Skip to content

[RA-L 2019] A Robust and Efficient Trajectory Planner for Quadrotors

License

Notifications You must be signed in to change notification settings

HuaYuXiao/Fast-Planner

 
 

Repository files navigation

Fast-Planner

Fast-Planner is developed aiming to enable quadrotor fast flight in complex unknown environments. It contains a rich set of carefully designed planning algorithms. Modified from dyn_planner

HitCount Static Badge Static Badge Static Badge Static Badge

Planning模块从fast-planner框架借鉴而来,可以实现无人机快速自主飞行。 框架前端kinodynamic路径搜索,后端采用基于样条的轨迹生成,同时还包含了时间调节系统。 Fast-planner可以在及其短的时间内(几毫秒)生成高质量轨迹(依赖增量式sdf地图构建,速度很快,但不是全局更新)。 这里由于建图工作(mapping模块)在别处完成,采用引入全局地图输入,利用sdf_tool生产全局sdf图,然后对系统做软约束优化。

参考文献
Robust and Efficient Quadrotor Trajectory Generation for Fast Autonomous Flight, Boyu Zhou, Fei Gao, Luqi Wang, Chuhao Liu and Shaojie Shen, IEEE Robotics and Automation Letters (RA-L), 2019.

Release Note

  • v1.1.0: upgrade sdf_map
  • v1.0.1:
    • remove flight_type
    • remove message_pub
    • remove load_map
    • remove pcdpubCallback
    • remove sim_mode

Installation

Before use, make sure you have installed following packages:

  • libboost: suggest version: 1.65.1
  • nlopt: NEVER install with apt install ros-noetic-nlopt!
catkin_make install --source src/Fast-Planner --build build/fast_planner

Launch

roslaunch fast_planner simulation.launch
  • 从rviz输入需要的期望goal, 选择3d navigation, 同时按下鼠标左右键,然后上下移动标记z大小.
  • 从rviz输入目标点信息,目标点高度不要为负值,x,y方向不要超出地图范围(地图参数在launch文件中设置)
  • 运行轨迹优化,加载离线地图,等待目标点输入.

img

About

[RA-L 2019] A Robust and Efficient Trajectory Planner for Quadrotors

Topics

Resources

License

Stars

Watchers

Forks

Languages

  • C++ 97.4%
  • Python 1.9%
  • CMake 0.7%