OpenMapper.h

Go to the documentation of this file.

/*
 * Copyright (C) 2006-2011, SRI International (R)
 *
 * This program 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.
 *
 * This program 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 this program.  If not, see <http://www.gnu.org/licenses/>.
 */
 
#pragma once
 
#ifndef __OpenKarto_Mapper_h__
#define __OpenKarto_Mapper_h__
 
#ifdef USE_TBB
#include <tbb/mutex.h>
#include <tbb/parallel_for.h>
#include <tbb/blocked_range.h>
#include <tbb/blocked_range3d.h>
#endif
 
#include <OpenKarto/Event.h>
#include <OpenKarto/Pair.h>
#include <OpenKarto/Geometry.h>
#include <OpenKarto/StringHelper.h>
#include <OpenKarto/SensorData.h>
#include <OpenKarto/Grid.h>
#include <OpenKarto/GridIndexLookup.h>
#include <OpenKarto/Module.h>
#include <OpenKarto/OccupancyGrid.h>
#include <OpenKarto/TypeCasts.h>
 
namespace karto
{
 
 
 
  class MapperEventArguments : public EventArguments
  {
  public:
    MapperEventArguments(const String& rMessage)
      : m_Message(rMessage)
    {
    }
 
    virtual ~MapperEventArguments()
    {
    }
 
  public:
    const String& GetEventMessage() const
    {
      return m_Message;
    }
 
  private:
    String m_Message;
  };
 
#ifdef WIN32
  EXPORT_KARTO_EVENT(KARTO_EXPORT, MapperEventArguments)
#endif
 
 
  class EdgeLabel
  {
  public:
    EdgeLabel()
    {
    }
 
    virtual ~EdgeLabel()
    {
    }
  }; // EdgeLabel
 
 
  // Contains the requisite information for the "spring"
  // that links two scans together--the pose difference and the uncertainty
  // (represented by a covariance matrix).
  class LinkInfo : public EdgeLabel
  {
  public:
    LinkInfo(const Pose2& rPose1, const Pose2& rPose2, const Matrix3& rCovariance)
    {
      Update(rPose1, rPose2, rCovariance);
    }
 
    virtual ~LinkInfo()
    {
    }
 
  public:
    void Update(const Pose2& rPose1, const Pose2& rPose2, const Matrix3& rCovariance)
    {
      m_Pose1 = rPose1;
      m_Pose2 = rPose2;
 
      // transform second pose into the coordinate system of the first pose
      Transform transform(rPose1, Pose2());
      m_PoseDifference = transform.TransformPose(rPose2);
 
      // transform covariance into reference of first pose
      Matrix3 rotationMatrix;
      rotationMatrix.FromAxisAngle(0, 0, 1, -rPose1.GetHeading());
 
      m_Covariance = rotationMatrix * rCovariance * rotationMatrix.Transpose();
    }
 
    inline const Pose2& GetPose1()
    {
      return m_Pose1;
    }
 
    inline const Pose2& GetPose2()
    {
      return m_Pose2;
    }
 
    inline const Pose2& GetPoseDifference()
    {
      return m_PoseDifference;
    }
 
    inline const Matrix3& GetCovariance()
    {
      return m_Covariance;
    }
 
  private:
    Pose2 m_Pose1;
    Pose2 m_Pose2;
    Pose2 m_PoseDifference;
    Matrix3 m_Covariance;
  }; // LinkInfo
 
 
  template<typename T>
  class Edge;
 
  template<typename T>
  class Vertex
  {
    friend class Edge<T>;
 
  public:
    Vertex(T pObject)
      : m_pObject(pObject)
    {
    }
 
    virtual ~Vertex()
    {
    }
 
    inline const List<Edge<T>*>& GetEdges() const
    {
      return m_Edges;
    }
 
    inline T GetVertexObject() const
    {
      return m_pObject;
    }
 
    List<Vertex<T>*> GetAdjacentVertices() const
    {
      List<Vertex<T>*> vertices;
 
      karto_const_forEach(typename List<Edge<T>*>, &m_Edges)
      {
        Edge<T>* pEdge = *iter;
 
