Advanced Kalman Filtering & Sensor Fusion
Building on the foundations of linear estimation, this course covers the nonlinear Kalman filter variants essential for real-world GNC and autonomous systems applications. Theory is developed from the ground up so you fully understand the implications of each design choice — then taken through to complete C++ implementations on practical problems.
The capstone project implements a full Unscented Kalman Filter running against a simulated autonomous vehicle sensor suite, as it would be deployed in a real self-driving car or UAV.
What You Will Learn
- How to use the Linear Kalman Filter to solve linear optimal estimation problems
- How to apply the Extended Kalman Filter (EKF) to nonlinear estimation problems
- How to apply the Unscented Kalman Filter (UKF) to nonlinear estimation problems
- How to fuse measurements from multiple sensors running at different update rates
- How to correctly initialise the Kalman Filter for robust operation
- How to model sensor errors inside the filter
- How to use fault detection to identify and reject bad sensor measurements
- How to tune the filter for best performance
- How to implement all three filter variants in C++
Capstone Project
Implement a full Unscented Kalman Filter for an autonomous vehicle sensor fusion problem, integrating GPS, IMU, and other sensor data at different rates — the same architecture used in real self-driving car and UAV navigation systems.
Who This Course Is For
- GNC and navigation engineers moving beyond linear estimation
- Engineers building autonomous vehicle, UAV, or spacecraft navigation systems
- Those who have completed Data Fusion with Linear Kalman Filter and want to go further
- Software engineers implementing sensor fusion in C++ for embedded or real-time systems
Prerequisites
A solid understanding of the Linear Kalman Filter and state space systems is assumed — the Data Fusion with Linear Kalman Filter course covers these prerequisites. Basic C++ programming knowledge is required for the implementation sections.
View Course on Udemy
A discount is available for this course — check back here once the mailing list is live.