[vlovich123]

It’s not uncommon to fuse magnetometer with other sensors. We did that for indoor navigation at Apple. In fact, we had prototypes that used only magnetometer and it worked fairly well for what it was but the low update rate and poor resolution meant that it worked to like 10-30 meters which wasn’t usable for indoor by itself. Of course, for indoor there was a lot more “texture” indoors for the commercial magnetometer of the time to pick up whereas outdoor it gets trickier. Is that similar to how MagNav works just with higher quality more sensitive magnetometers?

[shannonclaude]

So the role that magnetometers play in sensor fusion w/ IMU data is for yaw/heading/magnetic north estimation. In short, it aids your orientation (RPY estimate). However, with MagNav, they play a large role in supplying information that allows you to decrease your drift rate.

https://www.sagemotion.com/blog/how-does-imu-sensor-fusion-w...

[vlovich123]

Yes, we were doing all of that at Apple over ten years ago. I was specifically pointing out that just like WiFi has “texture” that lets you machine learn a position, so does the mag field, letting you determine an absolute position estimate purely from the magnetometer. I was wondering if the sensors and algorithms have advanced on that front.