        // check both source and target because we have a undirected graph
        if (pEdge->GetSource() != this)
        {
          vertices.Add(pEdge->GetSource());
        }
 
        if (pEdge->GetTarget() != this)
        {
          vertices.Add(pEdge->GetTarget());
        }
      }
 
      return vertices;
    }
 
  private:
    inline void AddEdge(Edge<T>* pEdge)
    {
      m_Edges.Add(pEdge);
    }
 
    T m_pObject;
    List<Edge<T>*> m_Edges;
  }; // Vertex<T>
 
 
  template<typename T>
  class Edge
  {
  public:
    Edge(Vertex<T>* pSource, Vertex<T>* pTarget)
      : m_pSource(pSource)
      , m_pTarget(pTarget)
      , m_pLabel(NULL)
    {
      m_pSource->AddEdge(this);
      m_pTarget->AddEdge(this);
    }
 
    virtual ~Edge()
    {
      m_pSource = NULL;
      m_pTarget = NULL;
 
      if (m_pLabel != NULL)
      {
        delete m_pLabel;
        m_pLabel = NULL;
      }
    }
 
  public:
    inline Vertex<T>* GetSource() const
    {
      return m_pSource;
    }
 
    inline Vertex<T>* GetTarget() const
    {
      return m_pTarget;
    }
 
    inline EdgeLabel* GetLabel()
    {
      return m_pLabel;
    }
 
    inline void SetLabel(EdgeLabel* pLabel)
    {
      m_pLabel = pLabel;
    }
 
  private:
    Vertex<T>* m_pSource;
    Vertex<T>* m_pTarget;
    EdgeLabel* m_pLabel;
  }; // class Edge<T>
 
 
  template<typename T>
  class Visitor
  {
  public:
    virtual kt_bool Visit(Vertex<T>* pVertex) = 0;
  }; // Visitor<T>
 
 
  template<typename T>
  class Graph;
 
  template<typename T>
  class GraphTraversal
  {
  public:
    GraphTraversal(Graph<T>* pGraph)
      : m_pGraph(pGraph)
    {
    }
 
    virtual ~GraphTraversal()
    {
    }
 
  public:
    virtual List<T> Traverse(Vertex<T>* pStartVertex, Visitor<T>* pVisitor) = 0;
 
  protected:
    Graph<T>* m_pGraph;
  }; // GraphTraversal<T>
 
 
  template<typename T>
  class Graph
  {
  public:
    typedef List<Vertex<T>*> VertexList;
    
    typedef List<Edge<T>*> EdgeList;
 
  public:
    Graph()
    {
    }
 
    virtual ~Graph()
    {
      Clear();
    }
 
  public:
    inline void AddVertex(Vertex<T>* pVertex)
    {
      m_Vertices.Add(pVertex);
    }
 
    inline void AddEdge(Edge<T>* pEdge)
    {
      m_Edges.Add(pEdge);
    }
 
    void Clear()
    {
      karto_const_forEach(typename VertexList, &m_Vertices)
      {
        // delete each vertex
        delete *iter;
      }
 
      m_Vertices.Clear();
 
      karto_const_forEach(typename EdgeList, &m_Edges)
      {
        // delete each edge
        delete *iter;
      }
 
      m_Edges.Clear();
    }
 
    inline const EdgeList& GetEdges() const
    {
      return m_Edges;
    }
 
    inline const VertexList& GetVertices() const
    {
      return m_Vertices;
    }
 
  protected:
    VertexList m_Vertices;
    
    EdgeList m_Edges;    
  }; // Graph<T>
 
 
  class OpenMapper;
  class ScanMatcher;
 
  class KARTO_EXPORT MapperGraph : public Graph<LocalizedObjectPtr>
  {
  public:
    MapperGraph(OpenMapper* pOpenMapper, kt_double rangeThreshold);
    
    virtual ~MapperGraph();
    
  public:
    void AddVertex(LocalizedObject* pObject);
    
