Filter Class Template Reference

List of all members.


Detailed Description

template<class T = unsigned int, class P = indii::ml::aux::GaussianPdf>
class indii::ml::filter::Filter< T, P >

Abstract filter.

Author:
Lawrence Murray <lawrence@indii.org>
Version:
Rev
Date:
Date
Parameters:
T The type of time.
P The type of probability distribution used to represent the system state.
See also:
indii::ml::filter for general usage guidelines.

Definition at line 24 of file Filter.hpp.


Public Member Functions

 Filter (const P &p_x0)
 Constructor.
virtual ~Filter ()
 Destructor.
getTime () const
 Get the current time.
void setTime (const T tn)
 Set the current time.
P & getFilteredState ()
 Get filter density.
void setFilteredState (const P &p_xtn_ytn)
 Set the filter density.
virtual void filter (const T tnp1, const indii::ml::aux::vector &ytnp1)=0
 Advance system to time of next measurement.
virtual P measure ()=0
 Apply the measurement function to the current filtered state to obtain an estimated measurement.

Protected Attributes

tn
 $t_n$; the current time.
p_xtn_ytn
 $p(\mathbf{x}(t_n)\,|\,\mathbf{y}_{1:n}$; the filter density at the current time.

Constructor & Destructor Documentation

Filter ( const P &  p_x0  )  [inline]

Constructor.

Parameters:
p_x0 $p(\mathbf{x}_0)$; prior over the initial state $\mathbf{x}_0$.

Definition at line 107 of file Filter.hpp.

~Filter (  )  [inline, virtual]

Destructor.

Definition at line 112 of file Filter.hpp.


Member Function Documentation

T getTime (  )  const [inline]

Get the current time.

Returns:
$t_n$; the current time.

Definition at line 117 of file Filter.hpp.

void setTime ( const T  tn  )  [inline]

Set the current time.

Parameters:
tn $t_n$; the current time.

Definition at line 122 of file Filter.hpp.

P & getFilteredState (  )  [inline]

Get filter density.

Returns:
$p(\mathbf{x}(t_n)\,|\,\mathbf{y}_{1:n}$; the filter density at the current time.

Definition at line 127 of file Filter.hpp.

void setFilteredState ( const P &  p_xtn_ytn  )  [inline]

Set the filter density.

Returns:
$p(\mathbf{x}(t_n)\,|\,\mathbf{y}_{1:n}$; the filter density at the current time.

Definition at line 132 of file Filter.hpp.

virtual void filter ( const T  tnp1,
const indii::ml::aux::vector ytnp1 
) [pure virtual]

Advance system to time of next measurement.

Parameters:
tnp1 $t_{n+1}$; the time to which to advance the system. This must be greater than the current time $t_n$.
ytnp1 $\mathbf{y}_{n+1}$; measurement at time $t_{n+1}$.

Implemented in KalmanFilter, ParticleFilter, and UnscentedKalmanFilter.

virtual P measure (  )  [pure virtual]

Apply the measurement function to the current filtered state to obtain an estimated measurement.

Returns:
The estimated measurement.

Implemented in KalmanFilter, ParticleFilter, and UnscentedKalmanFilter.


Member Data Documentation

T tn [protected]

$t_n$; the current time.

For internal use only.

Definition at line 92 of file Filter.hpp.

P p_xtn_ytn [protected]

$p(\mathbf{x}(t_n)\,|\,\mathbf{y}_{1:n}$; the filter density at the current time.

Definition at line 98 of file Filter.hpp.


Generated on Wed Dec 17 15:11:58 2008 for dysii Dynamical Systems Library by  doxygen 1.5.3