Requirement | Version / Release |
---|---|
Operating System | Ubuntu 16.04 |
ROS Distribution Release | ROS2 |
Pythin Version | python 3.9 |
The project involves the development and evaluation of two robot navigation models using reinforcement learning (RL). The goal is to navigate a robot in an unknown environment using ROS for real-time interaction. This includes path finding and obstacle avoidance. Two models are explored: the first utilizes an action value-function with linear function approximation, and the second employs a policy gradient method with non-linear function approximation. Each model is tuned and assessed based on its ability to effect
Create a new catkin workspace.
In your terminal shell, navigate to your intended directory for your new workspace and execute the following commands, replacing <wsname>
with your intended workspace name.
$ mkdir -p <wsname>/src
$ cd <wsname>/src/
$ catkin_init_workspace
$ cd ..
$ catkin_make