANDLING TRACK MATERIALS

1 . A method of handling track materials comprising:

identifying, using a vision sensor, a track material having a character;

determining, using a trained neural network, whether the character of the track material, for a rail assembly operation, satisfies an operation threshold; and

in response to determining whether the character satisfies the operation threshold:

placing, using a robotic arm, the track material at a target location.

2 . The method of claim 1 , wherein the method further comprises lifting, before identifying the character of the track material and using the robotic arm, the track material.

3 . The method of claim 2 , wherein the track material is in an assembled position, a distributed position, or a reinstallation position.

4 . The method of claim 3 , wherein the assembled position is a position in which the track material is installed for use within a rail system.

5 . The method of claim 3 , wherein the distributed position is a position in which the track material is placed before installation along a track bed.

6 . The method of claim 3 , wherein the reinstallation position is a position in which the track material is relocated from the assembled position along a track bed.

7 . The method of claim 2 , wherein the track material is lifted from a plurality of track materials from one or more of a bin, a pile, or a conveyor.

8 . The method of claim 1 , wherein the method further comprises lifting, after identifying the character of the track material and using the robotic arm, the track material.

9 . The method of claim 1 , wherein the target location is at an adjacent node, the adjacent node comprising one or more of:

a rail,

a tie,

a container associated with the robotic arm or a vehicle,

a conveyor, or

a rejection place.

10 . The method of claim 9 , wherein, in response to determining that the character satisfies the operation threshold, the target location comprises one or more of the rail, the tie, the container associated with the robotic arm or the vehicle, or the conveyor.

11 . The method of claim 9 , wherein, in response to determining that the character fails to satisfy the operation threshold, the target location comprises one or more of the rejection place comprising one or more of:

a discard container;

a discard chute;

a discard pallet; or

a discard conveyor.

12 . The method of claim 9 , wherein the target location is at one or more of a job site, a rail side area, or a factory.

13 . The method of claim 1 , wherein the track material comprises one or more of a tie plate, a spike, an anchor, a clip, a screw, a crosstie, a pad, a bolt, a nut, a joint, a switch, or a rail.

14 . The method of claim 1 , wherein the rail assembly operation comprises one or more of an installation, a replacement, a distribution, a removal, a bundling, a sliding, or a loading, related to a track.

15 . The method of claim 1 , wherein the character of the track material comprises one or more of:

a size,

a shape,

a weight,

a hole pattern,

a model,

an assembly arrangement, or

a quality.

16 . The method of claim 15 , further comprising determining the quality of the track material by evaluating one or more track material conditions against one or more defect conditions, the one or more defect conditions comprising:

a deformation,

a corrosion,

a deviation from an original manufacturing condition,

a decay, or

a structural damage.

17 . The method of claim 1 , wherein the trained neural network is trained based on a training set of:

one or more sample rail assembly operations,

sample characters of one or more sample track materials, and

operation parameters associated with the one or more sample rail assembly operations and the sample characters.

18 . The method of claim 1 , further comprising training the trained neural network based on at least one of the character or whether the character satisfies the operation threshold.

19 . A system for handling track materials comprising:

a robotic arm operable to manipulate a track material at a location;

a vision sensor associated with the robotic arm; and

a controller comprising a trained neural network;

wherein:

the vision sensor is configured to identify a character of the track material;

the trained neural network is configured to determine whether the character of the track material, for a rail assembly operation, satisfies an operation threshold; and

the robotic arm is configured to place the track material at a target location.

20 . A processing system comprising one or more processors and one or more memories coupled with the one or more processors and configured to cause the processing system to:

identify, using a vision sensor associated with a robotic arm, a character of track material, wherein the robotic arm is configured to manipulate a track material at a target location;

determine, using a trained neural network, that a character of the track material satisfies an operation threshold of a rail assembly operation; and

in response to determining that the character of the track material satisfies the operation threshold of the rail assembly operation:

place, using the robotic arm, the track material at the target location.