An End-to-End Framework for Modelling Pneumatic Soft Robots Based on Differentiable Finite Element Methods

Soft robots are known for their ability to perform tasks with great adaptability, enabled by their distributed, non-uniform stiffness and actuation. Bending is the most fundamental motion for soft robot design, but creating robust, and easy-to-fabricate soft bending joint with tunable properties remains an active problem of research. In this work, we demonstrate an inflatable actuation module for soft robots with a defined bending plane enabled by forced partial wrinkling. This lowers the structural stiffness in the bending direction, with the final stiffness easily designed by the ratio of wrinkled and unwrinkled regions. We show the stiffness properties of the actuation module through a first-principle model validated by experimental characterization, and demonstrate the module's ability to maintain the kinematic constraint over a large range of loading conditions. We illustrate how these properties give the potential for complex actuation in a soft continuum robot and for decoupling actuation force and efficiency from load capacity. The module provides a novel method for embedding intelligent actuation into soft pneumatic robots.

Soft robots have been attracting much research attention in the past decades owing to their unique properties that are inherently different from conventional rigid robots. The intrinsic characteristics of soft robots, such as compliance and flexibility, have made them a promising supplement for many existing robotic applications. For instance, soft robots provide safer human-machine interaction and better adaptability to unexplored environments than the conventional rigid ones. However, their “soft” feature would conversely cause complex material deformation due to their high-DoF nature, which has made their kinematic or dynamic models more difficult to deduce. Logically, the process of designing model-based controllers would therefore suffer from the lack of efficient and accurate modeling methods. In this letter, through adopting Finite Element Method (FEM), we demonstrate a FEM-based nonlinear control method for a soft trunk robot to achieve trajectory tracking objectives. The proposed controller is developed by utilizing the essential information from the dynamic FE model of the studied robot. The details of deriving this nonlinear control method are demonstrated, then followed by 4 practical experiments on the soft truck robot to verify the feasibility of the control method. The corresponding experimental results are presented accordingly.

Mechanoreception, the ability of robots to detect mechanical stimuli from the internal and external environments, contributes significantly to improving safety and task performance during the operation of robots in unstructured environments. Various approaches have been proposed to endow robot systems with mechanoreception. In the case of soft robots, the state-of-the-art mechanosensory solutions typically embedded dedicated deformable sensors into the soft body, giving rise to fabrication complexity and signal sophistication. In this study, we propose a novel mechanoreception scheme to enable pneumatic-driven soft robots to perceive proprioceptive movements as well as external contacts. Both internal and external mechanical parameters can be decoded from intuitive cues of body deformation and pneumatic pressure signals. In contrast to most existing solutions employing dedicated deformable sensors, the proposed approach only utilizes pressure feedback, which is typically available from the pneumatic pressure sensors incorporated in the control loop of most pneumatic soft robots. The concept was implemented and validated on a proprietary robotic gripper with a linear soft pneumatic actuator, demonstrating the capability in simultaneous detection of actuator position and external contact forceAfter the proposed approach, the gripper can achieve both active and passive mechanosensation, with demonstrated experiments in grasping force estimation, contact loss detection, object stiffness identification, and contour measurements. This approach offers an alternative route to achieving excellent internal/environmental awareness without requiring dedicated sensing modalities.

A key challenge in soft robotics is controlling the large deformation experienced as a result of high compliance nature of soft robots. In this work, a software control strategy for regulating the amount of internal positive and negative air pressure inside pneumatic soft robots is presented. Since the air pressure has a direct effect on the amount of deformation, the position of the robot is controlled. Pressure control was implemented with a fuzzy logic controller, which is described with its performance shown. The approach can be integrated into any specified soft robotic actuator requiring pneumatic actuation e.g. bending, triangular and muscle actuators.

The conventional pneumatic soft robot has the problem of insufficient stiffness, while in the magnetorheological soft robot, the magnetic field provided by electromagnet has the disadvantage of oversized structure and poor flexibility. This paper presents a variable stiffness pneumatic soft robot based on magnetorheological grease (MRG) to solve these problems. Its three soft fingers cooperate with the adjustable gripper to adjust the gripping range for the robot hand, and it is used to provide gripping driving force through the bending drive. The MRG layer is designed on the gripping surface to provide adaptivity and rigid support for the gripped objects. A magnetic-air structure consisting of a Halbach array and Halbach array actuator is designed inside the soft fingers to provide a flexible magnetic field for the MRG layer. Theoretical and simulation analysis is carried out, and the results show that the state of the MRG changes and the stiffness of the clamping surface changes under the working pressure of 30 kPa. Finally, the experiment further proves the variable and high adaptivity of the surface stiffness of the gripping surface to reduce the damage to the gripped objects.

