Current automated box moving solutions are in very structured environments – production lines, part of automated
processes, or built into material handling equipment. Neocortex Box Mover delivers patented, breakthrough box
moving that replicates human capabilities in unstructured environments where it is dangerous, difficult or costly:
Integrated with world-class dexterous Motoman SDA20D or SDA10D dual-arm robot
Speeds up to 7 random boxes per minute
One box payload weight up to 70 lbs.
Neocortex Box Mover knows the task of moving boxes, and analyzes each situation to determine the best way to
pick up a box – just like a human – and then lifts and moves it to the desired location:
1. Visually analyzes box in 3D
2. Determines box size
3. Identifies when robot is engaged with box
4. Determines best strategy to pick up box
5. Picks up box
6. Moves box to desired location
7. Sets box down
If any of these approaches don’t work, Neocortex adapts and suggests other approaches to try, without
programming. Here’s an example of different approaches to lifting boxes:
Since Neocortex is scalable to any platform, the various hardware and software componentsof the system can be adjusted to ensure overall performance and
accuracy requirements for the box moving solution are met. For instance, high-speed GigE cameras with higher resolution and greater focal length than USB cameras can be used for improved visual accuracy.
In general, the following performance is possible with standard components:
10 kg – 20 kg per arm payload
Height: 1.8M – 1.9M
Reach: 0.7M – 1.1M
5 – 10 boxes/minute
Typical Accuracy: 4 mm
Requires Universal Robotics Sensor Servoing
Requires Universal Robotics Neocortex
Neocortex Depalletizer incrementally adds patented-backed ability to automate the depalletization of a random stack
of boxes to the Neocortex Box Mover capability. This is something never before accomplished in an industrial
environment – while providing a great return on your investment.
Designed to work with world-class dexterous Motoman SDA20D or SDA10D dual-arm robot
Leverages Neocortex Box Mover
Designed to fit into an existing work cell designed for humans
Unload up to four rows of boxes within the robot height limits
Assesses how pallet is stacked; decides which box to pick up next
33% or greater improvement in efficiency
29% increase in throughput
1% increase in accuracy and reliability
1% less damage and theft
2% less scrap and raw materials
ROI typically less than 24 months
Operates on standard computers, uses off-the-shelf hardware
Operates in existing facilities in common human work cell; no major plant changes required
Spatial Vision Robotics delivers scalable 3D vision guidance at 2D prices for Motoman robots and retrofit situations of all existing robot brands. It delivers market leading price/performance vision guidance. It provides stereo vision using just the right mix of standard webcams or GigE cameras
with 3D world-class pattern matching to determine true 3D for six degrees of position and pose information. When combined with PC hardware, you have a complete vision guidance system integrated with your robotic application. Adding tailored high-speed servoing completes your part picking, box moving,
pallet sorting, or automated kitting job.
Key features
Hook up your webcams, verify image acquisition, run the alignment 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, C++ or MATLAB robotic real-time control program, or our robotic experts will do it for you.
Easily acquire images from USB webcams & GigE cameras
Align loosely-placed webcams without specialized hardware
Calibrate camera-to-camera and camera-to-robot (hand-eye)
Object tracking provided in six degrees of freedom (X, Y, Z, Rx, Ry, Rz)
Match parts based on 3D shape, even with partially occluded objects under varying lighting
Identifies 3 – 10 objects simultaneously
Real-time vision guidance
Can track stationary or moving parts
Use standard AutoCAD DXF file for parts of interest
Estimate 3D accuracy on-the-fly using vision processing tools
