Ottimizzazione della cinematica e della dinamica dei robot industriali mediante modellazione numerica e validazione sperimentale

The kinematics and dynamics of the Cartesian robot are optimized in this thesis with the goal to improve its adaptability in high-mix low-volume production environment. In this work, a complementary methodology is developed that integrates analytical models, virtual prototypes, and experimental validation. The main focus of this project is the design optimization and performance enhancement of the linear belt drive and rack-and-pinion drive of the Cartesian robot. Three-mass and reduced-order two-mass models are derived for the linear belt drive to capture the effects of elasticity, belt damping, and non-linear friction on the system and to analyze its characteristics. A simplified single-mass model for the rack-and-pinion drive is developed to identify the main parameters that affect its response. These models provide a basis for the design and optimization strategies. In addition to these models, virtual prototype based on multi-body dynamics is developed and then co-simulated with a control strategy designed in MATLAB/Simulink. This allows for a detailed examination of the dynamic behavior of the system and analysis of the trajectory tracking performance and controller robustness. The kinematic and dynamic results obtained from co-simulations are analyzed and discussed, demonstrating the capabilities of multi-body dynamic simulations. Optimum trajectory generation and robust control strategy that improves trajectory tracking accuracy, reduces drive vibrations, and improves energy efficiency are discussed. Parametric analysis is conducted using virtual prototyping and analytical modeling to identify the sensitive design parameters that influence the robot's performance. An energy optimization strategy is used to minimize the total mechanical energy by varying the critical design parameters and in the presence of design constraints. The results obtained from the virtual prototyping and analytical modeling are then experimentally validated using a Cartesian robot. This validation of results proves the reliability of both modeling strategies, which can be used in the future to analyze and optimize the robot without making an actual prototype. The results obtained demonstrate that the integration of analytical modeling, virtual prototyping, and experimental validation provides a reliable structure for the design and optimization of reconfigurable Cartesian robots. The developed prototypes can be used to analyze and improve the performance of any required configuration of the robot without the need for physical prototypes, thereby reducing development time and cost. This work focuses only on energy minimization, but in future work, other performance metrics would also be considered for the design and optimization of the robot. The scope of the work is limited to individual drives of the robot instead of a whole robot, but it will be extended to complete robot analysis in the future.