    Edge<LocalizedObjectPtr>* AddEdge(LocalizedObject* pSourceObject, LocalizedObject* pTargetObject, kt_bool& rIsNewEdge);
 
    void AddEdges(LocalizedObject* pObject);
    
    void AddEdges(LocalizedLaserScan* pScan, const Matrix3& rCovariance);
    
    kt_bool TryCloseLoop(LocalizedLaserScan* pScan, const Identifier& rSensorName);
    
    LocalizedLaserScanList FindNearLinkedScans(LocalizedLaserScan* pScan, kt_double maxDistance);
 
     LocalizedLaserScanList FindOverlappingScans(LocalizedLaserScan* pScan);
    
    inline ScanMatcher* GetLoopScanMatcher() const
    {
      return m_pLoopScanMatcher;
    }
 
    void LinkChainToScan(const LocalizedLaserScanList& rChain, LocalizedLaserScan* pScan, const Pose2& rMean, const Matrix3& rCovariance);
    
  private:
    inline Vertex<LocalizedObjectPtr>* GetVertex(LocalizedObject* pObject)
    {
      return m_Vertices[pObject->GetUniqueId()];
    }
        
    LocalizedLaserScan* GetClosestScanToPose(const LocalizedLaserScanList& rScans, const Pose2& rPose) const;
            
    void LinkObjects(LocalizedObject* pFromObject, LocalizedObject* pToObject, const Pose2& rMean, const Matrix3& rCovariance);
    
    void LinkNearChains(LocalizedLaserScan* pScan, Pose2List& rMeans, List<Matrix3>& rCovariances);
    
    List<LocalizedLaserScanList> FindNearChains(LocalizedLaserScan* pScan);
        
    Pose2 ComputeWeightedMean(const Pose2List& rMeans, const List<Matrix3>& rCovariances) const;
    
    LocalizedLaserScanList FindPossibleLoopClosure(LocalizedLaserScan* pScan, const Identifier& rSensorName, kt_int32u& rStartScanIndex);
    
    void CorrectPoses();
    
  private:
    OpenMapper* m_pOpenMapper;
    
    ScanMatcher* m_pLoopScanMatcher;    
    
    GraphTraversal<LocalizedObjectPtr>* m_pTraversal;    
  }; // MapperGraph
 
  
  class ScanSolver : public Referenced
  {
  public:
    typedef List<Pair<kt_int32s, Pose2> > IdPoseVector;
 
    ScanSolver()
    {
    }
 
  protected:
    //@cond EXCLUDE
    virtual ~ScanSolver()
    {
    }
    //@endcond
 
  public:
    virtual void Compute() = 0;
 
    virtual const IdPoseVector& GetCorrections() const = 0;
 
    virtual void AddNode(Vertex<LocalizedObjectPtr>* /*pVertex*/)
    {
    }
 
    virtual void RemoveNode(kt_int32s /*id*/)
    {
    }
 
    virtual void AddConstraint(Edge<LocalizedObjectPtr>* /*pEdge*/)
    {
    }
    
    virtual void RemoveConstraint(kt_int32s /*sourceId*/, kt_int32s /*targetId*/)
    {
    }
    
    virtual void Clear() {};
  }; // ScanSolver
 
  
  class CorrelationGrid : public Grid<kt_int8u>
  {    
  protected:
    //@cond EXCLUDE
    virtual ~CorrelationGrid()
    {
      delete [] m_pKernel;
    }
    //@endcond
    
  public:
    static CorrelationGrid* CreateGrid(kt_int32s width, kt_int32s height, kt_double resolution, kt_double smearDeviation)
    {
      assert(resolution != 0.0);
      
      // +1 in case of roundoff
      kt_int32u borderSize = GetHalfKernelSize(smearDeviation, resolution) + 1;
      
