Design and control of a four degrees of freedom manipulator for a mobile robot for automatic wine sampling in a winery.

The thesis project falls within the field of collaborative robotics and aims to automate the wine sampling process through the use of a mobile robot. The project can be divided into three main areas: the mobile platform, the robotic arm, and the end effector. This work focuses on the design of a four-degree-of-freedom robotic arm that allows the end effector to reach the necessary positions to complete the wine extraction from the barrels. After a brief review of existing technologies in the field of robotic arms, the study begins with the dimensioning of an arm featuring a double-parallelogram mechanism, defining the relevant component lengths and establishing an operational space to be satisfied. The double-parallelogram configuration allows to passively maintain a constant orientation of the end-effector. The work then proceeds with the analysis of forward and inverse kinematics to ensure that the chosen dimensions can cover the entire operational space. The D-H convention is employed for the kinematics analysis, which concludes with a dexterity analysis throughout the entire workspace. The core of the thesis is the mechanical design phase, where the creation of the prototype is achieved through the development of custom components and the selection of commercial ones. This is followed by a theoretical discussion on balancing systems for robotic arms, which can be integrated into the project to reduce the load on the actuators. Two balancing systems are analyzed: a spring-and-pulley system and a cam-spring-pulley system. Both systems are examined and compared for the balancing of each joint. Finally, the prototype is simulated and tested through the design of a controller, defining its logic to ensure that the wine sampling process is validated under realistic conditions. Through simulation testing, it has been verified that the manipulator is correctly designed and dimensioned to reach the entire operational space. The project shall proceed with the 3D-printing of components for testing on the physical model, incorporating the discussed balancing systems. The control algorithm needs to be extended to manage the entire mobile robot, integrating control for both the end effector and the mobile platform, which, in this initial phase, are only simulated to validate the controller of the robotic arm.