Crate embedded_kalman

source · [−]
Expand description

embedded-kalman

embedded-kalman is a Kalman filtering library written for Rust targeting:

  • Embedded or resource-constrained systems
  • High-reliability-systems where runtime failure is not an option

Although this library was designed with embedded systems in mind, there’s nothing wrong with using it in other applications. On the contrary, it should be quite fast on unconstrained systems. For now, embedded-kalman only features a standard optimal gain Kalman filter for linear time-invariant systems. Future versions are planned to contain:

  • Different variations such as the time-variant version of the regular Kalman Filter
  • Kalman filter varieties for non-linear systems such as the Extended Kalman Filter (EKF) and the Unscented Kalman Filter (UKF)

A quick example

ⓘ
use embedded_kalman::KalmanFilter;
use nalgebra::{SMatrix, SVector, Vector2};

let kf: KalmanFilter<f64, 2, 2, 2, _> = KalmanFilter::new(
    ... // matrices and initial state go here
);

let kf = kf
    .predict(Vector2::new(1.0, 0.0)) // Give an input to the system
    .update(Vector2::new(1.0, 0.0)) // Update the state from a given measurement
    .expect("Update step failed!");

let (x_posterior, P_posterior) = kf.get_posteriors(); // Get the posteriors

Features

  • Built on top of the beautiful nalgebra linear algebra library
  • embedded-kalman does not rely on the standard library and can be run in a #![no_std] environment
  • The filters do not require dynamic memory allocation, and can be run without an allocator
  • Plays nicely with embedded platforms such as the microcontrollers running on the ARM Cortex-M architecture
  • Uses Rust’s type-state programming to ensure correct usage of the filters in compile-time
  • Written in 100% safe Rust
  • Few direct dependencies

Non-features

  • Thus far, little effort has been made to optimize the code in this library
  • This library is pretty new and still not battle-tested. Please do not use this for anything important

Using embedded-kalman

You will need the last stable build of the rust compiler and the official package manager: cargo.

Simply add the following to your Cargo.toml file:

ⓘ
[dependencies]
embedded-kalman = "0.2.0"

Implemented Filters

Re-exports

pub use crate::covariance_matrix::CovarianceMatrix;

Modules

covariance_matrix
A module that defines a covariance matrix type for use in Kalman Filters

Structs

KalmanFilter
Kalman Filter

Traits

State
A marker trait to indicate what state the Kalman filter is currently in