Kalman Filters

Josh Leverette's picture

This evening I started looking over Kalman filters and trying to figure out how to apply those to my system. Hopefully I'll be able to get something demonstrable up and running in the next couple of days.


Be careful

Kalman Filters are a huge black hole in terms of time investment. You might as well use them, as they're the simplest thing that will work in terms of sensor fusion (unless you decide to go Bayesian Particle Filters). But do ask me or better yet some KF guru for help when you start finding problems; there's a lot of ways they can go wrong. In particular, getting the variance values right is tricky.

Josh Leverette's picture


I didn't check for new replies until I had posted my other post tonight, but I can definitely see how they might grow to be a huge time sink. I'm not at the point of determining the variance/covariance values yet. I may also look into Bayesian particle filters, since you've mentioned them. I'd like to have the most appropriate filter for the task at hand, but I wouldn't want to sink too much time into Kalman filters of there's something with similar effectiveness for this situation that could be implemented in less time.