Elastic-Tracker is a flexible trajectory planning framework that can deal with challenging tracking tasks with guaranteed safety and visibility.
Authors: Jialin Ji, Neng Pan and Fei Gao from the ZJU Fast Lab.
Paper: Elastic Tracker: A Spatio-temporal Trajectory Planner Flexible Aerial Tracking, Jialin Ji, Neng Pan, Chao Xu, Fei Gao, Accepted in IEEE International Conference on Robotics and Automation (ICRA 2022).
Video Links: youtube or bilibili
[NOTE] remember to change the CUDA option of src/uav_simulator/local_sensing/CMakeLists.txt
Preparation and visualization:
git clone https://github.com/ZJU-FAST-Lab/Elastic-Tracker.git
cd Elastic-Tracker
catkin_make
source devel/setup.zsh
chmod +x sh_utils/pub_triger.sh
roslaunch mapping rviz_sim.launch
A small drone with the global map as the chasing target:
roslaunch planning fake_target.launch
Start the elastic tracker:
roslaunch planning simulation1.launch
Triger the drone to track the target:
./sh_utils/pub_triger.sh
Comparision of the planners with and without
visibility guarantee:
roslaunch planning simulation2.launch
First start the stage of tracking:
roslaunch planning fake_car_target.launch
roslaunch planning simulation_landing.launch
./sh_utils/pub_triger.sh
Triger the drone to land on the moving vehicle:
./sh_utils/land_triger.sh
We use MINCO as our trajectory representation.
We use DecompROS for safe flight corridor generation and visualization.