visPlanner

visPlanner is a visibility-aware trajectory planning framework that can deal with visibility of the target in aerial tracking.

Authors: Qianhao Wang, Yuman Gao, Jialin Ji, Chao Xu and Fei Gao from the ZJU Fast Lab.

Paper: Visibility-aware Trajectory Optimization with Application to Aerial Tracking, Accepted in IEEE International Workshop on Intelligent Robots and Systems (IROS 2021).

Video Links: youtube or bilibili

[NOTE] remember to change the CUDA option of src/uav_simulator/local_sensing/CMakeLists.txt based on your GPU.

Run the visualization:

git clone https://github.com/ZJU-FAST-Lab/visPlanner.git
cd visPlanner
catkin_make
source devel/setup.zsh
roslaunch ego_planner rviz.launch

Run the visPlanner:

roslaunch ego_planner tracking.launch

Triger the target drone with the 3D Nav Goal in rviz to fly and the other drone will track the target drone:

<p align="center"> <img src="figs/rviz.gif" width="700"/> </p>

We made some improvements based on Ego-Planner and Fast-Planner to make the trajectory optimization in the framework work better.