0. 编译PX4固件

参考仿真平台基础配置教程（中文详细教程）

仿真平台基础配置 · 语雀 (yuque.com)https://www.yuque.com/xtdrone/manual_cn/basic_config

按照教程，成功编译后运行launch

一. 安装Ceres依赖：
1、软件包下载：
安装及下载地址：  Installation — Ceres Solverhttp://ceres-solver.org/installation.html

2、安装编译Ceres库所需依赖项：

$ sudo apt-get install liblapack-dev libsuitesparse-dev libcxsparse3.1.2 libgflags-dev libgoogle-glog-dev libgtest-dev

# CMake
sudo apt-get install cmake
# google-glog + gflags
sudo apt-get install libgoogle-glog-dev libgflags-dev
# Use ATLAS for BLAS & LAPACK
sudo apt-get install libatlas-base-dev
# Eigen3
sudo apt-get install libeigen3-dev
# SuiteSparse and CXSparse (optional)
sudo apt-get install libsuitesparse-dev
 可能会出现无法定位libcxsoarse3.1.2的问题

解决方法：

//第一步，打开sources.list

$ sudo gedit /etc/apt/sources.list

//第二步，将下面的源粘贴到最上方sources.list

$ deb http://cz.archive.ubuntu.com/ubuntu trusty main universe

//第三步，更新源

$ sudo apt-get update

//第四步，重新输入依赖项安装命令安装依赖项

$ sudo apt-get install liblapack-dev libsuitesparse-dev libcxsparse3.1.2 libgflags-dev libgoogle-glog-dev libgtest-dev

 3、编译安装：

$ mkdir build

$ cd build

$ cmake ..

$make -j4  //可用nproc这个命令查看cpu核心数，我的cpu核心数是8

4、测试
tar zxf ceres-solver-2.1.0.tar.gz
mkdir ceres-bin
cd ceres-bin
cmake ../ceres-solver-2.1.0
make -j3
make test
# Optionally install Ceres, it can also be exported using CMake which
# allows Ceres to be used without requiring installation, see the documentation
# for the EXPORT_BUILD_DIR option for more information.
make install

运行测试demo：

./simple_bundle_adjuster /home/haowang/workspace/ceres-solver/data/problem-16-22106-pre.t



This runs Ceres for a maximum of 10 iterations using the DENSE_SCHUR linear solver. The output should look something like this.
 
  iter      cost      cost_change  |gradient|   |step|    tr_ratio  tr_radius  ls_iter  iter_time  total_time
     0  4.185660e+06    0.00e+00    1.09e+08   0.00e+00   0.00e+00  1.00e+04       0    7.59e-02    3.37e-01
     1  1.062590e+05    4.08e+06    8.99e+06   5.36e+02   9.82e-01  3.00e+04       1    1.65e-01    5.03e-01
     2  4.992817e+04    5.63e+04    8.32e+06   3.19e+02   6.52e-01  3.09e+04       1    1.45e-01    6.48e-01
     3  1.899774e+04    3.09e+04    1.60e+06   1.24e+02   9.77e-01  9.26e+04       1    1.43e-01    7.92e-01
     4  1.808729e+04    9.10e+02    3.97e+05   6.39e+01   9.51e-01  2.78e+05       1    1.45e-01    9.36e-01
     5  1.803399e+04    5.33e+01    1.48e+04   1.23e+01   9.99e-01  8.33e+05       1    1.45e-01    1.08e+00
     6  1.803390e+04    9.02e-02    6.35e+01   8.00e-01   1.00e+00  2.50e+06       1    1.50e-01    1.23e+00
 
Solver Summary (v 2.1.0-eigen-(3.4.0)-lapack-suitesparse-(5.10.1)-cxsparse-(3.2.0)-acceleratesparse-eigensparse-no_openmp)
 
                                     Original                  Reduced
Parameter blocks                        22122                    22122
Parameters                              66462                    66462
Residual blocks                         83718                    83718
Residuals                              167436                   167436
 
