Linux 35.5 + JetPack v5.1.3@ego-planner-swarm编译安装

Linux 35.5 + JetPack v5.1.3@ego-planner-swarm编译安装

  • [1. 源由](#1. 源由)
  • [2. 编译&安装](#2. 编译&安装)
    • [Step 1:依赖库安装](#Step 1:依赖库安装)
    • [Step 2:建立工程](#Step 2:建立工程)
    • [Step 3:编译工程](#Step 3:编译工程)
    • [Step 4:安装工程](#Step 4:安装工程)
  • [3. 问题汇总](#3. 问题汇总)
    • [3.1 组件ros-noetic-roslint问题](#3.1 组件ros-noetic-roslint问题)
    • [3.2 uav_simulator/local_sensing - CUDA优化](#3.2 uav_simulator/local_sensing - CUDA优化)
    • [3.3 编译命令`catkin_make`问题](#3.3 编译命令catkin_make问题)
  • [4. 总结](#4. 总结)

1. 源由

ego-planner是一个单机版本,而ego-planner-swarm是一个多级飞行的版本,因此在此一起学习下,以便更好的理解该系列开源项目的整个进展和框架。

2. 编译&安装

  • 确认本地ROS版本:
bash 复制代码
$ rosversion --distro
noetic
  • 本次编译代码版本:
bash 复制代码
$ git log -n 1
commit 30de3c9e11e6a2e06d81561f9364f07ceba6774b (HEAD -> master, origin/master, origin/HEAD)
Author: Xin Zhou <923137104@qq.com>
Date:   Thu Mar 16 21:00:03 2023 +0800

    Update README.md

Step 1:依赖库安装

bash 复制代码
$ sudo apt-get install libarmadillo-dev

Step 2:建立工程

bash 复制代码
$ mkdir -p ${YOUR_WORKSPACE_PATH}/src
$ cd ${YOUR_WORKSPACE_PATH}/src
$ git clone git@github.com:ZJU-FAST-Lab/ego-planner-swarm.git

Step 3:编译工程

bash 复制代码
$ cd ego-planner-swarm
$ catkin_make -j1

或者

bash 复制代码
$ cd ego-planner
$ catkin_make -DCMAKE_BUILD_TYPE=Release -j1

Step 4:安装工程

略,因为暂时不涉及软件的部署应用,相关模拟命令参考:

bash 复制代码
$ source devel/setup.bash && roslaunch ego_planner simple_run.launch

或者

  1. 运行Rviz虚拟环境
bash 复制代码
$ source devel/setup.bash
$ roslaunch ego_planner rviz.launch
  1. 运行模拟程序
bash 复制代码
$ source devel/setup.bash
$ roslaunch ego_planner swarm.launch

3. 问题汇总

3.1 组件ros-noetic-roslint问题

bash 复制代码
-- +++ processing catkin package: 'drone_detect'
-- ==> add_subdirectory(planner/drone_detect)
-- Could NOT find roslint (missing: roslint_DIR)
-- Could not find the required component 'roslint'. The following CMake error indicates that you either need to install the package with the same name or change your environment so that it can be found.
CMake Error at /opt/ros/noetic/share/catkin/cmake/catkinConfig.cmake:83 (find_package):
  Could not find a package configuration file provided by "roslint" with any
  of the following names:

    roslintConfig.cmake
    roslint-config.cmake

  Add the installation prefix of "roslint" to CMAKE_PREFIX_PATH or set
  "roslint_DIR" to a directory containing one of the above files.  If
  "roslint" provides a separate development package or SDK, be sure it has
  been installed.
Call Stack (most recent call first):
  planner/drone_detect/CMakeLists.txt:12 (find_package)


-- Configuring incomplete, errors occurred!
See also "/home/daniel/ZJU-FAST-Lab/ego-planner-swarm/build/CMakeFiles/CMakeOutput.log".
See also "/home/daniel/ZJU-FAST-Lab/ego-planner-swarm/build/CMakeFiles/CMakeError.log".
Invoking "cmake" failed

