ubuntu16.04复现点线结合的双目算法：PL-SLAM

一、代码地址：

代码官网：https://github.com/rubengooj/pl-slam
注释：最终成功运行的是20180406提交的版本，不是最新的版本。

二、安装过程

1、安装依赖：

1）OpenCV 3.x.x

http://opencv.org

2）Eigen3 (tested with 3.2.92)

http://eigen.tuxfamily.org

3）Boost

sudo apt-get install libboost-dev

4）g2o - General Graph Optimization

https://github.com/RainerKuemmerle/g2o.git

5）YAML (tested with 0.5.2)

sudo apt-get install libyaml-cpp-dev

6）stvo-pl

https://github.com/rubengooj/stvo-pl

7）mrpt

https://github.com/MRPT/mrpt/tree/0c3d605c3cbf5f2ffb8137089e43ebdae5a55de3

2、编译生成

Executing the file build.sh will configure and generate the line_descriptor and DBoW2 modules, uncompress the vocabulary files, and then will 
configure and generate the PL-SLAM library for which we generate: libplslam.so in the lib folder, and the application plslam_dataset that works 
with our dataset format (explained in the next section).
运行文件build.sh，安装生成line_descriptor和DBoW2模块，解压vocabulary文件，会安装生成PL-SLAM库，有lib文件夹下的libplslam.so文件，以及plslam_dataset可执行文件。
解压vocabulary文件
chmod a+x build.sh
./build.sh

3、遇到问题

1）运行./build.sh时出现：

CMake Error at CMakeLists.txt:12 (find_package):
  By not providing "FindG2O.cmake" in CMAKE_MODULE_PATH this project has
  asked CMake to find a package configuration file provided by "G2O", but
  CMake did not find one.

  Could not find a package configuration file provided by "G2O" with any of
  the following names:

    G2OConfig.cmake
    g2o-config.cmake

  Add the installation prefix of "G2O" to CMAKE_PREFIX_PATH or set "G2O_DIR"
  to a directory containing one of the above files.  If "G2O" provides a
  separate development package or SDK, be sure it has been installed.


-- Configuring incomplete, errors occurred!

解决：

更改CMakeLists中寻找FindG2O的语句

#list(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake_modules")
更改为
list(APPEND CMAKE_MODULE_PATH /home/lexiaoxia/soft/lib/g2o_160424/g2o/cmake_modules)

config/kitti.yaml
vocabulary_p          : "/home/lexiaoxia/Codes/20210401/PLSLAM/pl-slam-20180406/vocabulary/mapir_orb.yml"
vocabulary_l          : "/home/lexiaoxia/Codes/20210401/PLSLAM/pl-slam-20180406/vocabulary/mapir_lsd.yml"

src/slamconfig.cpp:
    vocabulary_p          = "/home/lexiaoxia/Codes/20210401/PLSLAM/pl-slam-20180406/vocabulary/mapir_orb.yml";
    vocabulary_l          = "/home/lexiaoxia/Codes/20210401/PLSLAM/pl-slam-20180406/vocabulary/mapir_lsd.yml";

 

2）

/usr/bin/ld: 找不到 -lg2o_ext_csparse
collect2: error: ld returned 1 exit status
CMakeFiles/plslam.dir/build.make:255: recipe for target '../lib/libplslam.so' failed
make[2]: *** [../lib/libplslam.so] Error 1
CMakeFiles/Makefile2:99: recipe for target 'CMakeFiles/plslam.dir/all' failed
make[1]: *** [CMakeFiles/plslam.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2

解决：

在/usr/local/lib中看到啦相应的文件，所以应该是未更新库文件导致失败
sudo ldconfig -v//更新重新载入刚安装的g2o库文件

4、运行介绍

1）数据集配置：

We employ an environment variable, ${DATASETS_DIR}, pointing the directory that contains our datasets. Each sequence from each dataset 
must contain in its root folder a file named dataset_params.yaml, that indicates at least the camera model and the subfolders with 
the left and right images. We provide dataset parameters files for several datasets and cameras with the format xxxx_params.yaml.

