Reinforcement Learning Toolbox 2.0 last updated:

General Documentation Manual Tutorial Class Reference Master Thesis Examples Related Papers Downloads Links News mailto:webmaster Main Page     Class Hierarchy   Compound List   File List   Compound Members   File Members cagent.h Go to the documentation of this file. // Copyright (C) 2003 // Gerhard Neumann (gneumann@gmx.net) // Stephan Neumann (sneumann@gmx.net) // // This file is part of RL Toolbox. // http://www.igi.tugraz.at/ril_toolbox // // All rights reserved. // // Redistribution and use in source and binary forms, with or without // modification, are permitted provided that the following conditions // are met: // 1. Redistributions of source code must retain the above copyright // notice, this list of conditions and the following disclaimer. // 2. Redistributions in binary form must reproduce the above copyright // notice, this list of conditions and the following disclaimer in the // documentation and/or other materials provided with the distribution. // 3. The name of the author may not be used to endorse or promote products // derived from this software without specific prior written permission. // // THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. // IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, // INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT // NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF // THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #ifndef CAGENT_H #define CAGENT_H #include <list> #include "cagentcontroller.h" #include "caction.h" #include "cbaseobjects.h" #include "chierarchiccontroller.h" class CSemiMDPListener; class CEpisode; class CEnvironmentModel; class CStateCollection; class CStateCollectionImpl; class CSemiMDPSender { protected: std::list<CSemiMDPListener *> *SMDPListeners; public: CSemiMDPSender(); virtual ~CSemiMDPSender(); void addSemiMDPListener(CSemiMDPListener *listener); void removeSemiMDPListener(CSemiMDPListener *listener); bool isListenerAdded(CSemiMDPListener *listener); virtual void startNewEpisode(); virtual void sendNextStep(CStateCollection *lastState, CAction *Action, CStateCollection *currentState); virtual void sendIntermediateStep(CStateCollection *lastState, CAction *Action, CStateCollection *currentState); }; class CSemiMarkovDecisionProcess : public CDeterministicController, public CSemiMDPSender { protected: CAction *lastAction; int currentEpisodeNumber; int currentSteps; int totalSteps; bool isFirstStep; public: CSemiMarkovDecisionProcess(); ~CSemiMarkovDecisionProcess(); virtual void sendNextStep(CStateCollection *lastState, CAction *action, CStateCollection *currentState); CAction* getLastAction(); virtual void startNewEpisode(); int getCurrentEpisodeNumber() {return this->currentEpisodeNumber;}; int getCurrentStep() {return this->currentSteps;}; int getTotalSteps() {return this->totalSteps;}; virtual void addAction(CAction *action); virtual void addActions(CActionSet *action); }; class CHierarchicalSemiMarkovDecisionProcess : public CSemiMarkovDecisionProcess, public CHierarchicalStackListener, public CExtendedAction, public CStateModifiersObject { protected: virtual CAction *getExecutedAction(CHierarchicalStack *actionStack); CEpisode *currentEpisode; CStateCollectionImpl *pastState; CStateCollectionImpl *currentState; public: CHierarchicalSemiMarkovDecisionProcess(CEpisode *currentEpisode); CHierarchicalSemiMarkovDecisionProcess(CStateProperties *modelProperties, std::list<CStateModifier *> *modifiers = NULL); ~CHierarchicalSemiMarkovDecisionProcess(); virtual void setLoggedEpisode(CEpisode *loggedEpisode); virtual void nextStep(CStateCollection *oldState, CHierarchicalStack *actionStack, CStateCollection *newState); virtual void newEpisode(); virtual void sendNextStep(CAction *action); virtual bool isFinished(CStateCollection *, CStateCollection *) {return false;}; virtual CAction *getNextHierarchyLevel(CStateCollection *stateCollection, CActionDataSet *actionDataSet = NULL); virtual void addStateModifier(CStateModifier *modifier); virtual void removeStateModifier(CStateModifier *modifier); }; class CAgent : public CSemiMarkovDecisionProcess, public CStateModifiersObject { protected: CStateCollectionImpl *currentState; CStateCollectionImpl *lastState; int maxEpisodes; int maxSteps; bool keyboardBreaks; CEnvironmentModel *model; bool bLogEpisode; int doRun(bool bContinue); CEpisode *currentEpisode; public: CAgent(CEnvironmentModel *model); ~CAgent(); virtual void doAction(CAction *action); virtual void addStateModifier(CStateModifier *modifier); virtual void removeStateModifier(CStateModifier *modifier); int doControllerEpisode(int maxEpisodes = 1, int maxSteps = 5000); void setParameters(int maxEpisodes, int maxSteps); int doResume(); virtual void startNewEpisode(); void doControllerStep(); void setKeyboardBreak(bool keyboardBreak); bool getKeyboardBreak(); virtual void addAction(CPrimitiveAction *action); virtual void setLogEpisode(bool bLogEpisode); virtual CEpisode *getCurrentEpisode(); virtual CStateCollection *getCurrentState(); CEnvironmentModel *getEnvironmentModel(); }; class CHiearchicalAgent : public CAgent { protected: CAgent *realAgent; CHierarchicalSemiMarkovDecisionProcess *hierarchicSMDP; public: CHiearchicalAgent(CAgent *agent, CHierarchicalSemiMarkovDecisionProcess *hierarchicSMDP); virtual ~CHiearchicalAgent(); virtual void doAction(CAction *action); virtual void startNewEpisode(); }; #endif