**解决方案:**安装组件ros-noetic-roslint

bash 复制代码
$ sudo apt-get install ros-noetic-roslint

3.2 uav_simulator/local_sensing - CUDA优化

bash 复制代码
$ git diff src/uav_simulator/local_sensing/CMakeLists.txt
diff --git a/src/uav_simulator/local_sensing/CMakeLists.txt b/src/uav_simulator/local_sensing/CMakeLists.txt
index 13c9911..5362408 100644
--- a/src/uav_simulator/local_sensing/CMakeLists.txt
+++ b/src/uav_simulator/local_sensing/CMakeLists.txt
@@ -6,7 +6,7 @@ ADD_COMPILE_OPTIONS(-std=c++11 )
 ADD_COMPILE_OPTIONS(-std=c++14 )

 #set(ENABLE_CUDA false)
-set(ENABLE_CUDA false)
+set(ENABLE_CUDA true)

 if(ENABLE_CUDA)
   find_package(CUDA REQUIRED)
@@ -20,7 +20,8 @@ if(ENABLE_CUDA)
 #       -gencode arch=compute_52,code=sm_52;
 #       -gencode arch=compute_60,code=sm_60;
 #       -gencode arch=compute_61,code=sm_61;
-      -gencode arch=compute_75,code=sm_75;
+#       -gencode arch=compute_75,code=sm_75;
+        -gencode arch=compute_87,code=sm_87;
   )

   SET(CUDA_PROPAGATE_HOST_FLAGS OFF)

3.3 编译命令catkin_make问题

bash 复制代码
[ 56%] Building CXX object uav_simulator/Utils/multi_map_server/CMakeFiles/multi_map_visualization.dir/src/multi_map_visualization.cc.o
/home/daniel/ZJU-FAST-Lab/ego-planner-swarm/src/uav_simulator/Utils/multi_map_server/src/multi_map_visualization.cc:5:10: fatal error: multi_map_server/MultiOccupancyGrid.h: No such file or directory
    5 | #include <multi_map_server/MultiOccupancyGrid.h>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [uav_simulator/Utils/multi_map_server/CMakeFiles/multi_map_visualization.dir/build.make:63: uav_simulator/Utils/multi_map_server/CMakeFiles/multi_map_visualization.dir/src/multi_map_visualization.cc.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:4588: uav_simulator/Utils/multi_map_server/CMakeFiles/multi_map_visualization.dir/all] Error 2
make: *** [Makefile:141: all] Error 2
Invoking "make -j1" failed

**解决方案:**更换编译命令

bash 复制代码
$ catkin_make

4. 总结

本次顺利实现了Jetson Orin Nano上的模拟导航飞行,同时看到有模拟的飞机遇到障碍没有飞出来,具体问题待进一步的研究了。

Jetson Orin Nano | ego-planner-swarm simulation

相关推荐
渲染101专业云渲染2 分钟前
川翔云电脑优势总结
服务器·3d·电脑·blender·maya
夜幕龙10 分钟前
深度生成模型(二)——基本概念与数学建模
人工智能·深度学习·transformer
游王子15 分钟前
OpenCV(11):人脸检测、物体识别
人工智能·opencv·计算机视觉
山海青风16 分钟前
从零开始玩转TensorFlow:小明的机器学习故事 3
人工智能·机器学习·tensorflow
@心都17 分钟前
机器学习数学基础:35.效度
人工智能·机器学习
幻想趾于现实19 分钟前
傅里叶分析
人工智能
chian-ocean20 分钟前
Linux 文件缓冲区:高效数据访问的幕后推手
linux·运维·服务器
加油=^_^=21 分钟前
【Linux】进程优先级 | 进程调度(三)
linux·运维·服务器
春末的南方城市26 分钟前
VidSketch:具有扩散控制的手绘草图驱动视频生成
人工智能·深度学习·计算机视觉·aigc
橘子味的茶二31 分钟前
高级系统架构师--第十章:计算机网络
服务器·计算机网络·系统架构