00001 /* +---------------------------------------------------------------------------+ 00002 | The Mobile Robot Programming Toolkit (MRPT) C++ library | 00003 | | 00004 | http://mrpt.sourceforge.net/ | 00005 | | 00006 | Copyright (C) 2005-2010 University of Malaga | 00007 | | 00008 | This software was written by the Machine Perception and Intelligent | 00009 | Robotics Lab, University of Malaga (Spain). | 00010 | Contact: Jose-Luis Blanco <jlblanco@ctima.uma.es> | 00011 | | 00012 | This file is part of the MRPT project. | 00013 | | 00014 | MRPT is free software: you can redistribute it and/or modify | 00015 | it under the terms of the GNU General Public License as published by | 00016 | the Free Software Foundation, either version 3 of the License, or | 00017 | (at your option) any later version. | 00018 | | 00019 | MRPT is distributed in the hope that it will be useful, | 00020 | but WITHOUT ANY WARRANTY; without even the implied warranty of | 00021 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | 00022 | GNU General Public License for more details. | 00023 | | 00024 | You should have received a copy of the GNU General Public License | 00025 | along with MRPT. If not, see <http://www.gnu.org/licenses/>. | 00026 | | 00027 +---------------------------------------------------------------------------+ */ 00028 #ifndef CRangeBearingKFSLAM2D_H 00029 #define CRangeBearingKFSLAM2D_H 00030 00031 #include <mrpt/utils/CDebugOutputCapable.h> 00032 #include <mrpt/math/CVectorTemplate.h> 00033 #include <mrpt/math/CMatrixTemplateNumeric.h> 00034 #include <mrpt/utils/CConfigFileBase.h> 00035 #include <mrpt/utils/CLoadableOptions.h> 00036 #include <mrpt/opengl.h> 00037 #include <mrpt/bayes/CKalmanFilterCapable.h> 00038 00039 #include <mrpt/utils/safe_pointers.h> 00040 00041 #include <mrpt/slam/CSensoryFrame.h> 00042 #include <mrpt/slam/CActionCollection.h> 00043 #include <mrpt/slam/CObservationBearingRange.h> 00044 #include <mrpt/poses/CPoint2D.h> 00045 #include <mrpt/poses/CPosePDFGaussian.h> 00046 #include <mrpt/slam/CLandmark.h> 00047 #include <mrpt/slam/CSensFrameProbSequence.h> 00048 #include <mrpt/slam/CIncrementalMapPartitioner.h> 00049 #include <mrpt/slam/data_association.h> 00050 00051 namespace mrpt 00052 { 00053 namespace slam 00054 { 00055 using namespace mrpt::bayes; 00056 using namespace mrpt::poses; 00057 00058 /** An implementation of EKF-based SLAM with range-bearing sensors, odometry, and a 2D (+heading) robot pose, and 2D landmarks. 00059 * The main method is "processActionObservation" which processes pairs of action/observation. 00060 * 00061 * The following Wiki page describes an front-end application based on this class: 00062 * http://babel.isa.uma.es/mrpt/index.php/Application:2d-slam-demo 00063 * 00064 * \sa CRangeBearingKFSLAM 00065 */ 00066 class MRPTDLLIMPEXP CRangeBearingKFSLAM2D : 00067 public bayes::CKalmanFilterCapable<3 /* x y yaw */, 2 /* range yaw */, 2 /* x y */, 3 /* Ax Ay Ayaw */> 00068 // <size_t VEH_SIZE, size_t OBS_SIZE, size_t FEAT_SIZE, size_t ACT_SIZE, size typename kftype = double> 00069 { 00070 public: 00071 CRangeBearingKFSLAM2D( ); //!< Default constructor 00072 virtual ~CRangeBearingKFSLAM2D(); //!< Destructor 00073 void reset(); //!< Reset the state of the SLAM filter: The map is emptied and the robot put back to (0,0,0). 00074 00075 /** Process one new action and observations to update the map and robot pose estimate. See the description of the class at the top of this page. 00076 * \param action May contain odometry 00077 * \param SF The set of observations, must contain at least one CObservationBearingRange 00078 */ 00079 void processActionObservation( 00080 CActionCollectionPtr &action, 00081 CSensoryFramePtr &SF ); 00082 00083 /** Returns the complete mean and cov. 00084 * \param out_robotPose The mean & 3x3 covariance matrix of the robot 2D pose 00085 * \param out_landmarksPositions One entry for each of the M landmark positions (2D). 00086 * \param out_landmarkIDs Each element[index] (for indices of out_landmarksPositions) gives the corresponding landmark ID. 00087 * \param out_fullState The complete state vector (3+2M). 00088 * \param out_fullCovariance The full (3+2M)x(3+2M) covariance matrix of the filter. 