Context & Aim
Bin picking is one of the great areas of interest to manufacturers. To save space and the cost of automation, manufacturers would like a solution for robots to pick, randomly orientated parts, out of a bin. There are many commercially available bins picking solutions that address a subset of bin picking challenges one such system is Photoneo Bin Picking Studio supplied by Multipix.
The aim of the work was to build a physical implementation of Photoneo BPS in conjunction with a robot to:
- Build the team’s capability with a planned capital investment.
- Demonstrate our capabilities in bin picking using market-ready technology.
- Share information and share setup and capability of a commercial solution with REC members.
- Communicate and engender confidence in the market in our ability to take on their challenges even with traditionally difficult products.
Challenge
Build a working bin picking cell based on state of the art commercial solution picking multiple product types using Photoneo Bin Picking Studio in 3 weeks to DEMO to the REDF Consortium and Irish Industry.
While commercial bin picking solutions exist in the marketplace. Robotic systems still struggle with object identification when it comes to clear, deformable, variable , and reflective parts. Piled, interlocked, and jumbled up parts pose problems for robotics picking.
The technical challenges integrated into this project include:
- Identification and pick of non-symmetric complex shaped part.
- Complete random orientation.
- The solution should demo on 3 different products.
- Demonstrate optimum process speed for the product chosen.
Solution
Photoneo Bin Picking Studio and 3D scanner supplied by Multipix were integrated onto a work cell with a UR 5 robot and used to demonstrate random bin picking on 3 different shaped bread products.
Before implementation, the following setup and configuration requirements need to be met.
Pre-planning:
- Define end of arm tooling design suitable for the product.
- Mount EOAT gripper and test operation on Robot.
- Ensure Safety Requirements for the cell are correct (collaborative mode or Industrial mode)
- Have CAD models (in STL file format) of:
- A: The product you wish to use.
- B: The EOAT gripper you wish to use.
- C: The cell layout you wish to use.
- Have the Software files of the UR5 robot required for integration with BPS.
- Have the network cables required for a physical connection between BPS, UR5, and Scanner.
- Have n HDMI cable to connect between BPS hardware and the Monitor screen.
Setup:-
Once all of the above have been satisfied the next steps in implementation are as follows:
-
-
- Cell build and hardware integration.
- Bin Picking studio hardware ad software integration.
- Gripper grasp planning include all the potential grasp orientations and pick points.
- Calibration of vision robot and end of arm tooling.
- The setting of parameters.
- Optimisation of layout, path planning, and robot parameters.
- Repeatability testing.
-
Results
- The system has demonstrated repeatable picking and emptying of random bins of complex shaped product at a pick rate of 6ppm. Faster processing speeds would be achievable with symmetrical geometric shaped products.
- Picking of 3 separate products from Random bins.
Impact
- Increased knowledge for IMR and their partners.
- Investment in bin picking software and 3d Vision.
Benefits to the client:
- Provides IMR partners and customers insights into the requirements and capabilities of a typical bin picking solution.
- IMR has invested in a commercial bin picking solution that Irish manufacturers and IMR partners can avail of.
- IMR staff have been fully trained on the system and have built up considerable knowledge on vision robotics and bin picking solutions through this and other related projects.
THEMATIC PILLAR
Robotics and Automation
