Control of an inflatable robot for space applications

In recent years, the collaboration between humans and robots has become essential, revealing to be crucial in space environment exploration. The interest in soft systems for space missions represents a growing trend in recent years. Soft robots offer higher dexterity, larger variability of movements, better usability in otherwise inaccessible places and higher safety. The aim of this project is to propose a visual servoing control strategy that can be used for inflatable robots for space applications. Visual servoing is a well-known approach to guide robots using visual information. Image processing, robotics and control theory are combined in order to control the motion of a robot depending on the visual information extracted from the images captures by one or several cameras. The robot to be controlled, the POPUP robot, is a soft manipulator developed for space applications. A prototype made up of two inflatable links and three rigid joints was developed in the laboratories of the Polytechnic of Turin. The structure presents 3 degree of freedom, but it can be extended to 6 degree of freedom through the integration of a wrist. The thesis work aims to implement algorithms for controlling the prototype and its simulated model in a virtual environment. Indeed, simulation allows control strategies to be tested safely. To obtain realistic simulations of the robotic scenario, Gazebo was used as a 3D simulator and ROS 1.0 as a robotic interface. Combining both results in a powerful robot simulator. Therefore, first of all, a model of the robot is created on Gazebo. The Effort-Position control is implemented on ROS 1.0 and the vision system is integrated. It consists of one camera system on the end-effector of the robot. On the prototype a velocity control algorithm is implemented on ROS 1.0 and the depth camera Realsense d435 is integrated. In both real and simulated cases, vision systems are useful for performing tasks such as object recognition, obstacle avoidance or grasping. It also enables precise and accurate control even in the case of inflatable links.