Bin picking with 3D vision camera

Bin-Picking with 3D
Challenges & Solutions
Customer Stories
Find the Right Camera

Detect

Zivid Two delivers high resolution and extreme precision, native color 3D point clouds with minimal occlusion, and excellent artifact suppression. Significantly improving object recognition and increasing the number of
detectable parts.

Pick

Zivid Two enables more reliable detection of object boundaries for grasp planning and a true-to-reality representation of object size, rotation, and absolute position in relation to the robot coordinate system. Essential for accurate picking, avoiding mispicks, and crashes.

Place

Zivid Two, with its groundbreaking trueness, enables more demanding place operations. Placing or inserting more accurately for further processing, with known position and orientation, without colliding or damaging the parts.

Work faster. Zivid Two delivers significantly higher quality at faster speeds than typical laser scanners used in bin picking today. Capture speeds <300ms for high dynamic range scenes, including shiny and reflective objects.

Example of bin with metal parts from laser sensor

Benchmark 1. Point clouds from a laser structured light camera

Metal parts in a bin captured with a high-end stereo sensors produce point clouds

Benchmark 2. Point clouds from a high-end stereo camera

Zivid 3D camera captures high-quality point clouds of metallic cylinders in a bin for pick and place applications

Benchmark 3. Point clouds from Zivid Two

Challenges & Solutions

Bin picking is still a challenge for robots. Even at large manufacturers, adoption of bin picking stations is low, and at SMEs the number is still close to zero. Unpickable objects are typically complex-shaped, hard to manipulate, entangled, and/or highly reflective machined or polished metals. The difficulty stems from the inherent uncertainty in physics, perception, and control.

See tiny and detailed objects, densely stacked or randomly arranged.

Zivid Two provides high resolution and precision point clouds with minimal occlusion, capturing shapes and sizes with all the fine details.

Very small, thin sheets, highly detailed, and irregular objects.
Densely stacked with a little gap or randomly arranged and piled on top of each other.
Distinguish features smaller than 5mm for reliable object recognition and separation of boundaries for picking.

Image shows metal parts and cylinders in a bin captured by Zivid for picking applications

Point cloud example with Zivid Two - Various metallic objects in a bin, typical for industrial bin-picking scenarios. View in 3D HERE

See shiny, reflective, machined, and polished parts

3D HDR and Artifact Reduction Technology (ART) ensures excellent suppression of imaging artifacts from reflections, interreflections, specular highlights, and high contrast transitions.

Shiny and reflective, sheet metal, cylinders, washers, molded, machined, polished, and chrome-plated parts.
Capture speeds <300ms for high dynamic range scenes such as shiny and reflective

Zivid 3D camera captures high quality point clouds of reflective and shiny metal parts

Point cloud example with Zivid Two - Highly shiny and reflective, chrome plated, machined and polished parts. View in 3D HERE.

Experience fewer occlusion outliers

Zivid Two is ultra-compact with a small baseline and more optimal occlusion performance than larger baseline cameras and scanners. A large baseline gives excessive projector/camera shadowing and loss of scene details, creating holes, and missing data affecting the detection. There's also a risk of missing smaller objects hidden by shadows in the bin, for example, close to corners and edges, especially when picking from smaller boxes.

Learn more about occlusion here.