Multi-Rotor Obstacle Avoidance : Reinfocement Learning 4.0.0

An DRL-based path planning framework for real quadrotor

Video: Bilibili

Paper: DRL-based Path Planner and its Application in Real Quadrotor with LIDAR

Cite:

@article{yang2023drl,
  title={DRL-based Path Planner and its Application in Real Quadrotor with LIDAR},
  author={Yang, Yongsheng and Hou, Zhiwei and Chen, Hongbo and Lu, Peng},
  journal={Journal of Intelligent \& Robotic Systems},
  volume={107},
  number={3},
  pages={38},
  year={2023},
  publisher={Springer}
}

Latest Update

1. complete paper
2. improve documentation

1. feature

interface	version
ubuntu	18.04
ros	melodic
pytorch	1.4.0

2. installation

ROS melodic

refer to here

Gazebo 9

sudo apt install ros-melodic-gazebo-*

download models.zip and unpack under ~/.gazebo/.

Mavros

sudo apt install ros-melodic-mavros ros-melodic-mavros-extras
wget https://gitee.com/robin_shaun/XTDrone/raw/master/sitl_config/mavros/install_geographiclib_datasets.sh
sudo chmod a+x ./install_geographiclib_datasets.sh
sudo ./install_geographiclib_datasets.sh

PX4

git clone https://github.com/PX4/PX4-Autopilot.git
mv PX4-Autopilot PX4_Firmware
cd PX4_Firmware
git checkout -b xtdrone/dev v1.11.0-beta1
git submodule update --init --recursive
make px4_sitl_default gazebo

open ~/.bashrc, and insert

source ~/catkin_ws/devel/setup.bash
source ~/PX4_Firmware/Tools/setup_gazebo.bash ~/PX4_Firmware/ ~/PX4_Firmware/build/px4_sitl_default
export ROS_PACKAGE_PATH=$ROS_PACKAGE_PATH:~/PX4_Firmware
export ROS_PACKAGE_PATH=$ROS_PACKAGE_PATH:~/PX4_Firmware/Tools/sitl_gazebo

QGroundControl

refer to here

Anaconda

refer to here

virtual environment

create env

conda create -n my_env python=2.7
conda activate my_env

update pip

pip install --upgrade pip

install dependence

pip install requirements.txt

Building pkg

create workspace

mkdir catkin_ws
cd catkin_ws
mkdir src
cd src
catkin_init_workspace

Download this project and put it into src.

cd ..
catkin_make

copy models falco/imu_gazebo/UST_10LX to ~/PX4_Firmware/Tools/sitl_gazebo/models.

Run Samples

# open gazebo
roslaunch multi_rotor_avoidance_rl setup_env.launch

open QGroundControl load models/falco/params/sim.param.

# start task
rosrun multi_rotor_avoidance_rl training_node.py	
# rosrun multi_rotor_avoidance_rl testing_node.py