Functional restoration of the cerebral cortex relies on activity-dependent neuroplasticity following stroke. However, optimizing robotic rehabilitation for hand recovery remains a significant challenge. This proof-of-concept study investigates the feasibility of a configurable soft pneumatic robot in modulating the brain network among ten participants with hand motor impairments after stroke. The programmable robotic intervention was administered in a randomized sequence for resting-state assessment, slow-mode, and fast-mode robotic therapy by adjusting working mode, action time, and interaction duration. Functional near-infrared spectroscopy (fNIRS) was employed to measure cortical activity and functional connectivity. Additionally, graph theory metrics, including clustering coefficient, average path length, small-world index, global efficiency, degree centrality, and eigenvector centrality, were derived from fNIRS-based functional connectivity matrices. The results demonstrated that robotic intervention significantly improved clustering coefficient (P = 0.034), average path length (P = 0.007), and global efficiency (P = 0.001). While small-world index, degree centrality, and eigenvector centrality showed an increasing trend, these differences did not reach statistical significance. Moreover, fast-mode therapy induced more substantial changes in clustering coefficient compared to slow-mode therapy, suggesting a potentially stronger effect on neural reorganization. These findings preliminarily support the use of soft robotics with adjustable paradigms to enhance brain network connectivity and facilitate neuroplasticity following stroke. The observed improvements in global efficiency and small-world properties indicate that robotic therapy optimizes cortical organization, promoting functional recovery. Further studies with larger sample sizes and personalized intervention protocols are needed to confirm these results and explore their long-term effects.

This paper presents a study on improving the mechanical performance of soft pneumatic robots through parameterized layout and shape optimization of air chambers. Most soft robots are designed mainly by using biomimetics or engineering intuition. To fully explore the potential of the soft robots, however, it is highly desirable to develop a rational and systematic method to optimize the structural layout. In particular, structural stiffness of soft pneumatic robots is a key issue to be considered in practical design. This study first shows that the structural layout has a significant impact on the structural stiffness of the soft robot. In the optimization model, the design parameters explicitly defining the layout and shapes of the air chambers are to be optimized. In order to attain adequate structural stiffness so as to enhance the gripping force while improving the delivered motion, the output work is considered as the objective function to be maximized. This optimizaiton model enables a gradient-based mathematical programming algorithm to be used. It is shown through finite element analysis and physical experiments that the optimized layout and shape of the air chambers significantly improve the output work of the soft pneumatic gripper. The method is also adapted to be used in the structural design of a crawling soft robot.

How to model the dynamics and represent the behavior of soft continuum robot accurately and efficiently has always been a challenging problem. In this paper, a new soft continuum robot dynamic model modeling idea is proposed, that is, the pneumatic soft continuum robot is regarded as a variable-stiffness beam structure when bending. The stiffness of the variable-stiffness beam structure is related to the input air pressure, and a large deflection theorem is employed for force analysis of the variable-stiffness beam. Finally, the soft continuous robot pressure model is used to solve the deflection curve equation. In the experiment, the bending posture of the soft continuum robot obtained by the visual capture system was compared with the simulation results, and the average error of the model was 8.5%

Pneumatic soft robotic systems show remarkable potentials in producing versatile locomotion and manipulations, owing to their flexibility in structural design and material selections. However, fabrication of pneumatic soft robotic systems with complex 3D structures and material distribution still remains elusive. Herein, a mode‐free fabrication approach, called planar laser cutting and stacking fabrication (PLCSF), is proposed to create pneumatic soft robotic systems with multi‐material and complex structure, which involves the following steps: 1) slicing the 3D model of desired pneumatic soft robotic systems into 2D layers; 2) fabricating each layer via laser cutting corresponding 2D membrane; 3) stacking all layers together to finish the fabrication. With the PLCSF approach, various prevalent pneumatic soft actuators are fabricated with complex structures (including fiber‐reinforced and pneu‐net) and different actuation modes (such as bending, elongation, twisting, abduction, contraction, and grabbing). The scalability of the PLCSF approach to fabricate multiple degrees‐of‐freedom (DOFs) pneumatic soft actuators with integrated structures, such as twisting‐bending pneumatic soft actuators for delivering and bending‐elongation‐bending pneumatic soft actuators for crawling, is also demonstrated. It is further demonstrated that the PLCSF approach also enables creating a bio‐inspired soft hand with nine DOFs, capable of various dexterous motions and manipulations.

Living organisms intertwine soft (e.g., muscle) and hard (e.g., bones) materials, giving them an intrinsic flexibility and resiliency often lacking in conventional rigid robots. The emerging field of soft robotics seeks to harness these same properties to create resilient machines. The nature of soft materials, however, presents considerable challenges to aspects of design, construction, and control—and up until now, the vast majority of gaits for soft robots have been hand-designed through empirical trial-and-error. This article describes an easy-to-assemble tensegrity-based soft robot capable of highly dynamic locomotive gaits and demonstrating structural and behavioral resilience in the face of physical damage. Enabling this is the use of a machine learning algorithm able to discover effective gaits with a minimal number of physical trials. These results lend further credence to soft-robotic approaches that seek to harness the interaction of complex material dynamics to generate a wealth of dynamical behaviors.

