Skip to content

Code for the paper "Optimisation of formulations using robotic experiments driven by machine learning DoE"

Notifications You must be signed in to change notification settings

liweicao/centripeta

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

42 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Centripeta

This is a set of tools for controlling the centripeta robots in publication 'Optimisation of formulations using robotic experiments driven by machine learning DoE'.

top-view

This robot contains two parts: sample preparation part and sanple analysis part. A general scheme for the procedure is provided below. The continuous lines represent the materials flow, whereas dashed lines represent the information flow.

image

Installation

The easiest way to install centripeta is using pip:

pip install -e git+https://github.com/sustainable-processes/centripeta@master#egg=centripeta

To use the code for pH measurement, the PicoSDK C libraries need to installed. For details, please refer to Picotech website(https://www.picotech.com/downloads) and their github repository.

Running the experiments

Below is a simple example for running the experiments.

from centripeta import Dispenser
from pycont.controller import MultiPumpController
from commanduino import CommandManager
import pandas as pd

#This is the CSV file with the conditions to dispense into the vials
CONDITIONS_FILE="xp_initial_0.csv"

#Instantiate the communication with the robot
mgr = CommandManager.from_configfile('platform_config_ports.json')
pumps = MultiPumpController.from_configfile('pycont_config_less.json')
d = Dispenser(manager=mgr, pump_controller=pumps)

# Read in conditions and run the robot
conditions = pd.read_csv(CONDITION_FILE)
for i, condition in conditions.iterrows():
    #Dispense each surfactant
    d.dispense(pump_name="Plantacare818", volume=condition['Plantacare818'], speed_in=1500)
    d.dispense(pump_name="Texapon", volume =condition['Texapon'], speed_in=1500) 
    d.dispense(pump_name="DehytonAB30", volume=condition['DehytonAB30'], speed_in=1500)
    d.dispense(pump_name="ArlyponTT", volume=condition['ArlyponTT'])
    d.dispense(pump_name="CC7BZ", volume=condition['CC7BZ'])

    #Wait and dispense water
    time.sleep(10)
    d.dispense(pump_name="water", volume=condition['water'])

    #Wait and turn the wheel
    time.sleep(5)
    d.turn_wheel(n_turns=1)
    print('finish sample' + 'i')
print('Done!!')

To see more examples, take a look at the protocol directory, which contains the files for the robotic system control.

About

Code for the paper "Optimisation of formulations using robotic experiments driven by machine learning DoE"

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages