Getting Started with Bayes in Engineering: Implementing Kalman Filters with RxInfer.jl
summary
Shows a concrete implementation of state-space Bayesian estimation using RxInfer.jl, building A/B/C matrices from structural mass/damping/stiffness, and running an augmented Kalman filter to estimate an unknown input and hidden states. Demonstrates a four-story building with three noisy accelerometers, plots posterior mean plus uncertainty ribbons, and explains how domain matrices are the hard part while the probabilistic code (priors, likelihood, inference) runs in Julia and produces usable state estimates.
AI-written summary · about
chapters (12)
- 0:00Bayesian methods intro
- 1:11Why Bayesian for engineers
- 2:59Modeling uncertainty
- 8:59Bayesian linear regression
- 10:02State space models
- 16:23Kalman filters
- 17:28Tracking building dynamics
- 20:58RxInfer.jl implementation
- 21:40Interpreting results
- 23:08Why go Bayesian
- 24:12Final thoughts
- 24:32Question and answer session
related talks
42mBayes is BAE
47mIPython and Sympy to Develop a Kalman Filter for Multisensor Data Fusion
42mBayesDB: Query the Probable Implications of Data
43mHow to look like a statistician (...)
42mSolving Real-World Business Problems with Bayesian Modeling
01h 10mBayesian Optimization and Machine Learning for Accelerating Experiments in the Physical Sciences
41mJAX for Bayes
30mProbabilistic modelling with Scala