shen/deep-rl-navigation
1
0
Fork 0
mirror of https://github.com/vishweshvhavle/deep-rl-navigation.git synced 2024-11-25 16:32:50 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
main
deep-rl-navigation/README.md
Vishwesh Vhavle 6459b7f0c1
Update README.md
2023-12-13 13:08:16 +05:30

1.5 KiB
Raw Permalink Blame History

Goal-Driven Deep RL Policy for Robot Navigation

Deep Reinforcement Learning for mobile robot navigation in ROS2 Gazebo simulator. Using Twin Delayed Deep Deterministic Policy Gradient (TD3) neural network, a robot learns to navigate to a random goal point in a simulated environment while avoiding obstacles. Trained in ROS2 Humble & Gazebo simulator with PyTorch.

Screenshot from 2023-12-09 11-41-59

sim2

sim1

How To Run

Install Python 3.10, ROS2 Humble, Gazebo 11 on Ubuntu 22.04

git clone git@github.com:vishweshvhavle/deep-rl-navigation.git
cd deep-rl-navigation/DRL_robot_navigation_ros2/
sudo rosdep init
rosdep install --from-paths src --ignore-src -y
colcon build
cd ..

For Training -

ros2 launch td3 train_simulation.launch.py

For Testing -

ros2 launch td3 test_simulation.launch.py

Academic Integrity

If you are currently enrolled in this course, please refer to IIIT-Delhi's Policy on Academic Integrity before referring to any of the repository contents. This repository contains the work we did as undergrads at IIIT-Delhi in CSE-564 Reinforcement Learning course. We do not encourage plagiarism of any kind.