Minimizer                        TRUST_REGION
 
Dense linear algebra library            EIGEN
Trust region strategy     LEVENBERG_MARQUARDT
 
                                        Given                     Used
Linear solver                     DENSE_SCHUR              DENSE_SCHUR
Threads                                     1                        1
Linear solver ordering              AUTOMATIC                 22106,16
Schur structure                         2,3,9                    2,3,9
 
Cost:
Initial                          4.185660e+06
Final                            1.803390e+04
Change                           4.167626e+06
 
Minimizer iterations                        7
Successful steps                            7
Unsuccessful steps                          0
 
Time (in seconds):
Preprocessor                         0.121654
 
  Residual only evaluation           0.065968 (7)
  Jacobian & residual evaluation     0.303356 (7)
  Linear solver                      0.436650 (7)
Minimizer                            0.890535
 
Postprocessor                        0.001684
Total                                1.013873
 
Termination:                      CONVERGENCE (Function tolerance reached. |cost_change|/cost: 1.769756e-09 <= 1.000000e-06)

5、install安装

sudo make install

二、 编译VINS-Fusion

源码地址：
GitHub - HKUST-Aerial-Robotics/VINS-Fusion: An optimization-based multi-sensor state estimator
An optimization-based multi-sensor state estimator - GitHub - HKUST-Aerial-Robotics/VINS-Fusion: An optimization-based multi-sensor state estimator
https://github.com/HKUST-Aerial-Robotics/VINS-Fusion

    cd ~/catkin_ws/src
    git clone https://github.com/HKUST-Aerial-Robotics/VINS-Fusion.git
    cd ../
    catkin_make
    source ~/catkin_ws/devel/setup.bash
VINS-Fusion仿真
注意把EKF设置为视觉定位，详见PX4飞控EKF配置教程。

启动仿真程序，注意launch文件中使用的是iris_stereo_camera.sdf

cd PX4_Firmware
roslaunch px4 indoor1.launch
启动VINS-Fusion，注意其默认订阅的话题是iris_0的，如果要换别的vehicle，需要对应修改。另外临时文件存储路径也要对应修改。

gedit ~/catkin_ws/src/VINS-Fusion/config/xtdrone_sitl/px4_sitl_stereo_imu_config.yaml
imu_topic: "/iris_0/imu_gazebo"
image0_topic: "/iris_0/stereo_camera/left/image_raw"
image1_topic: "/iris_0/stereo_camera/right/image_raw"
output_path: "/home/robin/catkin_ws/vins_output"
.
.
.
pose_graph_save_path: "/home/robin/catkin_ws/vins_output/pose_graph/"  # save and load path
cd catkin_ws
bash scripts/xtdrone_run_vio.sh
这时VINS-Fusion就启动了

由于VINS-Fusion发布的是Odometry类型的话题，我们要将其对应转为PX4所需的话题

cd ~/XTDrone/sensing/slam/vio
python vins_transfer.py iris 0

这时PX4仿真终端出现如下所示输出，代表视觉信息融合成功

INFO  [ecl/EKF] 1213644000: reset position to ev position
INFO  [ecl/EKF] 1213644000: commencing external vision position fusion
INFO  [ecl/EKF] 1213644000: commencing external vision yaw fusion
然后建立通信，键盘控制起飞即可

cd ~/XTDrone/communication
python multirotor_communication.py iris 0 

cd ~/XTDrone/control/keyboard
python multirotor_keyboard_control.py iris 1 vel



VIO的初始化
VINS-Fusion的双目+IMU可以在静止条件下完成初始化，因此不用额外考虑。但单目+IMU以及ORBSLAM3的单目/双目+IMU均需要在运动中初始化，在实物系统中，我们可以用手晃一晃飞机实现初始化，而仿真中必须起飞才能让无人机运动，但PX4的offboard模式如果没有定位数据，又无法起飞，这就造成了矛盾。因此仿真里只能用Gazebo真值给PX4定位，然后在飞行过程中初始化VIO，VIO的输出只用于测试精度，并不实际给无人机定位用。

原文链接：https://haowang.blog.csdn.net/article/details/124124570