Plume – A long and diversified week !

It has been a long time since I last posted on this logbook, so I have many things to tell you ! Here is my day to day work of this week !

On Monday, I took a little too much time to write the description of our project on hackster.io (come check it out here), so on the afternoon I mainly worked on the lab work we have on the STM32 board.

On Tuesday, I have worked a lot on the mathematical model of our problem. I really think that thanks to that it will be easier for us to compute a precise location using our magnetic fields. The use of matrices to calculate the three different magnetic fields from the voltage measured in our coils, will enable us to use a precise calibration system taking into account not only the capacity of the individual coils to generate or sense the magnetic fields but also the mutual effects neighboring coils can have on each others. I have put all of these notations and calculation methods in a latex document so that anyone in the Plume team can see how the calculation will work.

On Wednesday, I haven’t really had time to work on the project because I was performing in an Improvisation show and thus I had to leave early. (By the way, if you want to see me on stage, I’d be glad to see you in the audience, so here is the fb page of my company : https://www.facebook.com/lacompagnieOnOff)

On Thursday and Friday, I tried to find a software way to suppress the noise we could have on the magnetic fields we will measure. Indeed, if we can have a perfect measure of the three magnetic fields our orthogonal coils emit, we will have no trouble computing the position and orientation of our sensing device, but as we will need to sense very weak magnetic fields, we might have a very low Signal-to-Noise Ratio. But we might reduce that noise considering that our 3×3 measured values should follow equations depending on only 3 variables, so if we were able to find the nearest point to our 3×3 matrix in this 3-dimensional variety, we would be able to get rid of a lot of the noise. It kind of looks like the way error-correcting codes work. Unfortunately, I haven’t found an efficient way of finding this nearest matrix, even with the help of a mathematician friend of mine, but I will keep this possibility in mind if the noise happens to be a too big handicap.

Finally, on Sunday I have worked with Olivier to conceive a good low noise amplifier for our lab tests. We have tested a few different amplifying circuit, with different resistor and capacitor values, and we finally found a great one, combining a resonant part with a low-pass filtering part. It works well, but we will still have to work on the coils and their power supply to have the range of use we wanted.

That’s it for this week ! It had the opportunity to work on very different subjects and it was really great !

See you soon !

Marc-O