Magneto Suture is the code repository for the MDPI Robotics journal paper submission titled: "A Control Interface for Autonomous Positioning of Magnetically Actuated Particles using an Artificial Neural Network."
By: Victor Huynh, Basam Butawak, Minh Quan Do, Elizabeth Ankrah, Irving N. Weinberg, Nathalia Peixoto, Qi Wei, Lamar O. Mair
Electromagnet arrays hold significant potential in untethered guidance of robotics. However, complications obtaining accurate models of electromagnetic fields pose challenges to precision control. Additionally, control schemes capable of being applied to electromagnet arrays of any configuration may significantly expand usefulness. Here, we develop and compare two regression neural networks for controlling a simple four-coil array of electromagnets. We demonstrate millimeter-scale resolution control of a spherical magnet using a controller trained using a 5-input regression neural network.
The data/ directory contains experimental results observed in the study and plotting utilities.