Soft Robotic Sim2Real via Conditional Flow Matching

A fluidic relaxation oscillator for reprogrammable sequential actuation in soft robots

Soft robots, inspired by living organisms in nature, are primarily made of soft materials, and can be used to perform delicate tasks due to their high flexibility, such as grasping and locomotion. However, it is a challenge to efficiently manufacture soft robots with complex functions. In recent years, 3D printing technology has greatly improved the efficiency and flexibility of manufacturing soft robots. Unlike traditional subtractive manufacturing technologies, 3D printing, as an additive manufacturing method, can directly produce parts of high quality and complex geometry for soft robots without manual errors or costly post-processing. In this review, we investigate the basic concepts and working principles of current 3D printing technologies, including stereolithography, selective laser sintering, material extrusion, and material jetting. The advantages and disadvantages of fabricating soft robots are discussed. Various 3D printing materials for soft robots are introduced, including elastomers, shape memory polymers, hydrogels, composites, and other materials. Their functions and limitations in soft robots are illustrated. The existing 3D-printed soft robots, including soft grippers, soft locomotion robots, and wearable soft robots, are demonstrated. Their application in industrial, manufacturing, service, and assistive medical fields is discussed. We summarize the challenges of 3D printing at the technical level, material level, and application level. The prospects of 3D printing technology in the field of soft robots are explored.

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.

Dielectric Elastomer Actuators (DEAs) are a type of electroactive polymers that transform electric energy into mechanical work and are capable of large strains and high energy densities. DEAs are often described as “artificial muscles” because they exhibit strains, energy density, and responsiveness similar to natural human muscles. Their applications as soft robotic grippers, artificial arms, underwater robots, and aerial robots have previously been demonstrated. Single-layer DEAs have a simple design, with a dielectric layer sandwiched between compliant electrodes. However, by alternating positive and negative electrodes, multilayer DEAs can exhibit larger forces and strains than single-layer DEAs. DEAs are actuated by applying voltage and forming an electric field, generating a Maxwell stress between the electrodes. The process of fabricating a multilayer DEA consists of spin coating and curing layers of elastomer, stamping carbon nanotubes (CNT), and making connections to the electrodes. By alternating positive and negative electrodes, multilayer DEAs are constructed, which can exhibit larger forces and strains than single-layer DEAs. When electrically activated, DEAs expand laterally; however, by fabricating a bimorph DEA actuator, we demonstrate bending DEAs. A bimorph DEA consists of a multilayer DEA layer attached to a thicker inactive elastomer layer. When the bimorph is actuated, only the DEA layer expands, causing the actuator to bend. We study the relationship between the bending and stiffness of the inactive layer. By changing the shape of the DEA to a triangle or wing-like shape, we also emulate the high-frequency flapping of birds. Work begun on characterizing the voltage-controlled deflection of bimorph DEAs and understanding the mechanics of the bending of a simple beam in terms of forces at Harvard University will be continued on the Navajo Reservation at Navajo Technical University.

Soft pneumatic actuators possess attributes of large deformation, high driving force and light weight in the application of soft robots and smart devices. However, most reported soft pneumatic actuators are with rigid hydraulic source such as motor driven pump, piston and pressurized reservoir. These rigid and heavy hydraulic sources limit the actuation and compliance of the soft robots. Inspired by the bladders and hydrostatic skeleton of natural creatures, we propose a soft hydraulic robot consisting of dielectric elastomer (DE) and hydrogel, exhibiting an excellent actuating performance. An inflated DE balloon functions as the soft hydraulic source, in which the pressure of the containing water can be tuned by voltage. Hydrogel chambers are connected to the DE balloon as the hydraulic actuator, deforming as a soft robotic gripper. A new analytical approach is proposed to describe the system’s behaviors, which couples the electromechanical actuation of DE and the hydraulic deformation of hydrogel chamber. The proposed model is validated by good agreement between the numerical and experimental data. The proposed model could serve as a new tool for modeling and characterizing soft robots with hydraulic actuation. The working principles can guide the design and control of soft robots and smart structures.

