Autonomous Navigation of TurtleBot3 Using DDPG Reinforcement Learning in ROS2 Framework

The Deep Deterministic Policy Gradient (DDPG) technique is used in this paper's deep reinforcement learning approach for TurtleBot3's autonomous navigation within a ROS2 environment. The DDPG algorithm uses an actor-critic structure to make decisions in real time while handling continuous action spaces. We include This study applies DDPG with Ornstein-Uhlenbeck noise to the TurtleBot3 platform. The capacity of the DDPG agent to navigate unfamiliar areas while avoiding obstacles and maximizing goal-reaching efficiency is demonstrated experimentally in a Gazebo simulation.