dragonpilot - 基於 openpilot 的開源駕駛輔助系統
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

290 lines
8.2 KiB

/*
* This file is part of ACADO Toolkit.
*
* ACADO Toolkit -- A Toolkit for Automatic Control and Dynamic Optimization.
* Copyright (C) 2008-2014 by Boris Houska, Hans Joachim Ferreau,
* Milan Vukov, Rien Quirynen, KU Leuven.
* Developed within the Optimization in Engineering Center (OPTEC)
* under supervision of Moritz Diehl. All rights reserved.
*
* ACADO Toolkit is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 3 of the License, or (at your option) any later version.
*
* ACADO Toolkit is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with ACADO Toolkit; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*
*/
/**
* \file include/acado/nlp_solver/scp_method.hpp
* \author Boris Houska, Hans Joachim Ferreau
*
*/
#ifndef ACADO_TOOLKIT_SCP_METHOD_HPP
#define ACADO_TOOLKIT_SCP_METHOD_HPP
#include <acado/utils/acado_utils.hpp>
#include <acado/function/ocp_iterate.hpp>
#include <acado/nlp_solver/nlp_solver.hpp>
#include <acado/conic_solver/dense_qp_solver.hpp>
#include <acado/conic_solver/banded_cp_solver.hpp>
#include <acado/conic_solver/condensing_based_cp_solver.hpp>
#include <acado/nlp_solver/scp_evaluation.hpp>
#include <acado/nlp_solver/scp_step_linesearch.hpp>
#include <acado/nlp_solver/scp_step_fullstep.hpp>
#include <acado/nlp_derivative_approximation/nlp_derivative_approximation.hpp>
BEGIN_NAMESPACE_ACADO
/**
* \brief Implements different sequential convex programming methods for solving NLPs.
*
* \ingroup NumericalAlgorithms
*
* The class SCPmethod implements different sequential convex programming methods
* for solving nonlinear programming problems.
*
* \author Boris Houska, Hans Joachim Ferreau
*/
class SCPmethod : public NLPsolver
{
//
// PUBLIC MEMBER FUNCTIONS:
//
public:
/** Default constructor. */
SCPmethod( );
/** Default constructor. */
SCPmethod( UserInteraction* _userInteraction,
const Objective *objective_ ,
const DynamicDiscretization *dynamic_discretization_,
const Constraint *constraint_,
BooleanType _isCP = BT_FALSE
);
/** Copy constructor (deep copy). */
SCPmethod( const SCPmethod& rhs );
/** Destructor. */
virtual ~SCPmethod( );
/** Assignment operator (deep copy). */
SCPmethod& operator=( const SCPmethod& rhs );
virtual NLPsolver* clone() const;
/** Initialization. */
virtual returnValue init( VariablesGrid* x_init ,
VariablesGrid* xa_init,
VariablesGrid* p_init ,
VariablesGrid* u_init ,
VariablesGrid* w_init );
/** Solves current, possibly parametric, optimization problem. */
virtual returnValue solve( const DVector &x0_ = emptyConstVector,
const DVector &p_ = emptyConstVector
);
/** Executes a complete real-time step. */
virtual returnValue step( const DVector &x0_ = emptyConstVector,
const DVector &p_ = emptyConstVector
);
/** Executes a real-time feedback step */
virtual returnValue feedbackStep( const DVector &x0_,
const DVector &p_ = emptyConstVector
);
virtual returnValue performCurrentStep( );
/** Executes a real-time preparation step */
virtual returnValue prepareNextStep( );
/** Sets the reference to be used in the LSQ tracking terms. If the objective \n
* has also non-LSQ terms or no LSQ terms, an error message will be returned \n
* (cf. objective.hpp). \n
* This routine has been designed for real-time applications where the reference \n
* is explicitly time-dependent. \n
* \n
* \return SUCCESSFUL_RETURN \n
*/
virtual returnValue setReference( const VariablesGrid &ref
);
// virtual returnValue enableNeedToReevaluate( );
/** Shifts the data for preparating the next real-time step.
*
* \return RET_NOT_YET_IMPLEMENTED
*/
virtual returnValue shiftVariables( double timeShift = -1.0,
DVector lastX = emptyVector,
DVector lastXA = emptyVector,
DVector lastP = emptyVector,
DVector lastU = emptyVector,
DVector lastW = emptyVector );
/** Returns a variance-covariance estimate if possible or an error message otherwise.
*
* \return SUCCESSFUL_RETURN
* RET_MEMBER_NOT_INITIALISED
*/
virtual returnValue getVarianceCovariance( DMatrix &var );
/** Prints the run-time profile. This routine \n
* can be used after an integration run in \n
* order to assess the performance. \n
*/
virtual returnValue printRuntimeProfile() const;
//
// PROTECTED MEMBER FUNCTIONS:
//
protected:
virtual returnValue setupLogging( );
returnValue setup( );
/** Prints the actual values of x, xa, p, u, and w.
*/
returnValue printIterate( ) const;
returnValue printIteration( );
returnValue checkForConvergence( );
returnValue computeHessianMatrix( const BlockMatrix& oldLagrangeGradient,
const BlockMatrix& newLagrangeGradient
);
returnValue initializeHessianProjection( );
returnValue checkForRealTimeMode( const DVector &x0_,
const DVector &p_
);
returnValue setupRealTimeParameters( const DVector &x0_ = emptyConstVector,
const DVector &p_ = emptyConstVector
);
returnValue stopClockAndPrintRuntimeProfile( );
virtual returnValue getDifferentialStates( VariablesGrid &xd_ ) const;
virtual returnValue getAlgebraicStates ( VariablesGrid &xa_ ) const;
virtual returnValue getParameters ( VariablesGrid &p_ ) const;
virtual returnValue getParameters ( DVector &p_ ) const;
virtual returnValue getControls ( VariablesGrid &u_ ) const;
virtual returnValue getFirstControl ( DVector &u0_ ) const;
virtual returnValue getDisturbances ( VariablesGrid &w_ ) const;
virtual double getObjectiveValue ( ) const;
virtual returnValue getSensitivitiesX( BlockMatrix& _sens
) const;
virtual returnValue getSensitivitiesXA( BlockMatrix& _sens
) const;
virtual returnValue getSensitivitiesP( BlockMatrix& _sens
) const;
virtual returnValue getSensitivitiesU( BlockMatrix& _sens
) const;
virtual returnValue getSensitivitiesW( BlockMatrix& _sens
) const;
virtual returnValue getAnySensitivities( BlockMatrix& _sens,
uint idx
) const;
inline uint getNumPoints( ) const;
inline uint getNX( ) const;
inline uint getNXA( ) const;
inline uint getNP( ) const;
inline uint getNU( ) const;
inline uint getNW( ) const;
inline uint getNC( ) const;
//
// DATA MEMBERS:
//
protected:
int timeLoggingIdx;
RealClock clock;
RealClock clockTotalTime;
OCPiterate iter;
OCPiterate oldIter;
SCPevaluation* eval;
SCPstep* scpStep;
NLPderivativeApproximation* derivativeApproximation;
BandedCP bandedCP;
BandedCPsolver* bandedCPsolver;
BlockStatus status;
BooleanType isCP;
BooleanType hasPerformedStep;
BooleanType isInRealTimeMode;
BooleanType needToReevaluate;
};
CLOSE_NAMESPACE_ACADO
#include <acado/nlp_solver/scp_method.ipp>
#endif // ACADO_TOOLKIT_SCP_METHOD_HPP
/*
* end of file
*/