Random Sheet Metal Picking

Sheet metal parts come in a wide range of specifications, with reflective surfaces and high output volumes. Manual loading is slow, physically demanding, and prone to part damage, while fixed fixtures struggle to adapt to multi-specification production. A DexForce 3D vision-guided robotic solution now achieves accurate sheet metal loading with ±1 mm repeat positioning accuracy.

• Slow and labor-intensive manual handling.
• Manual loading and unloading is inefficient and physically demanding, with inconsistent quality and a risk of part collision damage.
• Poor production flexibility.
• A wide variety of sheet metal specifications makes manual changeovers slow.
• Harsh working conditions.
• Oily and dusty shop-floor environments pose health risks to workers handling heavy parts.

The DexForce embodied AI spatial intelligent sensor performs ultra-fast imaging of stacked sheet metal parts on the conveyor, outputting complete, high-quality 3D point clouds. Powered by DexForce embodied AI algorithms, the system runs target segmentation and 6D pose estimation to identify the optimal grasp target. The system then guides the industrial robot to precisely pick each sheet metal part and place the part onto the fixture. Once processing is complete, finished parts are automatically unloaded into the parts basket.

• Complete imaging of highly reflective parts.
• The embodied AI spatial intelligent sensor generates high-quality point clouds of reflective sheet metal parts, with stable imaging performance under heavy oil contamination.
• Millimeter-level repeat positioning.
• Visual repeat positioning accuracy of ±1 mm meets high-precision production requirements.
• Fast changeovers across multiple variants.
• Precision calibration, multi-model compatibility, and collision avoidance optimization ensure efficient and stable loading, with one-click program switching for flexible production.
  

The DexForce spatial intelligence suite enables a single station to handle both random picking of small parts and orderly stacked picking of large parts. Mixed-model production of multiple sheet metal categories now runs smoothly, with significantly improved line stability and production flexibility, and measurable gains in both yield and cycle time.

Robotic picking and loading with 3D vision gives manufacturers a practical path to automate reflective sheet metal handling. Reliable point cloud imaging under oil contamination, sub-millimeter robot guidance, and one-click changeovers allow a single station to support multi-specification production without manual intervention.