recursivemake / kalman

kalman

An implementation of the Kalman filter based on https://home.wlu.edu/~levys/kalman_tutorial/

Installation

Download from https://github.com/recursiveMake/kalman

Usage

lein repl

;; Load matrix
(require '[clojure.core.matrix :as m])

(require '[kalman.system :refer [set-system-parameter, create-system]])

;; Create a system with parameters
(def start-state
  (set-system-parameter (create-system 1 1)
                        {:state (m/matrix [881])
                         :sensor-noise (m/matrix [[200]])
                         :state-predict (m/matrix [[0.75]])}))

;; Create observations
(def obs (map #(m/matrix [%]) [662 496 372 279 157 118 88 66]))

;; Run the filter
(kalman start-state obs)

;; Load sample data 
(require '[kalman.sample :as sample])
(def s-data (sample/data (sample/file-name)))

;; Create observables
(def obs (sample/observables s-data))

;; Create a system
(def system (sample/system (first obs)))

;; Run filter
(logging-kalman system (rest obs))

;; Visualize path
(require '[kalman.visual :as visual])
(def v-data (visual/data))

;; Plot raw data positions
(visual/v (visual/plot v-data))

License

Copyright © 2017 recursiveMake

Distributed under the Eclipse Public License either version 1.0 or (at your option) any later version.