jafar::ddf::Upd_lin_pred_model< PARAMS > Class Template Reference

This class derives from a Linear_invertable_predict_model such as x(k|k-1) = F(x(k-1|k-1)) + G(k)w(k) q(k) = state noise covariance, q(k) is covariance of w(k) G(k) = state noise coupling invF exists and PredictionEngineBase. More...

---

Detailed Description

template<typename PARAMS>
class jafar::ddf::Upd_lin_pred_model< PARAMS >

This class derives from a Linear_invertable_predict_model such as x(k|k-1) = F(x(k-1|k-1)) + G(k)w(k) q(k) = state noise covariance, q(k) is covariance of w(k) G(k) = state noise coupling invF exists and PredictionEngineBase.

This class provides functions for updating the prediction model (matrices such as Fk, Gk and q) and for predicting from k to k+1. This class cannot be copied

Author:

Pierre Lamon

Definition at line 34 of file upd_lin_pred.hpp.

#include <upd_lin_pred.hpp>

Inheritance diagram for jafar::ddf::Upd_lin_pred_model< PARAMS >:

List of all members.

Public Member Functions
	Upd_lin_pred_model (unsigned short sv_size, unsigned short q_size)
unsigned short	GetSvSize ()
unsigned short	GetqSize ()
void	SetNeedUpdate (NEED_UPD_TYPE what, bool needUpd)
void	SetNoUpdate ()
std::vector< bool > const &	GetNeedUpdate () const
bool	IsUpdated (NEED_UPD_TYPE what)
bool	IsPSD (bool use_mutex)
	Check if X or Y is PD.
void	PrintModel ()
void	Predict (time const &horizon)
	Predict state from k to k+1.
void	Init (InfoContainer const &inityY)
void	InitState (VEC const &x, MSYM const &X)
	Set the state vector and the covariance matrix (at time k) before calling Predict.
void	InitInformation (VEC const &y, MSYM const &Y)
	Set the information vector and information matrix (at time k) before calling Predict.
InfoContainer const &	GetPrediction ()
	Get the predicted state (k+1) the result of Predict (return x,X in the case of Kalman and y,Y in the case of information)
InfoContainer const &	GetInversePrediction ()
	Return the inverse of GetPrediction.
void	UpdatePredictionWith (InfoContainer const &info)
	In the case of an information filter the update is trivial (it is just a sum)
void	UpdateModel ()
void	PrintHistory ()
FSStorageBase< PARAMS > const &	RecallStoredState (unsigned int idx) const
void	ClearHistory ()
unsigned short	GetHistorySize () const
bool	InsertDelayedMeasure (InfoContainer const &ICont, time const &avail_time)
	Insert a measure that has been taken in the past (teturn true in case of success)
Protected Member Functions
virtual void	UpdateF (PARAMS &param)=0
virtual void	UpdateG (PARAMS &param)=0
virtual void	Updateq (PARAMS &param)=0
Private Member Functions
void	Init ()
void	StoreState (FS_STORE_TYPE fs_type)
void	StoreState (MSYM const &I, VEC const &i)
void	Constructor ()
Private Attributes
std::vector< bool >	m_need_update
Bayesian_filter::Information_scheme	m_info_engine
InfoContainer	m_yY
SpeedUpInfoStruct	m_sp
Bayesian_filter::Information_scheme::Linear_predict_byproducts	m_products

---

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

• modules/ddf/include/ddf/upd_lin_pred.hpp