      CorrelationGrid* pGrid = new CorrelationGrid(width, height, borderSize, resolution, smearDeviation);
      
      return pGrid;
    }
        
    virtual kt_int32s GridIndex(const Vector2i& rGrid, kt_bool boundaryCheck = true) const
    {
      kt_int32s x = rGrid.GetX() + m_Roi.GetX();
      kt_int32s y = rGrid.GetY() + m_Roi.GetY();
      
      return Grid<kt_int8u>::GridIndex(Vector2i(x, y), boundaryCheck);
    }
    
    inline const Rectangle2<kt_int32s>& GetROI() const
    {
      return m_Roi;
    }
    
    inline void SetROI(const Rectangle2<kt_int32s>& roi)
    {
      m_Roi = roi;
    }
    
    inline void SmearPoint(const Vector2i& rGridPoint)
    {
      assert(m_pKernel != NULL);
      
      int gridIndex = GridIndex(rGridPoint);
      if (GetDataPointer()[gridIndex] != GridStates_Occupied)
      {
        return;
      }
      
      kt_int32s halfKernel = m_KernelSize / 2;
      
      // apply kernel
      for (kt_int32s j = -halfKernel; j <= halfKernel; j++)
      {
        kt_int8u* pGridAdr = GetDataPointer(Vector2i(rGridPoint.GetX(), rGridPoint.GetY() + j));
        
        kt_int32s kernelConstant = (halfKernel) + m_KernelSize * (j + halfKernel);
        
        // if a point is on the edge of the grid, there is no problem
        // with running over the edge of allowable memory, because
        // the grid has margins to compensate for the kernel size
        SmearInternal(halfKernel, kernelConstant, pGridAdr);
      }      
    }
    
  protected:
    CorrelationGrid(kt_int32u width, kt_int32u height, kt_int32u borderSize, kt_double resolution, kt_double smearDeviation)
      : Grid<kt_int8u>(width + borderSize * 2, height + borderSize * 2)
      , m_SmearDeviation(smearDeviation)
      , m_pKernel(NULL)
    {            
      GetCoordinateConverter()->SetScale(1.0 / resolution);
 
      // setup region of interest
      m_Roi = Rectangle2<kt_int32s>(borderSize, borderSize, width, height);
      
      // calculate kernel
      CalculateKernel();
    }
    
    virtual void CalculateKernel()
    {
      kt_double resolution = GetResolution();
 
      assert(resolution != 0.0);
      assert(m_SmearDeviation != 0.0);      
      
      // min and max distance deviation for smearing;
      // will smear for two standard deviations, so deviation must be at least 1/2 of the resolution
      const kt_double MIN_SMEAR_DISTANCE_DEVIATION = 0.5 * resolution;
      const kt_double MAX_SMEAR_DISTANCE_DEVIATION = 10 * resolution;
      
      // check if given value too small or too big
      if (!math::InRange(m_SmearDeviation, MIN_SMEAR_DISTANCE_DEVIATION, MAX_SMEAR_DISTANCE_DEVIATION))
      {
        StringBuilder error;
        error << "Mapper Error:  Smear deviation too small:  Must be between " << MIN_SMEAR_DISTANCE_DEVIATION << " and " << MAX_SMEAR_DISTANCE_DEVIATION;
        throw Exception(error.ToString());
      }
      
      // NOTE:  Currently assumes a two-dimensional kernel
      
      // +1 for center
      m_KernelSize = 2 * GetHalfKernelSize(m_SmearDeviation, resolution) + 1;
      
      // allocate kernel
      m_pKernel = new kt_int8u[m_KernelSize * m_KernelSize];
      if (m_pKernel == NULL)
      {
        throw Exception("Unable to allocate memory for kernel!");
      }
      
      // calculate kernel
      kt_int32s halfKernel = m_KernelSize / 2;
      for (kt_int32s i = -halfKernel; i <= halfKernel; i++)
      {
        for (kt_int32s j = -halfKernel; j <= halfKernel; j++)
        {
#ifdef WIN32
          kt_double distanceFromMean = _hypot(i * resolution, j * resolution);
#else
          kt_double distanceFromMean = hypot(i * resolution, j * resolution);
#endif
          kt_double z = exp(-0.5 * pow(distanceFromMean / m_SmearDeviation, 2));
          
          kt_int32u kernelValue = static_cast<kt_int32u>(math::Round(z * GridStates_Occupied));
          assert(math::IsUpTo(kernelValue, static_cast<kt_int32u>(255)));
          
          int kernelArrayIndex = (i + halfKernel) + m_KernelSize * (j + halfKernel);
          m_pKernel[kernelArrayIndex] = static_cast<kt_int8u>(kernelValue);
        }
      }
    }
    
