Moat Audit: ABB's Object-Grasping Patent Reads Like Factory IP, Not Lab IP

The question a moat audit asks is whether IP is defensible where the company actually makes money. ABB's grant US11007648B2 ("Robotic system for grasping objects," issued May 18, 2021; inventors Steven Wertenberger, Thomas Watson, Matthew Sallee) answers it through classification: the B25J manipulator codes sit next to B65G 47/52, a conveyor-handling code. This is grasping fenced inside a material-handling line — ABB's home turf.

The mechanism is vision-guided grasping integrated with conveyance: B25J 9/1697 (control using a vision sensor) and B25J 15/0616 (vacuum/suction gripping) describe how objects are seen and held, while the B65G code describes them arriving on and leaving via a conveyor. The claim, in other words, is about the whole pick station, not just the gripper — the integration is the invention.

For the manipulation beat, integration claims are strategically distinct from mechanism claims. A pure gripper patent fences a part; a system claim that ties perception, control, and conveyance together fences a workflow. Workflow claims are harder for a competitor to avoid because avoiding them means re-architecting the line, not swapping a component. That's the kind of moat that holds in a real factory.

ABB is an incumbent industrial-automation giant with a deep B25J portfolio, and this grant fits that posture exactly: it protects the integrated grasping cell that ABB sells, against integrators and rivals building competing pick stations. The inventor list and the conveyor classification both say this is product IP, not speculative research — the boring-robot-that-already-ships kind of asset.

Caveats. System claims that recite many cooperating elements can be easier to design around than they look — omit or alter one recited element and you may fall outside the claim. Breadth in a system claim is a double-edged thing. And industrial pick-and-place has heavy prior art, so the inventive step is the specific cooperation recited, not the existence of a grasping line.

Moat verdict: defensible where ABB operates, because it fences an integrated workflow rather than a swappable part — but only as broad as the recited element set. Pull US11007648B2 on PatentBear and read claim 1 for which elements are mandatory; the ones a competitor can't avoid are the moat.