Operate: virtual training and commissioning of a simulated arm
Twin: a twin-ready view of the arm and its cell
How to read the labels
Every record on this site says how sure the project is of it, in words and with a shape of its own, so colour is never the only signal. This page explains each label and shows real records from the corpus as worked examples: the same cards you meet in the References band at the foot of every section.
Confidence: Verified
A manufacturer or IntelliMake document says so, or it was observed directly. It does not mean anyone measured it at S6.
Confidence: Inferred
Reasoned from Verified facts. The reasoning is printed with the record.
Confidence: Assumed
Plausible but untested. The record says how it would be checked.
Kind: Gap
Not known. No figure or answer is given in its place.
Kind: Contradiction
Sources disagree. Both sides are shown, and the site does not pick one silently.
Kind: Recommendation
Project advice drawn from the records it cites. Not a fact about the machine and not a confidence class, so the filter never hides it.
Awaiting cell access
Awaiting cell access: could be confirmed or corrected once the physical S6 cell can be observed.
This project's simulation and agent design, not the physical cellProject design, not the physical cell
This project's simulation and agent design, not the physical cell.
This course draws on 642 records: 518 Verified, 18 Inferred and 3 Assumed claims, 68 gaps, 23 contradictions and 12 pieces of project advice. 39 are awaiting cell access, and 20 describe this project's own design.
Verified
Verified means a manufacturer or IntelliMake document says so, or the project observed it directly. It does not mean anyone measured it at the physical station: a Verified figure is what its source states, no more.
UFACTORY states the xArm 6's maximum payload of 5 kg, so the record is Verified1.
This robot arm has 6 axes, with 5kg payload and 1m/s speed.
The difference between UFACTORY xArm5, UFACTORY xArm6 and UFACTORY xArm7 (Help Center article) · UFACTORY · Section 1 'Main difference in brief', xArm 6 entry
Open the sourcePlain http: this publisher has no certificate for that name, so what loads cannot be verified.
Inferred
Inferred means the project reasoned it from Verified facts; no source states it outright. The card prints the reasoning under "Why we infer this", so you can follow each step back to its own sources.
No source says in so many words that the collaborative label leaves a given installation unsafe to share. The project reasons it from two statements in the xArm manual: a safety assessment is required each time the arm is installed, and the whole system should be assessed when the arm works with other machinery2.
Calling the xArm 6 'collaborative' does not make a given installation safe to share with people: the xArm manual says a safety assessment is required each time the arm is installed, and that a comprehensive safety assessment of the whole system should be performed when the xArm works with other machinery.
Why we infer this:UFACTORY uses 'collaborative' as a product description (app-504, app-509). The xArm manual separately requires a safety assessment for each installation and for the whole system when the arm works with other machinery. So the label describes the arm's features, not a guarantee that a particular cell allows people to work beside it.
Evidence · 2 citations
A safety assessment is required each time installed.
Assumed means plausible but untested. Every Assumed card says how it would be checked, under "How this would be checked". Treat it as a working hypothesis, never as a fact.
This course holds only three Assumed records: where the stations sit relative to each arm, how the IntelliMake agents exchange information, and what Camera C1 can see345.
So the Assumed option in the filter looks nearly empty on most pages. That is expected. Where the sources are silent, the project records a gap rather than a guess, and keeps a guess only when it says how the guess would be tested.
The S2 and S6 stations sit on opposite sides of the laser processing stations (S4/S5), so a single arm probably could not reach both the receiving and staging area and the post-processing area.
Awaiting cell access: this could be confirmed or corrected once the physical S6 cell can be observed.
How this would be checked:Measure the as-built positions of S1, S3, S4 and S6 relative to each arm's base and compare them with the published reach of each arm.
Evidence · 2 citations
S2 UFactory xArm 850 Cobot
IntelliMake Phase 1 Factory: Autonomous Production Demonstrator (infographic) · IntelliMake.org · S2 station label
Camera C1's field of view covers the S1 receiving area that S2 picks from, so C1 could confirm material presence and state for S2.
Awaiting cell access: this could be confirmed or corrected once the physical S6 cell can be observed.
How this would be checked:Observe C1's physical mounting and field of view at the cell, and whether its output is available to S2 through S11 or the shared hub.
Evidence · 1 citation
Camera C1 – Receiving
IntelliMake Phase 1 Factory: Autonomous Production Demonstrator (infographic) · IntelliMake.org · Camera C1 icon placement
A gap is something the sources do not settle. It is never phrased as a fact, and where a gap stands in for a number, no number is shown. Each section lists its gaps under Open questions, and the card links the records the gap concerns.
