Inspiration
As cobots become increasingly popular in manufacturing with its diverse industrial applications, we hope to allow for companies to estimate the probability of protective stops or grip losses, which are both symptoms of machine failure.
What it does
Our project analyzes data, which came from a study on real-time operational data of the UR3 cobot to test for anomolies
How we built it
We have developed systems for evaluating new incoming data. This includes an API endpoint for a single-sequence in time that is connected to our central dashboard website and data model. We also have a continuous-time server that utilizes TCP sockets to stream continuous, real-time data to predict possible failure.
What's next for UR3 CobotOps Predictive Model
Ideally, this could severe as an automated way to manage UR3 cobots, for both human maintenance and check-up and also by scripts to stop the robot before any errors occur.
Built With
- fastapi
- github
- jupyter
- node.js
- numpy
- pandas
- python
- react
- scikit-learn
- sockets
- tensorflow
- typescript
- uvicorn
- websockets
Log in or sign up for Devpost to join the conversation.