https://github.com/qoretechnologies/module-kalman

Qore Kalman filter module - Linear/Extended/Unscented Kalman filters via mherb/kalman
https://github.com/qoretechnologies/module-kalman

Last synced: about 1 month ago
JSON representation

Qore Kalman filter module - Linear/Extended/Unscented Kalman filters via mherb/kalman

• Host: GitHub
• URL: https://github.com/qoretechnologies/module-kalman
• Owner: qoretechnologies
• License: mit
• Created: 2026-04-22T14:35:56.000Z (3 months ago)
• Default Branch: develop
• Last Pushed: 2026-04-23T09:47:34.000Z (3 months ago)
• Last Synced: 2026-04-23T11:24:29.854Z (3 months ago)
• Language: C++
• Size: 51.8 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README
  • License: COPYING.MIT
  • Authors: AUTHORS

Awesome Lists containing this project

README

          
Qore Kalman module (kalman)

INTRODUCTION

------------

The Qore kalman module provides Kalman filter bindings on top of

Eigen (https://eigen.tuxfamily.org), the header-only C++ template

linear-algebra library. Kalman filters are the standard approach for

denoising / tracking / sensor-fusing noisy sequential observations —

their canonical applications are navigation (GPS/INS fusion, object

tracking), control systems, sensor processing in IoT and industrial

monitoring, and smoothing noisy time series in finance.

The module exposes three filter classes and a matrix primitive:

Classes provided:

    - Matrix:         columnar dense matrix (Eigen-backed); arithmetic,

                      transpose, inverse, cholesky, determinant.

    - LinearFilter:   standard Kalman filter (LKF) for linear systems.

    - ExtendedFilter: extended Kalman filter (EKF) for non-linear

                      dynamics / observations via Qore callbacks

                      returning Jacobians.

Pre-baked business-case helpers (1.1+):

    - ConstantVelocity2D / ConstantAcceleration2D / GpsSmoother —

      one-call construction of correctly-parameterised filters for

      common tracking problems.

Standalone functions:

    - kalman_version():   module version string.

    - get_kalman_info():  structured version info.

Thread safety: Matrix is immutable after construction (methods return

new matrices); filter objects hold mutable state and must not be

shared across threads without external locking.

DEPENDENCIES

------------

    - Eigen >= 3.3 (header-only; no runtime library)

BUILD

-----

    mkdir build && cd build

    cmake ..

    make

    sudo make install

Requires a Qore development installation (qore-config, qpp) and Eigen

development headers.

Matching distribution packages:

    - Ubuntu / Debian:  apt install libeigen3-dev

    - Alpine:           apk add eigen-dev

    - Fedora / RHEL:    dnf install eigen3-devel

    - macOS (Homebrew): brew install eigen