写在前边ecl是开源无人机项目PX4使用的算法库，使用ekf(扩展卡尔曼滤波)进行imu等多种传感器的数据融合

然而ecl不提供数据后处理功能

能使用ecl进行多传感器数据融合的后处理是很有必要的，这样便于参数调试，和二次开发的调试，从而缩短开发周期，降低开发成本

鉴于以上几点作者fork了ecl的项目，并添加了后处理功能

在添加后处理功能时的基本原则是，尽量做到不更改ecl的任何代码，实际上却是没有更改任何代码文件

实际上整个更改只是在CMakeLists.txt中添加了下边这一行

add_subdirectory(airdata)

add_subdirectory(EKF) #

add_subdirectory(geo)

add_subdirectory(geo_lookup)

add_subdirectory(main) # 这一行是新添加的可以看出一个叫做main的文件夹被添加到了和ekf以及test等的同级目录，这个文件夹中包含了生成后处理程序所需要的CMakeLists.txt和所有代码文件，而ekf则使用add_subdirectory(EKF) #这一行生成的动态库

入门

按照以下步骤可完成整个项目的克隆，生成，编译和运行。clone 项目到本地仓库

git clone https://github.com/akstuki/PX4-ECL.git

2.(这一步不是必须的)建议新建目录build，这样会比较干净

mkdir build/

cd build/

3. 生成项目(生成makefile文件)

$ cmake ..

-- The CXX compiler identification is GNU 7.4.0

-- Check for working CXX compiler: /usr/bin/c++.exe

-- Check for working CXX compiler: /usr/bin/c++.exe -- works

-- Detecting CXX compiler ABI info

-- Detecting CXX compiler ABI info - done

-- Detecting CXX compile features

-- Detecting CXX compile features - done

-- build type is RelWithDebInfo

-- PX4 ECL: Very lightweight Estimation & Control Library v1.9.0-rc1-496-g5ca644e

-- Configuring done

-- Generating done

-- Build files have been written to: /cygdrive/f/1-own/0-code/0alforithmCode/px4/ecl/PX4-ECL/build

4. 编译生成可运行文件

$ make

Scanning dependencies of target matrix

[ 2%] Completed 'matrix'

[ 21%] Built target matrix

Scanning dependencies of target prebuild_targets

[ 21%] Built target prebuild_targets

[ 26%] Built target ecl_airdata

[ 31%] Built target ecl_geo_lookup

[ 36%] Built target ecl_geo

[ 89%] Built target ecl_EKF

Scanning dependencies of target postEcl

[ 92%] Building CXX object main/CMakeFiles/postEcl.dir/main.cpp.o

[ 94%] Building CXX object main/CMakeFiles/postEcl.dir/csvparser.cpp.o

[ 97%] Building CXX object main/CMakeFiles/postEcl.dir/PostEkf.cpp.o

[100%] Linking CXX executable postEcl.exe

[100%] Built target postEcl

5. 运行程序(程序仅仅需要一个参数，就是px4的日志文件)

$ ./postEcl.exe 09_26_14_sensor_combined_0.csv

start ecl main...

47634158: reset position to last known position

47634158: reset velocity to zero

51113397: mag yaw fusion numerical error - covariance reset

51534173: EKF aligned, (baro hgt, IMU buf: 12, OBS buf: 9)

57718962: mag yaw fusion numerical error - covariance reset

baro hgt timeout - reset to baro

67032581: mag yaw fusion numerical error - covariance reset

68114987: reset position to last known position

68114987: reset velocity to zero

68114987: stopping navigation

baro hgt timeout - reset to baro

baro hgt timeout - reset to baro

84041372: reset position to last known position

84041372: reset velocity to zero

84041372: stopping navigation

finished.

6. 可以看到程序运行结束后生成一个输出文件ecloutput.csv，包含ekf的所有状态

$ ls -l

总用量 6848

-rwxrwxr-x+ 1 xiaoqiang xiaoqiang 1492459 9月 29 2018 09_26_14_sensor_combined_0.csv

-rwxrw-r--+ 1 xiaoqiang xiaoqiang 1025 1月 4 22:37 cmake_install.cmake

drwxrwxr-x+ 1 xiaoqiang xiaoqiang 0 1月 5 21:51 CMakeFiles

-rwxrw-r--+ 1 xiaoqiang xiaoqiang 2612997 1月 5 22:08 ecloutput.csv

-rwxrw-r--+ 1 xiaoqiang xiaoqiang 7687 1月 5 21:50 Makefile

-rwxrwxr-x+ 1 xiaoqiang xiaoqiang 2889190 1月 5 21:52 postEcl.exe

下一步工作添加更多数据格式的支持框架

添加参数配置接口 (或许以配置文件方式)

另外，可以将CMakeList.txt放到外层，把ecl以external的方式导入(本项目暂时不考虑这种方式)