我们利用环境变量${DATASETS_DIR}，指向包含我们数据集的目录。数据集中的每一个序列都必须在它的根目录中包含一个文件dataset_params.yaml，
该文件表示至少相机模型和带有左右图像的子文件夹。我们也给一些数据集和相机提供了xxxx_params.yaml格式的数据集参数文件。

2）配置文件： 

For running SLAM we can load the default parameters file or employ the config_xxxx.yaml files provided for every dataset.
我们可以下载默认的参数文件或者利用给每个数据集提供的config_xxxx.yaml文件

3）使用说明：

SLAM Application
Usage: ./plslam_dataset <dataset_name> [options] Options: -c Config file -o Offset (number of frames to skip in the dataset directory -n Number of frames 
to process the sequence -s Parameter to skip s-1 frames (default 1)

A full command would be:

./plslam_dataset kitti/00 -c ../config/config_kitti.yaml -o 100 -s 2 -n 1000

where we are processing the sequence 00 from the KITTI dataset (in our dataset folders) with the custom config file, with an offset -c allowing to
skip the first 100 images, a parameter -s to consider only one every 2 images, and a parameter -n to only consider 1000 input pairs.

使用说明：
./plslam_dataset <dataset_name> [options] Options: -c Config file -o Offset (number of frames to skip in the dataset directory -n Number of frames 
to process the sequence -s Parameter to skip s-1 frames (default 1)
如：./plslam_dataset kitti/00 -c ../config/config_kitti.yaml -o 100 -s 2 -n 1000
我们运行的是数据集文件夹在的KITTI数据集，利用惯例的配置文件，参数-c允许跳过前100张图像的偏移，参数-s考虑每2张图像只考虑1个，使用参数-n只考虑1000对输入图像对。

4）实际操作：

设置环境变量

sudo gedit ~/.bashrc
添加：
export DATASETS_DIR=/home/lexiaoxia/Datasets/hjydataset/kitti/data_odometry_gray/dataset/sequences
在build文件夹内运行：
./plslam_dataset 00 -c ../config/config_kitti.yaml -o 100 -s 2 -n 1000

5、遇到问题：

1）

Initializing PL-SLAM.... ... done. 
[ INFO:0] Initialize OpenCL runtime...
[ INFO:7] Initialize OpenCL runtime...
beignet-opencl-icd: no supported GPU found, this is probably the wrong opencl-icd package for this hardware
(If you have multiple ICDs installed and OpenCL works, you can ignore this message)
beignet-opencl-icd: no supported GPU found, this is probably the wrong opencl-icd package for this hardware
(If you have multiple ICDs installed and OpenCL works, you can ignore this message)
plslam_dataset: /usr/local/include/eigen3/Eigen/src/Core/DenseStorage.h:128: Eigen::internal::plain_array<T, Size, MatrixOrArrayOptions, 32>::plain_array() [with T = double; int Size = 16; int MatrixOrArrayOptions = 0]: Assertion `(reinterpret_cast<size_t>(eigen_unaligned_array_assert_workaround_gcc47(array)) & (31)) == 0 && "this assertion is explained here: " "http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html" " **** READ THIS WEB PAGE !!! ****"' failed.
已放弃 (核心已转储)

解决：

sudo apt install ocl-icd-libopencl1
sudo apt install opencl-headers
sudo apt install clinfo
sudo apt install ocl-icd-opencl-dev

2）

Initializing PL-SLAM.... ... done. 
plslam_dataset: /usr/local/include/eigen3/Eigen/src/Core/DenseStorage.h:128: Eigen::internal::plain_array<T, Size, MatrixOrArrayOptions, 32>::plain_array() [with T = double; int Size = 16; int MatrixOrArrayOptions = 0]: Assertion `(reinterpret_cast<size_t>(eigen_unaligned_array_assert_workaround_gcc47(array)) & (31)) == 0 && "this assertion is explained here: " "http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html" " **** READ THIS WEB PAGE !!! ****"' failed.
已放弃 (核心已转储)

解决：重新下载早期的20180406版本

3）在下载的20180406的PLSLAM版本

Initializing PL-SLAM.... ... done. 
[ INFO:7] Initialize OpenCL runtime...
[ INFO:0] Initialize OpenCL runtime...
beignet-opencl-icd: no supported GPU found, this is probably the wrong opencl-icd package for this hardware
(If you have multiple ICDs installed and OpenCL works, you can ignore this message)
beignet-opencl-icd: no supported GPU found, this is probably the wrong opencl-icd package for this hardware
(If you have multiple ICDs installed and OpenCL works, you can ignore this message)
------------------------------------------   Frame #1   ----------------------------------------

VO Runtime: 86.4519

No new KF needed: 1	-32.6602 -32.6602 1.72525e-36	0.680963 0.279471 0

段错误 (核心已转储)

解决：

用带有参数的选项的命令运行

三、总结：

还是下载了早期版本运行成功，注意一些库文件的安装先后顺序。