STAR Autoload & 1D Barcode Reading Upgrade

Hi everyone,

The core autoload and the 1D barcode scanning features on a PLR-powered Hamilton STAR have not been updated since:

With one notable exception:

So I upgraded it now:


Key Changes:

  • Created detailed Jupyter notebook tutorial in the docs with examples and architecture diagrams

  • Renamed and deprecated autoload methods for consistency (e.g., initialize_autoload, move_autoload_to_track)

  • Added comprehensive carrier presence detection methods for both loading tray and deck

  • Introduced granular barcode reading methods: IF a barcode has been detected a Barcode class will be returned for both (1) carrier barcodes and (2) container barcodes.


Created master infographics for alignment on machine and command nomenclature:


There will be more autoload and barcode reading commands along the way; I’m quite excited to get the 2D barcode reader fully integrated into PLR :slight_smile:

Happy automation :mechanical_arm:

3 Likes

Which 2D scanner are you gonna go with, the cognex one Hamilton uses?

1 Like

@cwehrhan yep, already 3D-printed an adapter for it to “piggy-back” onto the autoload sled (I’ve been told there is no OEM-designed “standard” way of integrating the Cognex 2D barcode scanner :joy:)

Though I am unsure whether I want to use the STAR firmware commands for the Cognex… or just integrate the Cognex directly into PLR.
Do you have suggestions based on having worked with the Hamilton-Cognex fusion? :slight_smile:

Really? Which scanner model? We have little brackets I believe machined, that are all identical for our robots lol

Hmmm so I haven’t used the Cognex since they’ve moved to a newer smaller one, my experince with it is 5 years ago was a pain. TBH a huge bottleneck of setting up the Cognex on our current robots is working with the STAR lib… which I had issues with in the past. It’s “ok”. I have painful memories of not being able to tweak all the settings that you can, exposure, lighting, field of view…etc. It only exposed the bare minimum, but maybe all this has changed. I can take a peak at the library Monday. I think also its not in use as much as the 1D so it was not robust back then, maybe still is.

My vote would be interface directly with the cognex, they are robust and good tech. I actually have a handful at my local hack space from Amazon warehouse upgrading and they gave us a ton, they are battle tested

1 Like

Hi @cwehrhan (busy Christmas time :slight_smile: ),

We’ve got a simple Cognex DM260. There is indeed a Hamilton metal bracket to “piggy-back” onto the autoload 1D barcode scanner. However, (1) our service team told me this is no “official” way of doing it, and (2) I have encountered some hardware issues with it:
It places the Cognex into a suboptimal position in which its communication/power cables have to be cable-tied to the autoload sled, and depending on how one does that the Cognex still scratches the autoload’s 1D barcode scanner during 1D barcode scanner rotation.
This hinders the 1D barcode scanner from fully rotation and can lead to 1D barcode scanning errors.

I’m designing my own adapter at the moment to overcome this little issue, happy to share once we’ve got something nice :slight_smile:

Thank you for sharing your experience - this is highly appreciated!

I had no idea how to approach the Cognex, but your comments and Cognex’ open, very detailed documentation make clear what I am going to do: integrate the Cognex DM260 itself into PLR :slight_smile:

1 Like

The best thing to do is go straight to the Cognex DM260. The STAR library never really had enough knobs to make 2D scanning reliable over time, especially when the lighting or label quality changes. If you keep the integration inside PLR, you can change the exposure and decode settings yourself, and you don’t have to fight the STAR abstraction every time you need to make a change. The only thing I would watch is cable strain once you finish your custom bracket. The DM260 is solid, but if the routing isn’t clean, repeated sled motion will wear out connectors faster than you think.

2 Likes