▎Loyal Wingman
Particle Filter
Sequential Monte Carlo estimator for nonlinear, non-Gaussian state estimation.
Definition
Particle filters represent a posterior distribution by a weighted sample set, propagated by the dynamics model and reweighted by each observation. They handle nonlinear, multi-modal estimation problems where Kalman variants fail and are commonly used for terrain-relative navigation and feature-based SLAM in GPS-denied conditions.
Reference attributes
- Family
- Sequential Monte Carlo
- Use cases
- SLAM, GPS-denied nav, vision tracking
Related terms
#estimation#fusion