No source documents where S6 puts a part for each of its three routes: inspection, rework or shipping6.
Where S6 puts a part for each of its three routes (inspection, rework, shipping) is not documented. The layout has no labelled inspection station (the topic mapping's S9 does not appear; int-037), the rework return path is unlabelled (int-039), and no source names S6's drop-off point for shipping.
Awaiting cell access: this could be confirmed or corrected once the physical S6 cell can be observed.
When sources disagree, the card says so and shows both sides with their citations. The site never picks a side silently. Where the teaching uses one side's figure, the contradiction sits next to it.
UFACTORY's own documents give two different ranges for the xArm 6's J27.
Sources disagree on the range of the xArm 6's joint 2 (J2): the online hardware manual (specifications and preface) and the product page give -117° to 116°, but UFACTORY's support article comparing the xArm 5 Lite, 6 and 7 gives -118° to 120°, and the xarm6 URDF's default limits are -2.059 to 2.0944 rad (about -118° to 120°).
Awaiting cell access: this could be confirmed or corrected once the physical S6 cell can be observed.
Project advice is this project's recommendation, drawn from the records it cites. It is not a fact about the machine and not a fourth confidence class, so it carries no confidence badge and the filter never hides it. Its card lists the records it is based on, each with its own badge.
The project advises treating the xArm 6's collision detection, safety boundary and reduced mode as configurable controller functions, not validated safeguards8.
Kind: RecommendationThis project's simulation and agent design, not the physical cellProject design, not the physical cellsafety-594
Project advice
Treat the xArm 6's collision detection, safety boundary and reduced mode as configurable controller functions, not validated safeguards: they must not replace risk-assessed protective devices wired to the EI and SI inputs.
This is advice from this project, based on: safety-578 Gapsafety-559 Verifiedsafety-560 Verifiedsafety-561 Verifiedsafety-564 Verifiedsafety-565 Verifiedsafety-544 Verified
Why:No fetched UFACTORY document gives these functions a Performance Level or SIL. Collision detection is a current-model comparison that UFACTORY says can false-trigger and can be switched off; the safety boundary and reduced mode are settings. A function with no documented rating cannot stand in for a rated protective device.
This project's design, not the physical cell
Some records describe this project's own simulation and agent design rather than the physical cell. Whatever their class, they carry the label "This project's simulation and agent design, not the physical cell" beside the badge, and on the section pages they sit under a heading with the same label.
Awaiting cell access
Some records could be confirmed or corrected once the physical cell can be observed. They carry the marker "Awaiting cell access" on the card, and the same mark follows their reference chips. The filter can show these records on their own, so they can be found and checked together.
The two Assumed records shown above both wait on cell access: where the stations sit relative to each arm, and what Camera C1 can see35.
Reading the numbered markers
In the teaching, a claim ends in a small numbered marker. The number is the record's place in References at the foot of the page, and the shape before it is the record's class, so you can see at a glance how firm the sentence is. Screen readers announce the number and the class in words. Hover over or focus a marker to preview the record; follow it to jump to the full card with its evidence.
7 A screen reader hears: “Reference 7, Contradiction”.
8 A screen reader hears: “Reference 8, Recommendation”.
Using the filter
On each section page, References opens with "Filter records". Untick a class to hide its records. Teaching that relies on a hidden class folds to a short note with a button to show it again, and a line at the top counts what is hidden. The cell-access option shows only the records awaiting cell access. Project advice is always shown. Nothing is hidden until you ask, the choice is remembered only in this browser, and without JavaScript every record is shown with its label.
References
Every record this page cites is shown above as a worked example, or listed here with its evidence.
The current assumption for the IntelliMake agents is that they do not talk to each other directly: they exchange information, and may collaborate, through a central shared knowledge space (a hub). IntelliMake leadership has not finalised this.
How this would be checked:IntelliMake team leadership's design for the shared knowledge space: its protocol, schema and hosting (open question Q9).
Evidence · 1 citation
The agents will not talk directly to each other, rather will exchange information in a central hub and possibly colloborate in that space. That has not been fully fleshed out by the iIntelliMake team leadership.
Owner statements, 2026-09-21 (verbatim) · Derek Stringfellow (project owner) Derek Stringfellow, project record · Project-internal record · retrieved 2026-09-21 · cited by 1 record here · Project copy; not published on this site.