The defect detection model can be run on basic hardware and operates on-site, ensuring customer images do not leave the worksite. The startup charges customers per model and aims to have the model ready before retooling is complete when a line change to a new mold is planned. The founders have a background in automating manual tasks and have previously worked on self-driving cars at Argo AI, Latitude AI, and Stack AV.
Key takeaways:
- Bucket Robotics has launched a new product that uses CAD models to create custom defect detection models for manufacturing, specifically for injection molded and cast parts.
- The defect detection models are created by generating a large number of variations of the 3D model, simulating various defects and then training a vision model from these images.
- The model doesn't require high-end hardware and runs purely on-device, ensuring customer images don't need to leave their worksite. The company charges customers per model.
- Bucket Robotics was founded by Matt and Steph, who have a history of working on tasks performed manually that they want to automate, including self-driving cars at Argo AI, Latitude AI, and Stack AV.