AccuPickCase Study
Bin Picking Auto Parts Using AI
Customer
The customer is a European auto parts manufacturer with over 40 years of experience. Specializing in high-volume production, the company sought a solution to improve the efficiency and accuracy of its bin picking processes for small, plastic components.
Challenge
Bin Picking Non-Reflective Plastic Components
The customer needed a solution capable of picking non-reflective plastic components in random orientations and placing them individually into separate small bags, with a cycle time of 3 to 5 seconds per pick. The low contrast of the components posed a significant challenge for ordinary 3D vision systems, which struggled to generate high-quality point clouds for accurate grasping and placement.
Solution
AccuPick + SolScan Complete Bin Picking Solution
Solomon’s bin picking system, AccuPick, combined with the advanced structured light capabilities of the SolScan 3D scanner, provided a complete solution to the customer’s auto parts bin picking challenge. AccuPick captured precise point cloud data to identify and pick each component using a FANUC LR Mate 200iD Robot. With thousands of items in the customer’s inventory, AccuPick’s Auto Labeling tool was employed to efficiently replicate each sample, significantly reducing the hours typically spent on manual annotations required by ordinary AI programs. This streamlined approach significantly reduced the time required for training AI models and markedly improved picking accuracy and operational efficiency.
