In the field of “Clinical Robotics and Smart Systems”, the research team is currently working on the design and development of a modular platform for Minimally Invasive Surgery (MIS) (Figure 1). Minimally invasive surgery is recognized as the gold standard approach in surgery for its ability to reduce patient trauma and expedite recovery. However, it also presents limitations in the surgeon’s range of motion and precision [1]. One state-of-the-art solution [2] involves inserting tools into the surgical area, securely docked on a magnetically anchored platform attached to both sides of the surgical wall. Instead of being constrained by the insertion point of the trocar and limiting the workspace of the surgical tool, we are focusing on developing a modular robot that can enter cavities in a flexible manner and be assembled remotely once inserted into the body. Compared to the state of the art of laparoscopic tools [3], the research team is targeting an autonomous solution, where each tool is proximally actuated, ultimately reducing the tethering. This approach aims to address a wider range of procedures and provide greater adaptability during surgeries. To achieve this framework, four sub-activities are performed (i.e., magnetic actuation modeling and design, soft robot design, end-effector design, and control), organized among four Ph.D. students taking part in this project. The first step after insertion is to dock and actuate the surgical manipulator. When used in medical applications, whether used in diagnostics or interventional medicine, the magnetic coupling (sub-activity 1: magnetic actuation modeling and design for Minimally Invasive Surgery, Ph.D. student: Matteo Bernabei) between magnets represents a great opportunity to: (1) abandon wired solutions that pose a bottleneck on the overall invasiveness of a device; (2) lower the energy consumption often required for the actuation systems adopted at small scale; and (3) contribute to the issue of miniaturization of a surgical device. Such advantages come at the expense of an accurate analysis of the interaction taking place, as the latter is a function of several physical parameters of the magnetic components. The embedding of magnetic components in the proposed surgical device has the aim to allow for wireless anchoring of the device inserted into the patient’s cavity to the endoluminal wall, the remote assembly of the different modules together, and even further its actuation, through the exploitation of the torque transmission produced by the coupled magnets. But as magnetically actuated systems show limitations in the number of controllable DoFs, the research team is investigating a hybrid solution, coupling magnetic and pneumatic actuation (sub-activity 2: soft modular robotic arm for Minimally Invasive Surgery, PhD student: Alexia Le Gall). In medical soft robotics, pneumatic is a state-of-the-art solution, used for its safety and high strain and force output [4]; however, the technology stays intrinsically bulky, making its downscaling an ongoing theme. Therefore, we are looking into ways to harvest the properties of soft materials to reduce or even eliminate the tethering of our surgical robot to external equipment, while keeping the required dexterity and dimensions. Once the arm is in position, the end-effector tool comes into action (sub-activity 3: compliant and origami-inspired mechanisms to design surgical robotic tools, Ph.D. student: Lorenzo Mocellin). Given our framework, the requirements needed to introduce the system through a trocar or natural opening induce the need for miniaturized dexterous surgical instruments mounted on the tip. Targeting our system to be suited for minimally invasive procedures, the research team plans to design the end-effector of the system to be easily scalable, flexible, and optimized with respect to specific constraints. Therefore, the research team is exploring an origami-based approach and the compliant mechanism theory to achieve good mechanical properties, easiness in manufacturing, scalability, and improved reusability of the system. Finally, to operate the system, we need data-driven control strategies for modular soft robots (sub-activity 4: data-driven control strategies for modular soft robots, Ph.D. student: Zixi Chen). Soft robots provide nonlinear and time-delayed responses, which shows the difficulty of control. Moreover, modular soft robots have more degrees of freedom and actuation than single soft robots, hence improving the difficulty of controlling such a system. It is challenging to propose a control strategy based on physical models considering the randomness and large number of physical parameters in this system. Therefore, the research team plans to design a data-driven control strategy for modular soft robots.