System and method for inhibiting or causing automated actions based on person locations estimated from multiple video sources
AssigneeHelmerich & Payne Technologies, LLC
InventorsPeter A. Torrione, Kenneth D. Morton, Jr.
Life of this patent
Estimating the term…
Abstract
The invention relates to systems and method for inhibiting or causing automated actions based on estimated person locations comprising multiple video sources configured to detect the location of one or more persons wherein at least one video source is calibrated for a known location and pose. The invention further comprises at least one processor operably connected to a calibrated video source wherein said processor aggregates possible person locations. These systems and method may be useful for initiating or interrupting the automated activity of equipment in the presence of personnel.
Claims17
A system for inhibiting or causing automated actions based on estimated person locations comprising:
- a plurality of video sources configured to detect a location of one or more persons on a drilling site having a drilling rig and one or more pieces of machinery associated with the drilling rig, each of the one or more pieces of machinery having a respective known location, wherein at least one video source is calibrated using pre-existing fiducials in visual data from the at least one video source, wherein the pre-existing fiducials comprise the one or more pieces of machinery, wherein the calibration is based on the known locations of the one or more pieces of machinery, and wherein at least one video source is positioned to capture a perspective view, and
- at least one processor operably connected to the calibrated video sources wherein said processor aggregates possible person locations, generates bounding box data for at least one person in a frame, and processes the bounding box data to determine person locations in 3-dimensions;
- and
- a machinery control system configured to initiate, alter or interrupt automated activities in response to determined person locations in 3-dimensions, wherein the machinery control system is operably connected to equipment.
A method for inhibiting, altering or causing automated actions based on estimated person locations on a drilling site having a drilling rig, the method comprising:
- acquiring visual data from at least one video source, wherein at least one video source is positioned to capture a perspective view and calibrated using pre-existing fiducials in the visual data, wherein the pre-existing fiducials comprise one or more pieces of machinery associated with the drilling rig, and wherein the calibration is based on known locations of the one or more pieces of machinery,
- analyzing the visual data,
- generating bounding box data for at least one person in a frame;
- processing the bounding box data to determine person locations in 3-dimensions;
- and
- inhibiting or causing automated actions based on the person locations in 3-dimensions using a machinery control system, wherein the machinery control system is operably connected to equipment.
A system for monitoring personnel location on a drilling site having a drilling rig and controlling automated equipment in response to personnel location, the system comprising:
- at least one video stream provided by at least one camera operably connected to a processor, wherein at least one camera is positioned to capture a perspective view and calibrated using pre-existing fiducials in the video stream, wherein the pre-existing fiducials comprise one or more pieces of machinery associated with the drilling rig, wherein the calibration is based on known locations of the one or more pieces of machinery, and wherein the processor is configured to monitor a location of any personnel present in the video stream in 3-dimensions and generate bounding box data for any personnel present in the video stream in 3-dimensions;
- a machinery control system operably connected to the processor, wherein the machinery control system is configured to receive commands to either initiate or stop automated operations of equipment based on the bounding box data for personnel present in the video stream in 3-dimensions;
- and
- a graphical user interface configured to provide the video stream to an operator.
Dependencies are inferred from the claim text.