關於 高翔“一起做RGBD-SLAM” 因 g2o版本更新出錯的問題

主要參考了這篇博文：

https://blog.csdn.net/robinhjwy/article/details/78084210

在第六講中，運行g2o的時候，報錯信息如下：

/home/robot/rgbd_slam/src/slam.cpp: In function ‘int main(int, char**)’:
/home/robot/rgbd_slam/src/slam.cpp:75:70: error: no matching function for call to ‘g2o::BlockSolver<g2o::BlockSolverTraits<6, 3> >::BlockSolver(SlamLinearSolver*&)’
     SlamBlockSolver* blockSolver = new SlamBlockSolver( linearSolver );
                                                                      ^
In file included from /usr/local/include/g2o/core/block_solver.h:199:0,
                 from /home/robot/rgbd_slam/src/slam.cpp:22:
/usr/local/include/g2o/core/block_solver.hpp:40:1: note: candidate: g2o::BlockSolver<Traits>::BlockSolver(std::unique_ptr<typename Traits::LinearSolverType>) [with Traits = g2o::BlockSolverTraits<6, 3>; typename Traits::LinearSolverType = g2o::LinearSolver<Eigen::Matrix<double, 6, 6, 0> >]
 BlockSolver<Traits>::BlockSolver(std::unique_ptr<LinearSolverType> linearSolver
 ^
/usr/local/include/g2o/core/block_solver.hpp:40:1: note:   no known conversion for argument 1 from ‘SlamLinearSolver* {aka g2o::LinearSolverEigen<Eigen::Matrix<double, 6, 6, 0> >*}’ to ‘std::unique_ptr<g2o::LinearSolver<Eigen::Matrix<double, 6, 6, 0> >, std::default_delete<g2o::LinearSolver<Eigen::Matrix<double, 6, 6, 0> > > >’
/home/robot/rgbd_slam/src/slam.cpp:76:104: error: no matching function for call to ‘g2o::OptimizationAlgorithmLevenberg::OptimizationAlgorithmLevenberg(SlamBlockSolver*&)’
 ithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg( blockSolver );
                                                                              ^
In file included from /home/robot/rgbd_slam/src/slam.cpp:29:0:
/usr/local/include/g2o/core/optimization_algorithm_levenberg.h:47:16: note: candidate: g2o::OptimizationAlgorithmLevenberg::OptimizationAlgorithmLevenberg(std::unique_ptr<g2o::Solver>)
       explicit OptimizationAlgorithmLevenberg(std::unique_ptr<Solver> solver);
                ^

根據報錯位置，找到slam.cpp程序中發現是g2o的初始化求解器有問題，在根據報錯信息下面提示的位置：

In file included from /usr/local/include/g2o/core/block_solver.h:199:0,

找到這個文件並定位到199行，發現是一個#include "block_solver.hpp"的頭文件，進入這個頭文件中，可以發現求解器的構造函數是這樣寫的

//我用的g2o版本中  求解器的構造函數
BlockSolver<Traits>::BlockSolver(std::unique_ptr<LinearSolverType> linearSolver)
    :   BlockSolverBase(),
        _linearSolver(std::move(linearSolver))
{
  // workspace
  _xSize=0;
  _numPoses=0;
  _numLandmarks=0;
  _sizePoses=0;
  _sizeLandmarks=0;
  _doSchur=true;
}

而根據高翔書中程序中發現，他用的g2o版本中，求解器的構造函數跟我的這個g2o版本中的是不一樣的

//高翔所用g2o版本
BlockSolver<Traits>::BlockSolver(LinearSolverType* linearSolver) :
  BlockSolverBase(),
  _linearSolver(linearSolver)

所以就可以發現問題所在，BlockSolver構造時需要的是std::unique_ptr類型的linearSolver參數。因此修改slam.cpp中的程序如下：

    //SlamLinearSolver* linearSolver = new SlamLinearSolver();
    std::unique_ptr<SlamLinearSolver> linearSolver ( new SlamLinearSolver());
    linearSolver->setBlockOrdering( false );
    //SlamBlockSolver* blockSolver = new SlamBlockSolver( linearSolver );
    std::unique_ptr<SlamBlockSolver> blockSolver ( new SlamBlockSolver ( std::move(linearSolver)));
    g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg( std::move(blockSolver) );

再次編譯就OK啦