This repository contains code for TUNI acoustofluidic manipulation setup. The setup is documented in the following papers:
Kyriacos Yiannacou and Veikko Sariola, "Controlled Manipulation and Active Sorting of Particles Inside Microfluidic Chips Using Bulk Acoustic Waves and Machine Learning", Langmuir, 37 (14), 4192–4199, 2021.
Kyriacos Yiannacou, Vipul Sharma, and Veikko Sariola, "Programmable Droplet Microfluidics Based on Machine Learning and Acoustic Manipulation", Langmuir, 38 (38), 11557–11564, 2022.
Kyriacos Yiannacou and Veikko Sariola, "Acoustic Manipulation of Particles in Microfluidic Chips with an Adaptive Controller that Models Acoustic Fields", Advanced Intelligent Systems, 5 (9), 2300058, 2023.
The code was written and tested in Matlab 2019a&b and Matlab 2021a.
Probably the most interesting part of the repository is the script src/bandit_ctrl.m, which contains the implementation of our e-greedy and UCB1 control algorithms, and src/sfb_ctrl.m, which contains the script for AMA control algorithm (originally was about to be called "sparse fourier basis controller" but got renamed to AMA controller during the writing of the paper).
Also src/+experiments/ is of interest, as it contains basic scripts to run the single-particle, multi-particle manipulation experiments, scripts to learn the sparse regression controller, and a script to write the university logo.
To test the code, run runtests tests.