    static kt_int32s GetHalfKernelSize(kt_double smearDeviation, kt_double resolution)
    {
      assert(resolution != 0.0);
      
      return static_cast<kt_int32s>(math::Round(2.0 * smearDeviation / resolution));
    }
    
  private:
    // \todo 1/5/2011: was this separated from SmearPoint in preparation for optimization?
    inline void SmearInternal(kt_int32s halfKernel, kt_int32s kernelConstant, kt_int8u* pGridAdr)
    {
      kt_int8u kernelValue;
      kt_int32s kernelArrayIndex;
      kt_int32s i;
      
      for (i = -halfKernel; i <= halfKernel; i++)
      {
        kernelArrayIndex = i + kernelConstant;
        
        kernelValue = m_pKernel[kernelArrayIndex];
        if (kernelValue > pGridAdr[i])
        {
          // kernel value is greater, so set it to kernel value
          pGridAdr[i] = kernelValue;
        }
      }
    }
    
    kt_double m_SmearDeviation;
    
    // Size of one side of the kernel
    kt_int32s m_KernelSize;
    
    // Cached kernel for smearing
    kt_int8u* m_pKernel;
    
    // Region of interest
    Rectangle2<kt_int32s> m_Roi;
  }; // CorrelationGrid
  
 
  class ScanMatcherGridSet : public Referenced
  {
  public:
    ScanMatcherGridSet(CorrelationGrid* pCorrelationGrid,
      Grid<kt_double>* pSearchSpaceProbs,
      GridIndexLookup<kt_int8u>* pGridLookup)
      : m_pCorrelationGrid(pCorrelationGrid)
      , m_pSearchSpaceProbs(pSearchSpaceProbs)
      , m_pGridLookup(pGridLookup)
    {
    }
 
    virtual ~ScanMatcherGridSet()
    {
      delete m_pGridLookup;
    }
 
    SmartPointer<CorrelationGrid> m_pCorrelationGrid;
    SmartPointer<Grid<kt_double> > m_pSearchSpaceProbs;
    GridIndexLookup<kt_int8u>* m_pGridLookup;
  };
 
  
  class ScanMatcherGridSetBank;
  
  class KARTO_EXPORT ScanMatcher
  {
  public:
    virtual ~ScanMatcher();
    
  public:
    static ScanMatcher* Create(OpenMapper* pOpenMapper, kt_double searchSize, kt_double resolution, kt_double smearDeviation, kt_double rangeThreshold);
    
    kt_double MatchScan(LocalizedLaserScan* pScan, const LocalizedLaserScanList& rBaseScans, Pose2& rMean, Matrix3& rCovariance,
                        kt_bool doPenalize = true, kt_bool doRefineMatch = true);
    
    kt_double CorrelateScan(ScanMatcherGridSet* pScanMatcherGridSet, LocalizedLaserScan* pScan, const Pose2& rSearchCenter, const Vector2d& rSearchSpaceOffset, const Vector2d& rSearchSpaceResolution,
                            kt_double searchAngleOffset, kt_double searchAngleResolution,       kt_bool doPenalize, Pose2& rMean, Matrix3& rCovariance, kt_bool doingFineMatch);
    
    static void ComputePositionalCovariance(Grid<kt_double>* pSearchSpaceProbs, const Pose2& rBestPose, kt_double bestResponse, const Pose2& rSearchCenter,
                                            const Vector2d& rSearchSpaceOffset, const Vector2d& rSearchSpaceResolution,
                                            kt_double searchAngleResolution, Matrix3& rCovariance);
    
    static void ComputeAngularCovariance(ScanMatcherGridSet* pScanMatcherGridSet, const Pose2& rBestPose, kt_double bestResponse, const Pose2& rSearchCenter,
                                         kt_double searchAngleOffset, kt_double searchAngleResolution, Matrix3& rCovariance);
 
    CorrelationGrid* GetCorrelationGrid() const;
    