00089 * \sa getCurrentRobotPose 00090 */ 00091 void getCurrentState( 00092 CPosePDFGaussian &out_robotPose, 00093 std::vector<TPoint2D> &out_landmarksPositions, 00094 std::map<unsigned int,CLandmark::TLandmarkID> &out_landmarkIDs, 00095 CVectorDouble &out_fullState, 00096 CMatrixDouble &out_fullCovariance 00097 ) const; 00098 00099 /** Returns the mean & 3x3 covariance matrix of the robot 2D pose. 00100 * \sa getCurrentState 00101 */ 00102 void getCurrentRobotPose( 00103 CPosePDFGaussian &out_robotPose ) const; 00104 00105 /** Returns a 3D representation of the landmarks in the map and the robot 3D position according to the current filter state. 00106 * \param out_objects 00107 */ 00108 void getAs3DObject( mrpt::opengl::CSetOfObjectsPtr &outObj ) const; 00109 00110 /** Load options from a ini-like file/text 00111 */ 00112 void loadOptions( const mrpt::utils::CConfigFileBase &ini ); 00113 00114 /** The options for the algorithm 00115 */ 00116 struct MRPTDLLIMPEXP TOptions : utils::CLoadableOptions 00117 { 00118 /** Default values 00119 */ 00120 TOptions(); 00121 00122 /** Load from a config file/text 00123 */ 00124 void loadFromConfigFile( 00125 const mrpt::utils::CConfigFileBase &source, 00126 const std::string §ion); 00127 00128 /** This method must display clearly all the contents of the structure in textual form, sending it to a CStream. 00129 */ 00130 void dumpToTextStream(CStream &out) const; 00131 00132 00133 vector_float stds_Q_no_odo; //!< A 3-length vector with the std. deviation of the transition model in (x,y,phi) used only when there is no odometry (if there is odo, its uncertainty values will be used instead); x y: In meters, phi: radians (but in degrees when loading from a configuration ini-file!) 00134 float std_sensor_range, std_sensor_yaw; //!< The std. deviation of the sensor (for the matrix R in the kalman filters), in meters and radians. 00135 float quantiles_3D_representation; //!< Default = 3 00136 bool create_simplemap; //!< Whether to fill m_SFs (default=false) 00137 00138 // Data association: 00139 TDataAssociationMethod data_assoc_method; 00140 TDataAssociationMetric data_assoc_metric; 00141 double data_assoc_IC_chi2_thres; //!< Threshold in [0,1] for the chi2square test for individual compatibility between predictions and observations (default: 0.99) 00142 00143 // Limits of the sensor: 00144 //double sensor_max_range; //!< Maximum range (meters) of the sensor, or 0 (default) if there's no limit. 00145 //double sensor_fov; //!< Field of view (rads) of the sensor; default is 2·PI. 00146 }; 00147 00148 TOptions options; //!< The options for the algorithm 00149 00150 00151 /** Save the current state of the filter (robot pose & map) to a MATLAB script which displays all the elements in 2D 00152 */ 00153 void saveMapAndPath2DRepresentationAsMATLABFile( 00154 const std::string &fil, 00155 float stdCount=3.0f, 00156 const std::string &styleLandmarks = std::string("b"), 00157 const std::string &stylePath = std::string("r"), 00158 const std::string &styleRobot = std::string("r") ) const; 00159 00160 00161 /** Information for data-association: 00162 * \sa getLastDataAssociation 00163 */ 00164 struct MRPTDLLIMPEXP TDataAssocInfo 00165 { 00166 TDataAssocInfo() : 00167 Y_pred_means(0,0), 00168 Y_pred_covs(0,0) 00169 { 00170 } 00171 00172 void clear() { 00173 results.clear(); 00174 predictions_IDs.clear(); 00175 newly_inserted_landmarks.clear(); 00176 } 00177 00178 // Predictions from the map: 00179 CMatrixTemplateNumeric<kftype> Y_pred_means,Y_pred_covs; 00180 mrpt::vector_size_t predictions_IDs; 00181 00182 /** Map from the 0-based index within the last observation and the landmark 0-based index in the map (the robot-map state vector) 00183 Only used for stats and so. */ 00184 std::map<size_t,size_t> newly_inserted_landmarks; 00185 00186 // DA results: 00187 TDataAssociationResults results; 00188 }; 00189 00190 /** Returns a read-only reference to the information on the last data-association */ 00191 const TDataAssocInfo & getLastDataAssociation() const { 00192 return m_last_data_association; 00193 } 00194 00195 protected: 00196 00197 /** @name Virtual methods for Kalman Filter implementation 00198 @{ 00199 */ 00200 00201 /** Must return the action vector u. 00202 * \param out_u The action vector which will be passed to OnTransitionModel 00203 */ 00204 void OnGetAction( KFArray_ACT &out_u ); 00205 00206 /** Implements the transition model \f$ \hat{x}_{k|k-1} = f( \hat{x}_{k-1|k-1}, u_k ) \f$ 00207 * \param in_u The vector returned by OnGetAction. 