Module ddf

---

Detailed Description

Version:

0.1

Author:

pierre.lamon@laas.fr

DDF stands as Decentralized Data Fusion. This module proposes a framework for fusing data in a decentralized fashion and handels asequent data.

History

• 0.1 (2005-11-30) - Decentralized data fusion with asequent data using IPC communication

Theory

The main concepts have been taken from the following publications:

• Nettleton E., Durrant-Whyte H.F, Abdallah S (2003) "A robust architecture for decentralised data fusion, International Conference on Advanced Robotics, Portugal"

Libraries

The main library for probabilistic data fusion is Bayes++. It can be downloaded from http://bayesclasses.sourceforge.net/Bayes++.html

Macro

Extra doc for macro can go here... (you can delete this section if not relevant)

Tcl interface (generated by swig)

Classes
class	jafar::ddf::ChannelFilterBase
	Implements a channel filter belonging to node i and connected to another channel filter belonging to node j. More...
class	jafar::ddf::CommBase
	This is the base class for channel filter communication. More...
class	jafar::ddf::CommFactoryBase
	The base class for the communication factories. More...
class	jafar::ddf::CommFactoryGeneric< T >
	The generic class for the communication factories. More...
class	jafar::ddf::time
	A toolbox for time management (absolute and duration). More...
class	jafar::ddf::DdfException
	Base class for all exceptions defined in the module ddf. More...
class	jafar::ddf::InfoContainer
	Class for manipulating information This class is a container class for communicating, storing information in a network of sensor nodes. More...
class	jafar::ddf::CommThreadFunc< PARENT >
	Class in charge of listening to a communication channel (continuously call the function Read) More...
class	jafar::ddf::IPCProcComm
	An IPC implementation of CommBase. More...
class	jafar::ddf::Timer_Command
	This is a class for controlling the execution of a thread (a nodal filter, a channel manager, etc) One can run/stop the execution of the thread, and kill the process (return from the main function) More...
class	jafar::ddf::Thread_Periodic_Func< T >
	This is the class (functionnal object) associated to a nodal thread. More...
class	jafar::ddf::PredictionEngineBase
	This is a virtual class for state prediction. More...
class	jafar::ddf::PredictionEngineGeneric< PARAMS >
	This class allows to customize the set of parameters needed for updating the prediction model (specialization of PredictionEngineBase) More...
class	jafar::ddf::PredictModelFactoryBase
	The factory for prediction models. More...
class	jafar::ddf::PredictModelFactoryGeneric< T >
	Specialized factory for prediction models. More...
class	jafar::ddf::SensorNodeBase
	This class implements the core functionalitires for a sensornode i.e. More...
class	jafar::ddf::SensorNodeGeneric< PARAMS >
	This class uses template for binding a generic parameter container to the SensorNodeBase class This parameter is used for updating the prediction model i.e. More...
class	jafar::ddf::ManagerThreadFunc< PARENT >
	The main loop for the channel manager. More...
class	jafar::ddf::SensorNodeIPC< PARAMS >
	Implement a sensor node communications with IPC's (message queues) create a thread that handles connections/deconnections between nodes. More...
class	jafar::ddf::StopWatch
	Measure the elapsed time. More...
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. More...
class	jafar::ddfsimu::CstAccPredictModelFactory
	The factory for constant acceleration models. More...