    Grid<kt_double>* GetSearchGrid() const;
    
    static kt_double GetResponse(ScanMatcherGridSet* pScanMatcherGridSet, kt_int32u angleIndex, kt_int32s gridPositionIndex);    
    
  private:
    static void AddScans(CorrelationGrid* pCorrelationGrid, const LocalizedLaserScanList& rScans, const Vector2d& rViewPoint);
    static void AddScansNew(CorrelationGrid* pCorrelationGrid, const LocalizedLaserScanList& rScans, const Vector2d& rViewPoint);
    
    static void AddScan(CorrelationGrid* pCorrelationGrid, LocalizedLaserScan* pScan, const Vector2d& rViewPoint, kt_bool doSmear = true);
    static void AddScanNew(CorrelationGrid* pCorrelationGrid, const Vector2dList& rValidPoints, kt_bool doSmear = true);
    
    static Vector2dList FindValidPoints(LocalizedLaserScan* pScan, const Vector2d& rViewPoint);
    
  protected:
    ScanMatcher(OpenMapper* pOpenMapper)
      : m_pOpenMapper(pOpenMapper)
      , m_pScanMatcherGridSet(NULL)
      , m_pScanMatcherGridSetBank(NULL)
    {
    }
    
  private:
    OpenMapper* m_pOpenMapper;
    
    SmartPointer<ScanMatcherGridSet> m_pScanMatcherGridSet;
    ScanMatcherGridSetBank* m_pScanMatcherGridSetBank;
  }; // ScanMatcher
  
 
  class SensorDataManager;
  struct MapperSensorManagerPrivate;
  
  class KARTO_EXPORT MapperSensorManager
  {    
  public:
    MapperSensorManager(kt_int32u runningBufferMaximumSize, kt_double runningBufferMaximumDistance);
    
    virtual ~MapperSensorManager();
    
  public:
    void RegisterSensor(const Identifier& rSensorName);
    
    LocalizedObject* GetLocalizedObject(const Identifier& rSensorName, kt_int32s stateId);
    
    List<Identifier> GetSensorNames();
    
    LocalizedLaserScan* GetLastScan(const Identifier& rSensorName);
    
    void SetLastScan(LocalizedLaserScan* pScan);
    
    void ClearLastScan(const Identifier& rSensorName);
    
    LocalizedObject* GetLocalizedObject(kt_int32s uniqueId);
    
    void AddLocalizedObject(LocalizedObject* pObject);
    
    void AddRunningScan(LocalizedLaserScan* pScan);
    
    LocalizedLaserScanList& GetScans(const Identifier& rSensorName);
    
    kt_int32s GetScanIndex(LocalizedLaserScan* pScan);
    
    LocalizedLaserScanList& GetRunningScans(const Identifier& rSensorName);
    
    LocalizedLaserScanList GetAllScans();
    
    LocalizedObjectList GetAllObjects();
 
    void Clear();
    
  private:
    inline SensorDataManager* GetSensorDataManager(LocalizedObject* pObject)
    {
      return GetSensorDataManager(pObject->GetSensorIdentifier());
    }
    
    SensorDataManager* GetSensorDataManager(const Identifier& rSensorName);
    
  private:
    MapperSensorManagerPrivate* m_pMapperSensorManagerPrivate;    
  }; // MapperSensorManager
  
  
  class KARTO_EXPORT OpenMapper : public Module
  {
    friend class MapperGraph;
    friend class ScanMatcher;
 
  public:
    OpenMapper(kt_bool multiThreaded = true);
 
    OpenMapper(const char* pName, kt_bool multiThreaded = true);
 
  public:
    BasicEvent<MapperEventArguments> Message;
 
    BasicEvent<MapperEventArguments> PreLoopClosed;
 
    BasicEvent<MapperEventArguments> PostLoopClosed;
    
    BasicEvent<EventArguments> ScansUpdated;
 
  protected:
    //@cond EXCLUDE
    virtual ~OpenMapper();
    //@endcond
    
  public:
    inline kt_bool IsMultiThreaded()
    {
      return m_MultiThreaded;
    }
 
    void Initialize(kt_double rangeThreshold);
    
    void Reset();
 
    virtual kt_bool Process(Object* pObject);
    
    const LocalizedLaserScanList GetAllProcessedScans() const;
 
    const LocalizedObjectList GetAllProcessedObjects() const;
 
