Controlling soft robots is a significant challenge due to the nonlinear elastic nature of the soft materials that conform their structure. This paper studies the identification and control problems of a novel two-degrees-of-freedom, tendon-actuated, soft robotic arm. A decoupled identification approach is presented; later, a fractional order control strategy is proposed and tested experimentally, in comparison with PI solutions. The simulation and experimental results show the goodness of the modeling and control approaches discussed.

A soft joint has been designed and modeled to perform as a robotic joint with 2 Degrees of Freedom (DOF) (inclination and orientation). The joint actuation is based on a Cable-Driven Parallel Mechanism (CDPM). To study its performance in more detail, a test platform has been developed using components that can be manufactured in a 3D printer using a flexible polymer. The mathematical model of the kinematics of the soft joint is developed, which includes a blocking mechanism and the morphology workspace. The model is validated using Finite Element Analysis (FEA) (CAD software). Experimental tests are performed to validate the inverse kinematic model and to show the potential use of the prototype in robotic platforms such as manipulators and humanoid robots.

This study presents a novel method for controlling the position and orientation of the bioinspired I-SUPPORT soft robot, which represents a relevant advancement in the field of soft robotics. The approach is based on module actuation decoupling and fractional-order control, offering a more advanced and robust control solution. This innovation enhances the versatility of the robot and illustrates the efficacy of fractional-order controllers, which are comparable to current meta-learning-based controllers. The research involves experiments in both vertical and horizontal configurations, addressing tasks ranging from simple orientation to complex interactions, such as gentle rubbing during bathing activities with the robot. These experimental results exemplify the efficacy of the proposed control strategy and provide a foundation for future research in soft robotics control, underscoring its potential for broader applications and further technological advancement.

Resumen La identificación adecuada del sistema es esencial para lograr un buen rendimiento de las tareas y una buena interacción con el entorno mediante la robótica blanda. Aunque se han utilizado técnicas analíticas tradicionales basadas en modelos, a veces resultan insuficientes. Los métodos basados en datos son una alternativa prometedora. Sin embargo, su uso en robótica blanda ha sido limitado y poco explorado. Este trabajo explora el uso de procesos Gaussianos para identificar la cinemática directa e inversa de un brazo robótico blando de dos grados de libertad accionado por tres tendones.

Soft robotic grippers offer exceptional adaptability and safety for grasping diverse objects. However, achieving dexterous in-hand manipulation remains a significant challenge. Traditional tendon-driven designs are optimized for stable holding but lack the intrinsic degrees of freedom (DoF) required for dexterous object reorientation, often relying instead on complex hybrid mechanisms or repeated regrasping. This study presents a novel modular soft robotic gripper specifically designed to bridge the gap between stable grasping and active, continuous in-hand manipulation. The system features three independently actuated soft fingers, each providing three DoF through a combination of asymmetric tendon-driven bending and active base rotation. A core principle of this design is an asymmetric soft joint that exploits passive geometric blocking to provide direction-dependent compliance and mechanical protection against excessive deformation. By decoupling the bending and spatial reorientation motions, the gripper generates coordinated kinematic patterns across the fingers. Experimental validation demonstrates the system’s ability to robustly grasp objects spanning a wide spectrum of sizes and stiffnesses using a unified actuation strategy. Furthermore, the gripper successfully performs complex in-hand manipulation tasks, specifically, continuous object rotation along its axis without releasing the initial grasp. These results indicate that the proposed modular architecture significantly extends the dexterity of purely tendon-driven soft grippers while preserving their inherent compliance, scalability, and mechanical simplicity.