4 ms·
Perhaps to get the basics sure, but to go through all the edge cases and parameter tuning and failure modes will easily fill a small book. That's not even going
by elcritch 2y ago
Perhaps to get the basics sure, but to go through all the edge cases and parameter tuning and failure modes will easily fill a small book. That's not even going into modifying Kalman filters for non-linear systems.
- esafak 2y agoHere's a survey that covers all of Bayesian filtering in 60 pages, plus references. https://www.researchgate.net/publication/238689222 https://www.researchgate.net/publication/238689222
- elcritch 2y agoThat's a nice survey. However for kalman filters it doesn't give practical examples or the different techniques to estimate the parameters. It doesn't cover adaptive modifications to kalmans like you often want in robotics or other applications to handle non-process external events which kalmans don't handle well. Et cetera, there's lots of reasons book forms on a topic can be useful.