    ScanSolver* GetScanSolver() const;
 
    void SetScanSolver(ScanSolver* pSolver);
 
    virtual MapperGraph* GetGraph() const;
 
    ScanMatcher* GetSequentialScanMatcher() const;
 
    ScanMatcher* GetLoopScanMatcher() const;
    
    inline MapperSensorManager* GetMapperSensorManager() const
    {
      return m_pMapperSensorManager;
    }
    
    inline kt_bool TryCloseLoop(LocalizedLaserScan* pScan, const Identifier& rSensorName)
    {
      return m_pGraph->TryCloseLoop(pScan, rSensorName);
    }
    
  protected:
    virtual void ScanMatched(LocalizedLaserScan* pScan) {};
    
    virtual void ScanMatchingEnd(LocalizedLaserScan* pScan) {};
 
  private:
    void InitializeParameters();
 
    kt_bool HasMovedEnough(LocalizedLaserScan* pScan, LocalizedLaserScan* pLastScan) const;
 
  public:
    // fire information for listeners!!
 
    kt_bool IsInitialized()
    {
      return m_Initialized;
    }
 
  private:
    // restrict the following functions
    OpenMapper(const OpenMapper&);
    const OpenMapper& operator=(const OpenMapper&);
 
  protected:
    SmartPointer<ScanSolver> m_pScanSolver;
 
  private:
    kt_bool m_Initialized;
    kt_bool m_MultiThreaded;
 
    ScanMatcher* m_pSequentialScanMatcher;
 
    MapperSensorManager* m_pMapperSensorManager;
 
    MapperGraph* m_pGraph;
 
    // Parameters
    // NOTE: Maintain the descriptions to these parameters in the comment above the class!
 
    Parameter<kt_bool>* m_pUseScanMatching;
    Parameter<kt_bool>* m_pUseScanBarycenter;    
    Parameter<kt_double>* m_pMinimumTravelDistance;
    Parameter<kt_double>* m_pMinimumTravelHeading;
 
    // scan matcher parameters
    
    Parameter<kt_int32u>* m_pScanBufferSize;
    Parameter<kt_double>* m_pScanBufferMaximumScanDistance;
    Parameter<kt_bool>* m_pUseResponseExpansion;
 
    Parameter<kt_double>* m_pDistanceVariancePenalty;
    Parameter<kt_double>* m_pMinimumDistancePenalty;
    Parameter<kt_double>* m_pAngleVariancePenalty;
    Parameter<kt_double>* m_pMinimumAnglePenalty;
    
    Parameter<kt_double>* m_pLinkMatchMinimumResponseFine;
    Parameter<kt_double>* m_pLinkScanMaximumDistance;
 
    // correlation parameters
 
    Parameter<kt_double>* m_pCorrelationSearchSpaceDimension;
    Parameter<kt_double>* m_pCorrelationSearchSpaceResolution;
    Parameter<kt_double>* m_pCorrelationSearchSpaceSmearDeviation;
    Parameter<kt_double>* m_pCoarseSearchAngleOffset;
    Parameter<kt_double>* m_pFineSearchAngleOffset;
    Parameter<kt_double>* m_pCoarseAngleResolution;
 
    // loop correlation parameters
 
    Parameter<kt_double>* m_pLoopSearchSpaceDimension;
    Parameter<kt_double>* m_pLoopSearchSpaceResolution;
    Parameter<kt_double>* m_pLoopSearchSpaceSmearDeviation;
 
    // loop detection parameters
    
    Parameter<kt_double>* m_pLoopSearchMaximumDistance;    
    Parameter<kt_int32u>* m_pLoopMatchMinimumChainSize;    
    Parameter<kt_double>* m_pLoopMatchMaximumVarianceCoarse;    
    Parameter<kt_double>* m_pLoopMatchMinimumResponseCoarse;    
    Parameter<kt_double>* m_pLoopMatchMinimumResponseFine;
  };
 
}
 
#endif // __OpenKarto_Mapper_h__