What does particle filter mean?
An estimator that represents its belief as a cloud of guesses, called particles. Each tick it moves every particle with the motion model plus noise, weights each by how well it explains the measurement, and resamples, so it can hold a belief of any shape.
In a lesson
The particle filter's idea is blunt and it works: instead of describing the belief with a formula, keep a few hundred guesses and treat the collection as the belief.
Learn it
- Taught in U7.2 A cloud of guesses Localisation, University
- Guide with live demos: Particle filters explained
Used in
- U6.2 Predict and correct
- U6.7 Project: navigate on the estimate
- U7.1 Where am I?
- U8.5 Ray casting the map