Multi-Station Layout, Dual Robots & Four Pallet Positions: How a Single 3D Vision System Addresses Complex Part Feeding

One 3D camera paired with two robots and four pallet positions can simultaneously identify three distinct workpieces — evaporators, black foam and brackets. This is not an "extreme challenge", but a real deployment completed by Transfer Technology at an automotive parts manufacturing site.

01 Site Conditions

The project involves multi-station linkage: 1 camera → 2 robots → 4 pallet positions, with diverse workpieces, strict placement rules, and stringent precision requirements.

Handling workpieces: 50 types of evaporators, 1 type of black foam, 7 types of brackets

Stacking method: Alternating layers of evaporators and black foam, with the top layer always an evaporator

Pickup requirements: Collision avoidance, layer-by-layer identification, model verification, and abnormality alarm

Challenges: Black foam is prone to deformation; evaporators feature a dimensional tolerance of ±3 mm; brackets require high-precision positioning. Traditional teaching and 2D vision solutions cannot deliver stable performance.

02 Workflow

Transfer Technology deployed the Epic Eye Laser L 3D camera. Mounted on a moving module, the camera switches shooting positions among four pallet positions (two for brackets, two for evaporators and foam) according to PLC commands.

Bracket identification: The camera moves to the bracket pallet position, recognizes the workpiece pose, and guides Robot 1 to pick and place the bracket onto the conveyor line.

Evaporator identification: The camera switches to the evaporator pallet position. The AI algorithm first verifies whether the top layer is an evaporator or black foam. If an evaporator is detected, the system calculates its precise pose and guides Robot 1 to pick the evaporator and fit it accurately into the bracket.

Foam identification: After evaporator removal, the camera performs a second scan to detect the quantity and pose of the remaining black foam on the current layer, guiding Robot 2 to pick and stack the foam sequentially.

Cyclic operation: The above workflow repeats continuously until all pallet positions are empty.

This entire process enables one vision system to drive two robots and process three types of workpieces, significantly cutting hardware costs and deployment complexity. Benefiting from Epic Eye Laser L’s strong resistance to ambient light interference and outstanding imaging performance on dark materials, complete and clear point cloud data can be captured even from light-absorbing black foam, laying a solid foundation for accurate calculation via subsequent AI algorithms.

03 Value Delivered

High Precision

With ±0.3 mm recognition accuracy and ±0.5 mm gripping accuracy, the system ensures one-shot precision for picking and placing evaporators, brackets and foam. It effectively eliminates line faults such as jamming, misalignment and collision, greatly reducing material waste and unplanned equipment downtime.

High Efficiency

The single recognition cycle of the entire vision system is controlled within 3.5 seconds. The streamlined workflow of the Epic Eye Laser L camera drastically cuts idle waiting time and substantially improves overall operational efficiency.

High Flexibility

The all-in-one solution supports 50+ evaporator models, 7 types of brackets and 1 type of black foam, enabling rapid integration of new product models. Engineers can conduct independent debugging and expansion without frequent line changeovers or manual teaching. It perfectly meets the production demands of multi-variety mixed-line manufacturing.