Bin picking
Empty a bin of mixed, randomly oriented parts, even shiny or reflective ones, no fixtures required.
ThRiVer sees the parts, picks the best one, and repeats until the bin is empty. New parts need no training data and no per-part tuning, just a description of the job. It handles the parts that usually defeat vision: shiny, reflective, jumbled, and never seen before.
How ThRiVer runs it.
What it delivers.
How it handles the hard cases.
A demo runs once. Production runs all shift. ThRiVer is built to keep going when an item is not where it should be.
No clear grasp
When a part is out of reach or buried under others, ThRiVer takes the next best pick instead of stopping, and comes back for it once the pile shifts.
Overlapping parts
It works out which part is on top and picks in that order, so it does not disturb the rest of the bin.
Every miss is logged
Each cycle records success, time, and cause, so you can see and fix the biggest source of lost picks first.
Built for production, across industries.
The same platform, set up by description, drops into very different lines.
Frequently asked questions
- What is robotic bin picking?
- Bin picking is a robot taking parts out of a bin where they lie in no particular order. The hard part is perception, not motion: the cell has to find a specific part in a random heap, work out how it is sitting in three dimensions, and plan a path to it without hitting the bin or the parts around it.
- Can a robot pick shiny or reflective parts?
- Yes, although conventional 3D vision struggles with them, because reflective surfaces scatter the projected pattern and parts reflect off each other and the bin walls. ThRiVer picks jumbled, shiny sheet-metal parts at up to 500 picks per hour with a >95% pick success rate, and needs no training images to do it.
- How long does it take to run a new part?
- The same day. A new part needs a description of the job or a CAD file, not a set of training images and not per-part tuning, so adding one is a configuration change rather than a project.