I was wondering if we have any reference implementations or packages that implement (convex) optimization based kalman filters.
Such as discussed in this paper by Boyd: Real-Time Convex Optimization in Signal Processing
Not that I’m aware of, but I happen to be working on a very related problem at the moment, for which it would be trivial to incorporate, e.g., the robust measurement update in eq 6. Give it a day or so and it might be available in GitHub - baggepinnen/LowLevelParticleFilters.jl: State estimation, smoothing and parameter estimation using Kalman and particle filters.
1 Like
That’s awesome! I’ll be watching it 
It turns out that I found a closed-form solution to what I was trying to do, so I do not need the convex optimization anymore 
Oops. What are you actually trying to do?
It’s all in the link I posted, in particular the PR title
I read the new thermal_nn_example, I get it now. If nothing else exists I think I will explore writing a package for generic optimization based kalman filters. In terms of computational efficiency, it will of course never be close to LowLevelParticleFilters.jl.