Hi Guys - I'm searching for a C++ matrix math library suitable for Kalman filter implementation on MCUs. I found a few but they may be zombies, for example Eigen
How big are your matrices? If they are just 2x2, you can probably roll your own code as quickly as you can find and test something from the net. Templates are sometimes very useful and elegant, but if you have fixed known types and sizes, specific classes and functions are often easier to debug.
Certainly a fair bit bigger than 2x2, non-sparse. Eigen does seem to be active but don't know about use on MCUs. M7 with adequate RAM should be OK though. Using templates for decades during which diagnostics have improved ;-)
I'm unfamiliar with Eigen and only vaguely informed about Kalman filters, but doing eigenvalue calculation properly takes some know-how, and you're better off using a library rather than rolling your own. Does the MCU angle actually matter in relation to the library choice? There are a zillion linear algebra packages out there, the best known probably being LAPACK.
I have no idea if this is suitable for your requirements:
for current project... Eigen license is non-polluting.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required