Pipette

• 

  The setup

Inspiration

Tedious tasks in labs

What it does

Automates decision making based on sensor data for the OpenTrons One platform (OT1).

How I built it

Adapted an optical spectroscopy device to the OT1. Implemented control system with feedback from a classifier using a Support Vector Machine and spectral data gathered from samples in the lab as training samples.

Challenges I ran into

Tuning of the classification algorithm. Talking to the OT1 in a consistent manner and integrating the decision into the protocol to be uploaded to the machine.

Accomplishments that I'm proud of

It works!

What I learned

Don't overengineer. We basically reimplemented some of the OpenTrons SDK for this task, duplicating efforts unnecessarily, in particular for a demo

What's next for Pipette

Integration into the SDK, abstraction of the classifier and sensor in order to enable easier addition of new sensors, methods, etc.

Built With

• gcode
• nano-drop
• opentrons-one
• python
• support-vector-machines