Hook up your webcams or cameras, run the wizard, take 15 snapshots of a pattern in the robot workcell, load a standard CAD file of your part, run the software ‘training’ step, and you’re ready to go. The calibration files can interface to your C or C++ real-time control program, or our robotic experts will do it for you.
Start your Universal application, such as the Box Mover for any additional sensor integration, such as a structured light sensor. After calibration of cameras and sensors, Spatial Vision Robotics software coordinates 3D input to the robot by getting an overall 3D view of the work space, detecting all objects possible, selecting an object to be moved, and monitoring the motion of the robot as it places the object.
Key Features
• Easy camera/sensor alignment
• Intrinsic calibration of sensor-to-sensor
• Estimates 3D camera accuracy using vision processing tools
• Extrinsic calibration of sensor-to-robot (hand-eye) using patterned fiducial
• Highly accurate software with inherent 0.1 sub-pixel accuracy
• Acquires 3D point cloud of entire work space with structured light sensors
• Acquires images of objects from webcams and cameras (USB & GigE)
• Object tracking provided in six degrees of freedom (X, Y, Z, Rx, Ry, Rz)
• Object recognition by matching parts based on 3D shape, even with partially occluded objects under varying lighting conditions
• Identifies up to 3 objects simultaneously
• Real-time vision guidance
• Tracks stationary or moving objects
• Uses standard AutoCAD DXF file for object recognition
• High speed recognition enables typical system speeds of 10 -12 objects per minute
Integration with Advanced Software
• Integrates with Antonomy for >15 Degree-Of-Freedom reactive control of robots
• Integrates with Neocortex for robotic adaptive learning of tasks
About Universal Robotics
Universal Robotics’ signature technology, Neocortex™, is a sensory-motor based form of artificial intelligence that enables moving machines to learn from their experiences and perform tasks that are unsafe or difficult for humans. Neocortex was co-developed over seven years by NASA and Vanderbilt University, and was funded by U.S. Department of Defense. The company is automating intelligence.
Because Neocortex’s learning requires observation and interaction with the physical world, Universal created an, accurate 3D vision guidance product called Spatial Vision™ Robotics that provides Neocortex with information about its surroundings using 3D sensors.
The company has also developed motion control software, Autonomy, that automatically generates robot motion control code that allow machines to react and move dynamically in real-time to random input.
Includes Spatial Vision Tools
• Estimate intrinsic calibration accuracy
• View or save left/right .AVI video or BMP images; combined 3D anaglyph or Laplacian of Gaussian BMP images; or raw feed or rectified .AVI videos.
• Intrinsic calibration files available with C or C++ interface for custom applications
Below are examples of Universal Robotics Applications. All offer under 24-month payback.
• Six axis robot from Motoman, Kuka, or Fanuc
• 4 – 6 axis robot from Motoman, Kuka or Fanuc
• Six axis robot from Motoman, Kuka, or Fanuc
• 5 kg – 50 kg payload
• Vertical Reach: 0.9M – 3.5M
• Horizontal Reach: 0.7M – 2.0M
• 3 – 8 parts/minute (often includes surface inspection)
• Enhanced Sensor Suite (± 2.5mm)
Two pairs of GigE cameras
Structured light sensor
Universal Robotics Mixed Depalletizer
• 10 kg – 500 kg payload
• Vertical Reach: 2.7M – 3.0M (pallet) or 5.5M (box)
• Horizontal Reach: 2.0M – 3.1M
• 6 – 15 boxes/minute
• Partial or full pallets up to 60” high
• Pallets up to 48” square
• Any number of pallet layers
• Standard Sensor Suite (± 5.0mm)
Two pairs of USB cameras
Structured light sensor
• 5 kg – 50 kg payload
• Vertical Reach: 0.9M – 3.5M
• Horizontal Reach: 0.7M – 2.0M
• 8 – 20 bags/minute
• Reflective Surface Capability
• Standard Sensor Suite (± 5.0mm)
Two pairs of USB cameras
Structured light sensor
