kalman

README

Adaptive Kalman filtering in Golang

go get github.com/konimarti/kalman

Adaptive Kalman filtering with Rapid Ongoing Stochastic covariance Estimation (ROSE)

Usage for standard Kalman filtering

	// create matrices and vectors
	...

	// create Kalman filter
	filter := kalman.NewFilter(
		X, // initial state (n x 1)
		P, // initial process covariance (n x n)
		F, // prediction matrix (n x n)
		B, // control matrix (n x k)
		Q, // process model covariance matrix (n x n)
		H, // measurement matrix (l x n)
		R, // measurement errors (l x l)
	)

	// get measurement (l x 1) and control (k x 1) vectors
	..

	// apply filter
	filtered := filter.Apply(measurement, control)
}

Results with standard Kalman filter

See example here.

Results with Rapid Ongoing Stochasic covariance Estimation (ROSE) filter

See example here.

Credits

This software package has been developed for and is in production at Kalkfabrik Netstal.

Documentation

Index

• type Filter
  • func NewFilter(X *mat.VecDense, P, F, B, Q, H, R *mat.Dense) Filter
  • func NewRoseFilter(X *mat.VecDense, P, F, B, H, G *mat.Dense, gammaR, alphaR, alphaM float64) Filter

Types

type Filter

type Filter interface {
	Apply(z, ctrl mat.Vector) mat.Vector
	State() mat.Vector
}

Filter interface for using the Kalman filter

func NewFilter

func NewFilter(X *mat.VecDense, P, F, B, Q, H, R *mat.Dense) Filter

NewFilter returns a Kalman filter X: initial state P: initial covariance matrix F: prediction matrix B: control matrix Q: system noise covariance matrix H: scaling matrix for measurements R: measurement error matrix

func NewRoseFilter

func NewRoseFilter(X *mat.VecDense, P, F, B, H, G *mat.Dense, gammaR, alphaR, alphaM float64) Filter

NewRoseFilter returns a ROSE Kalman filter Rapid Ongoing Stochasic covariance Estimation (ROSE) Filter X: initial state P: initial covariance matrix F: prediction matrix B: control matrix H: scaling matrix for measurements G: discretized G matrix for system noise gammaR: Gain factor for measurement noise alphaR: Kalman gain for measurment covariance noise alphaM: Kalman gain for covariance M