This article describes an open-source hardware platform for controlling pneumatic soft robotic systems and presents the comparison of control schemes with on-off and proportional valves. The Pneumatic Soft Robotics Driver (PneuSoRD) can be used with up to one pump and pressure accumulator, 26 on-off valves, and 5 proportional valves, any of which can be operated in open or closed-loop control using up to 12 sensor inputs, which allows for the simultaneous control of a large number of soft actuators. The electronic driver connects to a National Instruments myRIO controller or an Arduino Due with the use of an adapter shield. A library of pressure control algorithms in both LabVIEW and Simulink is provided that includes bang-bang control, hysteresis control and PID control using on-off or proportional valves. LabVIEW and Simulink provide user-friendly interfaces for rapid prototyping of control algorithms and real-time evaluation of pressure dynamics. The characteristics and performance of these control methods and pneumatic setups are evaluated to simplify the choice of valves and control algorithm for a given application.

Soft robots actuate themselves and their world through induced pressure and strain, and can often sense these quantities as well. We hypothesize that coordination in a tightly coupled collective of soft robots can be achieved with purely proprioceptive sensing and no direct communication. In this paper, we target a platform of soft pneumatic modules capable of sensing strain on their perimeter, with the goal of using only the robots' own soft actuators and sensors as a medium for distributed coordination. However, methods for modelling, sensing, and controlling strain in such soft robot collectives are not well understood. To address this challenge, we introduce and validate a computationally efficient spring-based model for two-dimensional sheets of soft pneumatic robots. We then translate a classical consensus algorithm to use only proprioceptive data, test in simulation, and show that due to the physical coupling between robots we can achieve consensus-like coordination. We discuss the unique challenges of strain sensors and next steps to bringing these findings to hardware. These findings have promising potential for smart materials and large-scale collectives, because they omit the need for additional communication infrastructure to support coordination.

Rapid motion in soft pneumatic robots is typically achieved through actuators that either use a fast volume input generated from pressure control, employ an integrated power source, such as chemical explosions, or are designed to embed elastic instabilities in the body of the robot. This paper presents a bi‐shell valve that can fast actuate soft actuators neither relying on the fast volume input provided by pressure control strategies nor requiring modifications to the architecture of the actuator. The bi‐shell valve consists of a spherical cap and an imperfect shell with a geometrically tuned defect that enables shell snapping interaction to convert a slowly dispensed volume input into a fast volume output. This function is beyond those of current valves capable to perform fluidic flow regulation. Validated through experiments, the analysis unveils that the spherical cap sets the threshold of the snapping pressure along with the upper bounds of volume and energy output, while the imperfect shell interacts with the cap to store and deliver the desired output for rapid actuation. Geometry variations of the bi‐shell valve are provided to show that the concept is versatile. A final demonstration shows that the soft valve can quickly actuate a striker.

Millions of years’ evolution has imparted life on earth with excellent environment adaptability. Of particular interest to scientists are some plants capable of macroscopically and reversibly altering their morphological and mechanical properties in response to external stimuli from the surrounding environment. These intriguing natural phenomena and underlying actuation mechanisms have provided important design guidance and principles for man-made soft robotic systems. Constructing bio-inspired soft robotic systems with effective actuation requires the efficient supply of mechanical energy generated from external inputs, such as temperature, light, and electricity. By combining bio-inspired designs with stimuli-responsive materials, various intelligent soft robotic systems that demonstrate promising and exciting results have been developed. As one of the building materials for soft robotics, hydrogels are gaining increasing attention owing to their advantageous properties, such as ultra-tunable modulus, high compliance, varying stimuli-responsiveness, good biocompatibility, and high transparency. In this review article, we summarize the recent progress on plant-inspired soft robotics assembled by stimuli-responsive hydrogels with a particular focus on their actuation mechanisms, fabrication, and application. Meanwhile, some critical challenges and problems associated with current hydrogel-based soft robotics are briefly introduced, and possible solutions are proposed. We expect that this review would provide elementary tutorial guidelines to audiences who are interested in the study on nature-inspired soft robotics, especially hydrogel-based intelligent soft robotic systems.

Soft robots have the potential to significantly change the way that robots interact with the environment and with humans. However, accurately modeling soft robot dynamics in order to do model-based control is extremely time consuming and difficult. neural networks are a powerful tool for modeling systems with complex dynamics such as an inflatable robot link with antagonistic pneumatic actuation. Unfortunately it is also difficult to apply standard model-based control techniques using a neural net. In this work, we show that the gradients used within a neural net to relate system states and inputs to outputs can be used to formulate a linearized discrete state space representation of the system. Using the state space representation, model predictive control can be developed with a one degree of freedom soft robot to achieve position control within 2° of the commanded joint angle. Additionally, control using the model derived from the neural net has similar performance to control using a model derived from first principles that took significantly longer to develop. This shows the potential of combining empirical modeling approaches with model-based control for soft robots.