Soft robots are distinguished by their flexibility and adaptability, allowing them to perform nearly impossible tasks for rigid robots. However, controlling their behavior is challenging due to their nonlinear material response and infinite degrees of freedom. A potential solution to these challenges is to discretize their infinite‐dimensional configuration space into a finite but sufficiently large number of functional modes with programmed dynamics. A strategy is presented for co‐designing the desired tasks and morphology of pneumatically actuated soft robots with multiple encoded stable states and dynamic responses. This approach introduces a general method to capture the soft robots' response using an energy‐based analytical model, the parameters of which are obtained using Recursive Feature Elimination. The resulting lumped‐parameter model enables the inverse co‐design of the robot's morphology and planned tasks by embodying specific dynamics upon actuation. This approach's ability to explore the configuration space is shown by co‐designing kinematics with optimized stiffnesses and time responses to obtain robots capable of classifying the size and weight of objects and displaying adaptable locomotion with minimal feedback control. This strategy offers a framework for simplifying the control of soft robots by exploiting the mechanics of multistable structures and embodying mechanical intelligence into soft material systems.

Soft robots are flexible and adaptable allowing them to conform to objects of different shapes and sizes. However, the flexibility of soft robots leads to challenges in designing and fabricating them to satisfy application-specific requirements. In this paper, we present a class of 3-D printable three-chambered actuator modules, along with analytical models to predict actuation behavior based on a set of design parameters related to the bellows. The actuator modules are designed parametrically and fabricated on a commercial 3-D printer, which improves the speed, accessibility, and repeatability of fabrication. The approach presented here provides a framework for tailoring actuators to specific soft robotic applications. This process relates actuation characteristics (e.g., bend angle and blocked force) to a small set of design parameters with the goal of reducing the design-fabrication cycle time. The forward and inverse kinematics are defined for the three-chambered actuator. The kinematics are experimentally validated using a motion capture system. We performed a sensitivity analysis to understand which of the geometric parameters of the bellows have the greatest effect on bending and blocked force. This design framework can be tailored to realize soft actuator modules, for example applications such as fingers for a soft robotic gripper, as a robotic arm for manipulating objects, or as legs for a soft walking robot.

Minimally invasive surgery (MIS), e.g., interventional endoscopy and single-incision laparoscopy, has paved the way toward increased patient safety, fewer postop- erative complications, and shorter recovery times [1]. The inherent compliant nature of soft robots makes them suitable for addressing current limitations in MIS, such as the lack of dexterity of rigid robotic devices; however, soft robots often lack adequate shape and contact sensing which makes them difficult to control [2], [3]. The future of laparoscopic surgical tools is moving toward flexible and soft robotic solutions for safer interaction with hu- man tissue [4]. However, newly developed technologies lack effective embedded sensing [5]. This work presents a fully soft robot combining i) a soft optical sensorized multi-modal gripper for tip tracking and contact force sensing with ii) a multi-directional bending module with integrated 3D shape sensing (Fig. 1, a-b). The gripper embeds two soft pneumatic actuators to deploy the jaws, two soft pneumatic actuators to control grasping, and two fully soft optical waveguide sensors (WG) (one in each jaw) with tuned roughness to monitor both actuator tip positions and subsequent contact force on an object.

