Abstract: This study evaluates the performance and robustness of a port-Hamiltonian controller for the links of a liquid mirror telescope and a PI controller for the rotation of its liquid mirror, using ROS2 and Gazebo. The telescope links track a star’s apparent daily motion, while the liquid mirror achieves the required angular speed for a desired focal length. These references are computed based on the star’s name and focal length input. The telescope’s physical properties, including dimensions, masses, inertia, and 3D models, are stored in a Unified Robot Description Format (URDF) file, enabling Gazebo to initialize accurate simulations. Joint information—state interfaces (angular position and speed) and command interfaces (effort)—is also defined in the URDF. Controller configurations, including gains, bounds, and control parameters, are stored in a YAML file, ensuring seamless integration with Gazebo. The evaluation encompasses key performance metrics. For the two-link telescope, tracking accuracy, settling time, control effort, and energy efficiency are analyzed. For the liquid mirror, the primary focus is on tracking precision. The port-Hamiltonian controller’s performance is compared to inverse dynamics and super-twisting sliding mode controllers. Results show that the port-Hamiltonian controller achieves a favorable balance between accuracy and energy efficiency, exhibiting smoother control actions that reduce energy consumption and actuator wear. Its stability under varying conditions ensures high precision for astronomical observations. Furthermore, ROS2 and Gazebo provide a risk-free environment for extensive testing, facilitating a smooth transition to real-world implementation.

Keywords: Port-Hamiltonian Control, Liquid Mirror Telescope, ROS2 and Gazebo Simulation, Tracking Accuracy and Energy Efficiency

Pages: 471-486