Rover Tracking is a Smart Robotics project focused on the development of an autonomous mobile robot capable of tracking objects in a simulated domestic environment using ROS2, Gazebo, and OpenCV.

The system is based on a differential-drive robot model equipped with an RGB camera and LiDAR sensor, described through URDF/Xacro and simulated inside Gazebo. ROS2 provides the modular communication architecture between perception, control, mapping, and navigation components.

Mapping and localization are performed using SLAM Toolbox, while autonomous navigation relies on the Navigation2 framework. Object tracking is implemented through a computer vision pipeline based on the OpenCV CSRT algorithm, enabling the robot to follow a target in real time by estimating its position from camera images and adjusting angular velocity accordingly.

Simulation results demonstrate effective tracking performance in controlled environments. Future developments include integrating deep-learning-based tracking methods and deploying the system on a real robotic platform.