00208 * \param inout_x At input has \f[ \hat{x}_{k-1|k-1} \f] , at output must have \f$ \hat{x}_{k|k-1} \f$ . 00209 * \param out_skip Set this to true if for some reason you want to skip the prediction step (to do not modify either the vector or the covariance). Default:false 00210 */ 00211 void OnTransitionModel( 00212 const KFArray_ACT &in_u, 00213 KFArray_VEH &inout_x, 00214 bool &out_skipPrediction 00215 ); 00216 00217 /** Implements the transition Jacobian \f$ \frac{\partial f}{\partial x} \f$ 00218 * \param out_F Must return the Jacobian. 00219 * The returned matrix must be \f$V \times V\f$ with V being either the size of the whole state vector (for non-SLAM problems) or VEH_SIZE (for SLAM problems). 00220 */ 00221 void OnTransitionJacobian( KFMatrix_VxV &out_F ); 00222 00223 00224 /** Implements the transition noise covariance \f$ Q_k \f$ 00225 * \param out_Q Must return the covariance matrix. 00226 * The returned matrix must be of the same size than the jacobian from OnTransitionJacobian 00227 */ 00228 void OnTransitionNoise( KFMatrix_VxV &out_Q ); 00229 00230 /** This is called between the KF prediction step and the update step, and the application must return the observations and, when applicable, the data association between these observations and the current map. 00231 * 00232 * \param out_z N vectors, each for one "observation" of length OBS_SIZE, N being the number of "observations": how many observed landmarks for a map, or just one if not applicable. 00233 * \param out_data_association An empty vector or, where applicable, a vector where the i'th element corresponds to the position of the observation in the i'th row of out_z within the system state vector (in the range [0,getNumberOfLandmarksInTheMap()-1]), or -1 if it is a new map element and we want to insert it at the end of this KF iteration. 00234 * \param in_S The full covariance matrix of the observation predictions (i.e. the "innovation covariance matrix"). This is a M·O x M·O matrix with M=length of "in_lm_indices_in_S". 00235 * \param in_lm_indices_in_S The indices of the map landmarks (range [0,getNumberOfLandmarksInTheMap()-1]) that can be found in the matrix in_S. 00236 * 00237 * This method will be called just once for each complete KF iteration. 00238 * \note It is assumed that the observations are independent, i.e. there are NO cross-covariances between them. 00239 */ 00240 void OnGetObservationsAndDataAssociation( 00241 std::vector<KFArray_OBS> &out_z, 00242 vector_int &out_data_association, 00243 const vector<KFArray_OBS> &in_all_predictions, 00244 const KFMatrix &in_S, 00245 const vector_size_t &in_lm_indices_in_S, 00246 const KFMatrix_OxO &in_R 00247 ); 00248 00249 void OnObservationModel( 00250 const vector_size_t &idx_landmarks_to_predict, 00251 std::vector<KFArray_OBS> &out_predictions 00252 ); 00253 00254 /** Implements the observation Jacobians \f$ \frac{\partial h_i}{\partial x} \f$ and (when applicable) \f$ \frac{\partial h_i}{\partial y_i} \f$. 00255 * \param idx_landmark_to_predict The index of the landmark in the map whose prediction is expected as output. For non SLAM-like problems, this will be zero and the expected output is for the whole state vector. 00256 * \param Hx The output Jacobian \f$ \frac{\partial h_i}{\partial x} \f$. 00257 * \param Hy The output Jacobian \f$ \frac{\partial h_i}{\partial y_i} \f$. 00258 */ 00259 void OnObservationJacobians( 00260 const size_t &idx_landmark_to_predict, 00261 KFMatrix_OxV &Hx, 00262 KFMatrix_OxF &Hy 00263 ); 00264 00265 /** Computes A=A-B, which may need to be re-implemented depending on the topology of the individual scalar components (eg, angles). 00266 */ 00267 void OnSubstractObservationVectors(KFArray_OBS &A, const KFArray_OBS &B); 00268 00269 /** Return the observation NOISE covariance matrix, that is, the model of the Gaussian additive noise of the sensor. 