tingeypa

Thanks! I have lots of pics of a small set of pieces, but they are not the ones in your batches. I think your project is a great idea! I was considering just making a fully automated scanner for my competition submission, but I had already spent time on the arms. The oak-d cameras would be easy to automate together with a rotating/rocking platform, or a rotating platform with an arm to arc the camera over the subject. You could automate scanning, uploading, and running the model training. You could even use OpenCV or other AI models on the camera to detect the location of pieces (i.e. define the bounding box) to assist in a detection model (rather than classification).

Thanks Thorsten. Those sound like good alternatives. I will take a look. PI4 was good for this application for a few reasons. It can connect to the AI camera directly over USB3. The screens showing the camera output can be connected to the PI, or it can be controlled via VNC Viewer which avoids having many cables going to the robot. Lastly, for this sort of robot you need to coordinate the three motors to move in sync, ideally starting and ending each movement together moving at controlled speeds. If they don't move in sync the gripper may move through a path that bumps into things. These are a slightly unusual requirements, but they could be achieved with the PI + Buildhat.

Thank you! The Pi foundation recently created the BuildHAT which can control four Lego motors/sensors using LPF2 connectors. It has an API in Python and C# and appears to be supported by Lego. The LEGO Education SPIKE Prime Expansion Set has a new PCB Panel designed specifically to mount a Pi (or similar PCB). The robot programming was not too hard as the OAK-D APIs (DepthAI) and the BuildHAT both have Python APIs. That said, I had to learn Python.

I am new here so hello all! I am a software engineer with a long history in Lego and Mindstorms. Recently I have been working on an entry for an AI competition which I thought may be of interest. The competition was to make something related to industrial automation that used a DepthAI camera (Oak-D-Lite) combined with Lego. I have been wanting to build a Delta Robot for a while and here's my submission. This uses a Raspberry Pi BuildHAT to control 3 Large Angular motors to drive the arms and one Technic motor for the gripper pump. The BuildHAT is a great way to control Lego motors and its cheap and open source. The DepthAI camera (which is an amazing bit of kit!) detects objects and sends their location to the Robot for pickup. You can do lots more with Oak-D cameras, but I am just doing basic object locations. There are quite a few other entries in the competition that are worth a look. Search #oakdlitecontest, or you can see the Top10 here - https://form.jotform.com/221086334784156