Neocortex Real-Time 3D Inspection Uses Scalable Off-the-Shelf Sensors
Universal Robotics has combined automated multi-dimensional inspection with Neocortex interactive intelligence to provide 2½D, 3D, and multi-dimension inspection of random objects in large areas for a wide range of real-time logistics applications. Neocortex software accurately identifies and tracks objects that are difficult to see or change shape, such as bags. Since the software is capable of integrating with any sensor appropriate for the job, sensors can cost only one-half of traditional approaches.
The Application leverages Spatial Vision Inspection to excel at recognition and examination of complex objects 50mm in size up to an entire 52 foot long truck trailer. It does it fast – with complete spatial image capture 10X faster than most 3D approaches, while delivering up to ± 0.25mm accuracy. It does not require fixturing, additional equipment, special safety precautions, or lighting beyond typical ambient conditions.
Fast – Up to 50 objects per minute
Adaptable – Objects in any orientation or tilt
Flexible – Rigid/semi-rigid, opaque/clear objects
Scalable – Off-the-shelf 3D sensors for just right accuracy
Tested – With major robot manufacturers; 3 sigma reliability
ROI – Less than 24 months payback
Whether it is inspecting a machined part at close range, or looking into a 52-foot long truck trailer, the Application integrates a wide range of sensor types – such as lasers, structured light sensors, time of flight sensors, 2D and 3D cameras – from any vendor. The 3D accuracy and size of field of view are key parameters of sensor choice.
It can stitch together up to eight 3D maps (point clouds) from a variety of sensors, and process and analyze the integrated result faster than the physical movement of a machine or robot.
Universal Robotics experts collaborated with Nvidia engineers to optimized parallel processing to ensure massive data sets with sophisticated algorithms deliver real-time results.
• Compare – actual 3D to CAD/specs
• Align – lids, caps, tops, fasteners, subassemblies
• Assess – surface splits, gashes, holes, missing or loose items
• Complex Objects – molded plastics, curved objects, machined parts
• Difficult Objects – clear glass, black rubber, Mylar, celophane-wrapped, stacked metal
• Combine – with Spatial Vision Robotics for vision guidance or Neocortex for automated intelligence for random situations
Traditionally, pallet inspection has been done visually with varying levels of manual handling or automated machinery, requiring the pallet to be lifted and flipped to see all surfaces. Stringent and frequent audits are required to reduce the variability of visual inspection. Traditional 2D vision is not as reliable as 3D due to the limitations of the technology and the structure of a pallet. Replicating manual inspection requires a flexible, automated vision system that quickly identifies a wide range of defects, including raised nails, and wood damage – whether split, loose, or missing – at a productive line speed. Universal’s 3D solution sees more of the pallet at a faster rate than current approaches.
CHEP, the global leader in pallet and container pooling services, has selected Universal’s 3D Inspection Application for automated 3D pallet inspection worldwide.