00270 * \param out_R The noise covariance matrix. It might be non diagonal, but it'll usually be. 00271 */ 00272 void OnGetObservationNoise(KFMatrix_OxO &out_R); 00273 00274 /** This will be called before OnGetObservationsAndDataAssociation to allow the application to reduce the number of covariance landmark predictions to be made. 00275 * For example, features which are known to be "out of sight" shouldn't be added to the output list to speed up the calculations. 00276 * \param in_all_prediction_means The mean of each landmark predictions; the computation or not of the corresponding covariances is what we're trying to determined with this method. 00277 * \param out_LM_indices_to_predict The list of landmark indices in the map [0,getNumberOfLandmarksInTheMap()-1] that should be predicted. 00278 * \note This is not a pure virtual method, so it should be implemented only if desired. The default implementation returns a vector with all the landmarks in the map. 00279 * \sa OnGetObservations, OnDataAssociation 00280 */ 00281 void OnPreComputingPredictions( 00282 const vector<KFArray_OBS> &in_all_prediction_means, 00283 vector_size_t &out_LM_indices_to_predict ); 00284 00285 /** If applicable to the given problem, this method implements the inverse observation model needed to extend the "map" with a new "element". 00286 * \param in_z The observation vector whose inverse sensor model is to be computed. This is actually one of the vector<> returned by OnGetObservations(). 00287 * \param out_yn The F-length vector with the inverse observation model \f$ y_n=y(x,z_n) \f$. 00288 * \param out_dyn_dxv The \f$F \times V\f$ Jacobian of the inv. sensor model wrt the robot pose \f$ \frac{\partial y_n}{\partial x_v} \f$. 00289 * \param out_dyn_dhn The \f$F \times O\f$ Jacobian of the inv. sensor model wrt the observation vector \f$ \frac{\partial y_n}{\partial h_n} \f$. 00290 * 00291 * - O: OBS_SIZE 00292 * - V: VEH_SIZE 00293 * - F: FEAT_SIZE 00294 * 00295 * \note OnNewLandmarkAddedToMap will be also called after calling this method if a landmark is actually being added to the map. 00296 */ 00297 void OnInverseObservationModel( 00298 const KFArray_OBS & in_z, 00299 KFArray_FEAT & out_yn, 00300 KFMatrix_FxV & out_dyn_dxv, 00301 KFMatrix_FxO & out_dyn_dhn ); 00302 00303 /** If applicable to the given problem, do here any special handling of adding a new landmark to the map. 00304 * \param in_obsIndex The index of the observation whose inverse sensor is to be computed. It corresponds to the row in in_z where the observation can be found. 00305 * \param in_idxNewFeat The index that this new feature will have in the state vector (0:just after the vehicle state, 1: after that,...). Save this number so data association can be done according to these indices. 00306 * \sa OnInverseObservationModel 00307 */ 00308 void OnNewLandmarkAddedToMap( 00309 const size_t in_obsIdx, 00310 const size_t in_idxNewFeat ); 00311 00312 00313 /** This method is called after the prediction and after the update, to give the user an opportunity to normalize the state vector (eg, keep angles within -pi,pi range) if the application requires it. 00314 */ 00315 void OnNormalizeStateVector(); 00316 00317 /** @} 00318 */ 00319 00320 00321 void getLandmarkIDsFromIndexInStateVector(std::map<unsigned int,CLandmark::TLandmarkID> &out_id2index) const 00322 { 00323 out_id2index = m_IDs_inverse; 00324 } 00325 00326 protected: 00327 00328 /** Set up by processActionObservation 00329 */ 00330 CActionCollectionPtr m_action; 00331 00332 /** Set up by processActionObservation 00333 */ 00334 CSensoryFramePtr m_SF; 00335 00336 /** The mapping between landmark IDs and indexes in the Pkk cov. matrix: 00337 */ 00338 std::map<CLandmark::TLandmarkID,unsigned int> m_IDs; 00339 00340 /** The mapping between indexes in the Pkk cov. matrix and landmark IDs: 00341 */ 00342 std::map<unsigned int,CLandmark::TLandmarkID> m_IDs_inverse; 00343 00344 /** The sequence of all the observations and the robot path (kept for debugging, statistics,etc) 00345 */ 00346 CSensFrameProbSequence m_SFs; 00347 00348 TDataAssocInfo m_last_data_association; //!< Last data association 00349 00350 00351 }; // end class 00352 } // End of namespace 00353 } // End of namespace 00354 00355 #endif
| Page generated by Doxygen 1.6.2 for MRPT 0.8.1 SVN:exported at Mon Feb 15 22:01:07 UTC 2010 |