Self-paced robotics courses. Every course has 8 lessons, the equations, a worked example and a Python simulation you run yourself. The first lesson is free.
The algorithm that guided Apollo to the Moon and now runs in every drone, phone and rover. Learn where it came from, derive its equations and build a working tracker in Python.
Real robots turn, and turning is nonlinear. Learn how the EKF linearises motion and sensor models with Jacobians, and build a rover that localises itself from landmark ranges and bearings.
No Jacobians, better accuracy. Learn the sigma-point trick behind the UKF, compare it head-to-head with an EKF and build one that tracks a swinging pendulum.
When the world is not Gaussian, use a swarm of guesses. Learn sequential importance resampling, the 1993 bootstrap filter and how to beat particle degeneracy, then run the classic benchmark in Python.