Kalman< n, p, m, T > Class Template Reference

Templated linear kalman filter. More...

#include <kalman.hpp>

Collaboration diagram for Kalman< n, p, m, T >:

List of all members.

Public Member Functions
	Kalman (void)
	Default constructor.
	Kalman (Mat< n, 1 > x, Mat< n, n > P, Mat< n, n > F, Mat< n, n > Q, Mat< m, n > H, Mat< m, m > R, Mat< n, p > B=Mat< n, p >())
	Constructor.
const Mat< n, 1 > &	x (void) const
	Get current state.
const Mat< n, n > &	P (void) const
	Get current state covariance.
void	predict (void)
	Predict next state without input.
void	predict (Mat< p, 1, T > u)
	Get current state with input.
void	update (Mat< m, 1, T > z)
	Update.
template<uint32_t mm>
void	update (Mat< mm, 1, T > z, Mat< mm, n, T > H, Mat< mm, mm, T > R)
	Update using different measurement matrix and noise.
Protected Attributes
Mat< n, 1 >	x_
	State.
Mat< n, n >	P_
	State covariance.
Mat< n, n >	F_
	Process.
Mat< n, n >	Q_
	Process noise.
Mat< m, n >	H_
	Measurement.
Mat< m, m >	R_
	Measurement noise.
Mat< n, p >	B_
	Input.
Mat< m, m >	S_
	Innovation covariance.
const Mat< n, n >	I_
	Identity matrix.
Mat< n, m >	K_
	Kalman gain.
Mat< m, 1 >	y_
	Innovation.

---

Detailed Description

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>
class Kalman< n, p, m, T >

Templated linear kalman filter.

Template Parameters:

n	Size of state vector
p	Size of input vector
m	Size of measure vector
T	Type of data (default: float)

---

Constructor & Destructor Documentation

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>

Kalman< n, p, m, T >::Kalman	(	Mat< n, 1 >	x,
		Mat< n, n >	P,
		Mat< n, n >	F,
		Mat< n, n >	Q,
		Mat< m, n >	H,
		Mat< m, m >	R,
		Mat< n, p >	B = Mat<n,p>()
	)		[inline]

Constructor.

Parameters:

x	State vector (initial value)
P	State covariance (initial value)
Q	Process noise covariance
H	Measurement matrix
R	Measurement noise
B	Input matrix (default null)

---

Member Function Documentation

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>

const Mat<n,n>& Kalman< n, p, m, T >::P

(

void

)

const [inline]

Get current state covariance.

Returns:

Reference to state covariance (const)

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>

void Kalman< n, p, m, T >::update

(

Mat< m, 1, T >

z

)

[inline]

Update.

Parameters:

z	Measurement vector

Here is the caller graph for this function:

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>

template<uint32_t mm>

void Kalman< n, p, m, T >::update	(	Mat< mm, 1, T >	z,
		Mat< mm, n, T >	H,
		Mat< mm, mm, T >	R
	)		[inline]

Update using different measurement matrix and noise.

The measurement vector can be of any size

Parameters:

z	Measurement vector
H	Measurement matrix
R	Measurement noise

Template Parameters:

mm	Size of measurement vector

template<uint32_t n, uint32_t p, uint32_t m, typename T = float>

const Mat<n,1>& Kalman< n, p, m, T >::x

(

void

)

const [inline]

Get current state.

Returns:

Reference to state (const)

---

The documentation for this class was generated from the following file:

• /home/travis/build/lis-epfl/MAVRIC_Library/util/kalman.hpp