Home

ParticleFilters.jl provides a basic particle filter, along with some useful tools for constructing more complex particle filters. In particular it provides both weighted and unweighted particle belief types that implement the POMDPs.jl distribution interface including sampling and automatic caching of probability masses. Additionally, an important requirement for a particle filter is efficient resampling. This package provides O(n) sampling.

Dynamics and measurement models for the filters can be specified with a few functions or a POMDP. The simplest Bootstrap Particle filter can be constructed with BootstrapFilter. BasicParticleFilter provides a more flexible structure.

There are tutorials for three ways to use the particle filters:

1. As an estimator for feedback control,
2. to filter time-series measurements, and
3. as an updater for POMDPs.jl.

For documentation on all aspects of the package, see the contents below.

Contents

• Home
• • Contents
• Bootstrap Filter
• • Quick Start
  • More on the Bootstrap Filter
• Beliefs
• • Types
  • Interface
• Basic Particle Filter
• • Reference
• Handling Particle Depletion
• • Example
  • Additional Examples of Postprocessing Functions
• Sampling
• Example: Filtering Preexisting Data
• • Model
  • Data
  • Filtering
  • Plotting the Results
  • Appendix
• Example: Feedback Control
• • System Description
  • Control Law Description
  • Dynamics model
  • Observation Model
  • Particle Filter
  • Running a Simulation
• Example: Use with POMDPs.jl
• • Running a Simulation
  • Visualization