Autonomous welding robots

Abstract

In various examples, a computer-implemented method of generating instructions for a welding robot. The computer-implemented method comprises identifying an expected position of a candidate seam on a part to be welded based on a Computer Aided Design (CAD) model of the part, scanning a workspace containing the part to produce a representation of the part, identifying the candidate seam on the part based on the representation of the part and the expected position of the candidate seam, determining an actual position of the candidate seam, and generating welding instructions for the welding robot based at least in part on the actual position of the candidate seam.

1 . (canceled) 
     
     
         2 . A welding robotic system, comprising:
 a robot arm positioned in a workspace, the robot arm coupled to a welding tool configured to weld two objects together along a gap between the two objects positioned in the workspace; and   a robot controller configured to:
 determine, based on a computer aided design (CAD) model depicting the two objects, an expected position of the gap within the workspace; 
 receive one or more images from one or more sensors, each sensor configured to generate sensor data associated with the workspace; 
 determine a position of the gap in the workspace based on: 
 the expected position, 
 the one or more images, or 
 a combination thereof; 
   identify, based on the one or more images, at least one former weld along the gap; and   from the at least one former weld, generate a welding path for the robotic arm to follow to weld at least a portion of the gap.   
     
     
         3 . The welding robotic system of  claim 2 , wherein the expected position is determined in accordance with one or more annotations associated with the gap in the CAD model. 
     
     
         4 . The welding robotic system of  claim 3 , wherein the one or more annotations are provided by a user. 
     
     
         5 . The welding robotic system of  claim 2 , wherein the robot controller is configured to identify the at least one former weld in the one or more images using pixel-wise classification. 
     
     
         6  The welding robotic system of  claim 2 , wherein:
 the robot controller is configured to transform the one or more images into a point cloud; and 
 the robot controller is configured to identify a representation of the at least one former weld in the point cloud using point-wise classification. 
 
     
     
         7 . The welding robotic system of  claim 2 , wherein the one or more sensors are coupled to the robot arm. 
     
     
         8 . The welding robotic system of  claim 2 , wherein the at least one former weld includes a tack weld. 
     
     
         9 . A welding robotic system, comprising:
 a robot arm positioned in a workspace, the robot arm coupled to a welding tool configured to weld two objects together along a gap between the two objects positioned in the workspace; and   a robot controller configured to:
 determine a position of the gap in the workspace based on:
 a computer aided design (CAD) model depicting the two objects; 
 one or more images associated with the workspace; or 
 a combination thereof; 
 
 identify, based on the one or more images, a former weld positioned along the gap; and 
 generate, based on a position of the gap in the workspace and based on the former weld, a welding path for the robotic arm to follow to weld at least a portion of the gap. 
   
     
     
         10 . The welding robotic system of  claim 9 , wherein, to identify the former weld, the robot controller is further configured to perform a finding operation. 
     
     
         11 . The welding robotic system of  claim 10 , wherein, to perform the finding operation, the robot controller is further configured to:
 perform a pixel-wise classification operation on the one or more images; or   perform a point-wise classification operation on a point cloud, the point cloud generated based on the one or more images.   
     
     
         12 . The welding robotic system of  claim 9 , wherein:
 the weld path is determined with respect to the former weld;   the former weld includes a tack weld; or   a combination thereof.   
     
     
         13 . The welding robotic system of  claim 9 , wherein the position of the gap in the workspace is determined based on the CAD model and the one or more images. 
     
     
         14 . The welding robotic system of  claim 9 , wherein:
 the robot controller is further configured to determine, based on the CAD model, an expected position of the gap within the workspace; and   the position of the gap in the workspace is determined based on the expected position.   
     
     
         15 . The welding robotic system of  claim 14 , wherein:
 the robot controller is further configured to identify an annotation included in the CAD model, the annotation associated with the gap; and   the expected position of the gap within the workspace is determined based on the .   
     
     
         16 . The welding robotic system of  claim 9 , wherein:
 the robot controller is further configured to receive the one or more images from one or more sensors; and   each sensor of the one or more sensors is configured to generate sensor data associated with the workspace.   
     
     
         17 . The welding robotic system of  claim 9 , wherein:
 the robot controller is further configured to:
 generate welding instructions based on the welding path; and 
 transmit the welding instructions to the robot arm coupled to the welding tool; and 
   the robot arm is configured to operate the welding tool to weld the two objects together along at least a portion of the gap based on the welding instructions.   
     
     
         18 . The welding robotic system of  claim 9 , wherein the portion of the gap is adjacent to the former weld. 
     
     
         19 . A computer-implemented method of generating welding instructions for a welding robot, the computer-implemented method comprising ::
 determining a position of a gap between multiple objects positioned in a workspace, the
 position of the gap determined based on: 
 a computer aided design (CAD) model depicting the multiple objects; 
 one or more images associated with the workspace; or 
 a combination thereof; 
   identifying, based on the one or more images, a former weld positioned along the gap; and   generating, based on a position of the gap in the workspace and based on the former weld, welding instructions for the welding robot to weld at least a portion of the gap.   
     
     
         20 . The computer-implemented method of  claim 19 , wherein:
 the welding robot includes a robot arm positioned in the workspace;   the robot arm is coupled to a welding tool configured to weld two objects together along the gap; and   the former weld includes a tack weld.   
     
     
         21 . The computer-implemented method of  claim 19 , further comprising:
 identifying an expected position of the gap based on the CAD model; and   receiving, from a sensor, the one or more images associated with the workspace; and   wherein the position is determined based on the expected position and the one or more images; and   wherein the welding instructions cause the welding robot to weld the portion of the gap in a direction away from the former weld.