Simulation of magnetic field coupling for electromagnetic pose tracking

The present document is about the thesis work that has been done together with Rokoko Co. between February and June 2024 and it was focused on simulating the magnetic field of the resembling structure of the Coil Pro product, a device launched by the company in 2023. This device generates a magnetic field in all directions and is detected by the other devices in the Rokoko ecoystem (e.g. Smartgloves, Smartsuit Pro). These are then able to compute their position and orientation in space relative to the Coil Pro, thanks to the sensors inside them who send their data in matrix form over WiFi to the connected computer, which is responsible for the post-processing and the real-time showing of the movement. Chapter 1 will present the company and in particular the Coil Pro product, which is the focus of our work. We'll enter into details on how it works and why we want to simulate the magnetic field on it. The results and discussions given in this paper expect that the reader is familiar with concepts of electromagnetism, although Chapter 2 revisits basic and more advanced concepts, which might be useful to understand this document in almost its entirety. We'll mathematically define the magnetic field, we'll describe its properties in matter and we'll introduce some examples to apply the theoretical formulas; these will be especially useful in the following chapter. The Appendix completes the review of magnetism by giving the reader additional results on not just magnetism but also about Coil Pro measurements and details. As a matter of fact, in Chapter 3, we'll describe the software used in the internship, kindly offered by DTU, which is COMSOL, a simulation software for physics systems. On top of that, we'll show some meaningful results to compare known analytical results from the field of magnetism with the simulation (e.g. a square and circular loop, etc...), which are presented theoretically in Chapter 2. These examples will be object of discussion as results from theory and simulation can differ significantly; this will give the chance to consider the reasons why there are errors and how to minimize them. Later, in Chapter 4, we'll cover all the setup and present results that we got from the simulations on the Coil Pro in a graphical and intuitive way. The analysis will be focused on how the design was brought out on Comsol with emphasis on the decisions taken for the drawing regarding objects, sizes and angles, on the presentation of the dimensions and parameters in the geometry, the mesh choices and the assumptions made. It will be on Chapter 5 that we compare these results with actual measurements and draw conclusions on the goodness of the product and the possible improvements that could be actuated. Finally, on Chapter 6, we'll investigate the effects that the windings have on each other, or to put it best, we'll calculate and possibly estimate the values of mutual inductances on the Coil Pro, as a non-ideality property present in any real-world application.