Mobile work machine control using a subterranean feature model

Abstract

A plurality of detection points is identified from a collection of subterranean scan data collected using a sensor on a mobile work machine. A predicted utility segment is generated based on a pattern reflected from the plurality of detection points. The predicted utility segment is incorporated into a subterranean feature model. A component of the mobile work machine is controlled, using the subterranean feature model, to perform an operation.

WHAT IS CLAIMED IS: 
     
         1 . A method of controlling a mobile work machine on a worksite to perform an operation, the method comprising: 
 identifying a plurality of detection points from a collection of subterranean scan data collected using a sensor on the mobile work machine;   generating a predicted utility segment based at least in part on a pattern reflected in the plurality of detection points;   incorporating the predicted utility segment into a subterranean feature model; and   controlling a component of the mobile work machine, using the subterranean feature model, to perform the operation.   
     
     
         2 . The method of  claim 1 , wherein generating the predicted utility segment further comprises generating the predicted utility segment based at least in part on a comparison of a depth at which detection points in the plurality of detection points is located. 
     
     
         3 . The method of  claim 2 , wherein generating the predicted utility segment further comprises generating the predicted utility segment based at least in part on a position of detection points the plurality of detection points relative to one another. 
     
     
         4 . The method of  claim 1 , wherein generating the predicted utility segment further comprises generating the predicted utility segment such that the predicted utility segment aligns with at least three detection points in the plurality of detection points determined programmatically to have a same depth and fall on a same line. 
     
     
         5 . The method of  claim 1 , further comprising generating a projected utility segment and incorporating the projected utility segment into the subterranean feature model, wherein the projected utility segment is in line with the predicted utility segment. 
     
     
         6 . The method of  claim 1 , wherein identifying the plurality of detection points further comprises identifying the plurality of detection points from a collection of subterranean scan data collected using a ground penetrating radar sensor that is part of the mobile work machine. 
     
     
         7 . The method of  claim 6 , wherein the ground penetrating radar sensor is a part of a bucket of the mobile work machine. 
     
     
         8 . The method of  claim 1 , wherein identifying the plurality of detection points from the collection of subterranean scan data collected using the sensor on the mobile work machine further comprises identifying the plurality of detection points from the collection of subterranean scan data collected using the sensor during a plurality of separate scan operations along a plurality of separate scan paths. 
     
     
         9 . The method of  claim 1 , wherein controlling the component of the mobile work machine further comprises physically moving the mobile work machine based at least in part on a location of the predicted utility segment. 
     
     
         10 . The method of  claim 1 , wherein controlling the component of the mobile work machine further comprises displaying an indication of the predicted utility segment. 
     
     
         11 . The method of  claim 1 , wherein controlling the component of the mobile work machine further comprises transferring an indication of the predicted utility segment to another system or machine that is separate from the mobile work machine. 
     
     
         12 . A computing system for a mobile work machine, the computing system comprising: 
 a detection point identification component that provides a plurality of detection points collected during a subterranean scan conducted by an operator of the mobile work machine while operating the mobile work machine;   a segment prediction component that processes the plurality of detection points and generates a predicted segment based on a comparison of a depth at which detection points in the plurality of detection points is determined to be located;   an output generation component that creates a subterranean feature model incorporating the predicted segment; and   a control system that controls the mobile work machine, using the subterranean feature model, to perform an operation.   
     
     
         13 . The computing system of  claim 12 , further comprising a segment projection component that generates a projected segment that aligns with the predicted segment, and wherein the output generation component incorporates the projected segment into the subterranean feature model. 
     
     
         14 . The computing system of  claim 12 , wherein the subterranean scan further comprises at least two separate subterranean scans in a same area of interest conducted by the operator of the mobile work machine while operating the mobile work machine. 
     
     
         15 . The computing system of  claim 12 , wherein the segment prediction component generates the predicted segment based on a determination that at least three detection points in the plurality of detection points are at a same depth and fall along a same line. 
     
     
         16 . The computing system of  claim 12 , wherein the operation provides a display that includes a representation of the predicted segment.  
     
     
         17 . A mobile work machine, comprising: 
 a ground penetrating radar sensor that moves across a ground surface while performing a subterranean scan;   a detection point identification component that identifies a plurality of detection points from the subterranean scan;   a segment prediction component that creates a predicted segment based on a pattern identified in depth and location of detection points in the plurality of detection points; and    a control system that controls the mobile work machine, using the predicted segment, to perform an operation.   
     
     
         18 . The mobile work machine of  claim 17 , further comprising a segment projection component that creates a projected segment that is an extension of the predicted segment, and wherein the control system controls the mobile work machine, using the projected segment, to perform the operation. 
     
     
         19 . The mobile work machine of  claim 18 , wherein the operation is a display of the predicted segment and the projected segment on a user interface of the mobile work machine.  
     
     
         20 . The mobile work machine of  claim 17 , wherein the ground penetrating radar sensor moves with a bucket of the mobile work machine across a ground surface while performing a subterranean scan.