3D printing failures are a waste of time, material, and energy. PrintGuard by Oliver Bravery offers real-time 3D print failure detection using AI on edge devices that addresses this. It’s designed specifically for edge devices like personal computers or embedded systems, and uses a custom computer vision model to achieve live monitoring through camera feeds and detect failures as and when they happen.
The tool has a web-based interface where the user can watch feeds from multiple cameras that are connected to the local system. Each camera is typically positioned to monitor a different printer. In addition to the live feed, the interface displays detection status and also allows the toggling of failure detection for each camera so the user can selectively monitor their ongoing prints. It integrates with 3D printer management tools like OctoPrint, which enables the automation of actions like pausing or cancelling a print job when an issue is spotted.
Users receive alerts promptly through push notifications on their desktop or mobile devices. Notifications include a snapshot of the failure that was detected and options to take action manually. The system will deal with the situation based on user-defined actions after a countdown that’s customizable. PrintGuard is installable via pip or Docker, and supports secure remote access using tools like Cloudflare Tunnel or ngrok, along with local network access. The project being open-source brings advanced failure detection to hobbyists and professionals alike.