Data-driven calibration methods have shown promising results for accurate proprioception in soft robotics. This process can be greatly benefited by adopting numerical simulation for computational efficiency. However, the gap between the simulated and real domains limits the accurate, generalized application of the approach. Herein, we propose an unsupervised domain adaptation framework as a data-efficient, generalized alignment of these heterogeneous sensor domains. A dual cross-modal autoencoder was designed to match the sensor domains at a feature level without any extensive labeling process, facilitating the computationally efficient transferability to various tasks. Moreover, our framework integrates domain adaptation with anomaly detection, which endows robots with the capability for external collision detection. As a proof-of-concept, the methodology was adopted for the famous soft robot design, a multigait soft robot, and two fundamental perception tasks for autonomous robot operation, involving high-fidelity shape estimation and collision detection. The resulting perception demonstrates the digital-twinned calibration process in both the simulated and real domains. The proposed design outperforms the existing prevalent benchmarks for both perception tasks. This unsupervised framework envisions a new approach to imparting embodied intelligence to soft robotic systems via blending simulation.

Soft robots demonstrate great potential compared with traditional rigid robots owing to their inherently soft body structures. Although researchers have made tremendous progress in recent years, existing soft robots are in general plagued by a main issue: slow speeds and small forces. In this work, we aim to address this issue by actively designing the energy landscape of the soft body: the total strain energy with respect to the robot's deformation. With such a strategy, a soft robot's dynamics can be tuned to have fast and strong motion. We introduce the general design principle using a soft module with two stable states that can rapidly switch from one state to the other under external forces. We characterize the required triggering (switching) force with respect to design parameters (e.g., the initial shape of the module). We then apply the soft bistable module to develop fast and strong soft robots, whose triggering forces are generated by a soft actuator - twisted-and-coiled actuator (TCA). We demonstrate a soft gripper that can hold weights more than 8 times its own weight, and a soft jumping robot that can jump more than 5 times its body height. We envision our strategies will overcome the weakness of soft robots to unleash their potential for diverse applications.

A meshfree model of hard-magnetic soft materials

In recent times, robotics and especially soft robotics has attracted a wide range of researchers and scientists. As oft robotics has an extensive number of advantages in the real environment, due to their less complex system and cost. Soft grippers are more adaptive as compared to the rigid robotic grippers. The grasping performance of soft robots can be improved without bringing major changes in the control inputs. Machine learning has played a vital role in improving the controls and increasing the number of applications of these kinds of robots in the real world. In this paper relevant research in modeling, design, intelligent control, sensing, and practical applications of soft robots has been discussed.

With the rapid development of soft robotics, there is a growing demand for compact, high-performance, and flexible actuators. Pneumatic actuators are particularly favored by researchers due to their high efficiency, low cost, and ease of fabrication. However, the existing pneumatic actuators are faced with challenges such as large volume, low actuating frequency, and insufficient propulsion. For the first time, this work presents a novel thin-plate pneumatic actuator based on prestressed principles, featuring an efficient energy storage and release mechanism. The actuator consists of a prestretched layer, a constrained layer, and a chamber (PCC), with a total thickness of less than 1 mm. Therefore, the elastic potential energy is preserved in the prestretched film within the actuator body, which enables high actuating frequency and large propulsion. Effects of key structural parameters on the propulsion of the PCC actuator are investigated by a mathematical model, finite element analysis, and experiments to further elucidate its energy storage and release mechanism. Moreover, the PCC actuator is applied to realize a soft gripper, a prestressed hinge, and a wireless jellyfish-like robot, whose performances are verified by experiments. Results demonstrate the high flexibility and rapid response of the soft gripper and prestressed hinge. In particular, the gripper is capable of stably gripping objects 40 times its weight for extended periods. In addition, the wireless jellyfish-like robot achieves an upward swimming speed of 58.03 mm/s, which is superior to the existing soft jellyfish-like robots of similar size. Overall, the PCC actuator features a lightweight structure and high energy storage capacity, providing significant potential for innovative applications in soft robotics.

High-force soft pneumatic actuators based on novel casting method for robotic applications

Soft robotics is an emerging field, since it offers distinct opportunities in areas where conventional rigid robots are not a feasible solution. However, due to the complex motions of soft robots and the stretchable nature of soft building materials, conventional electronic and fiber optic sensors cannot be used in soft robots, thus, hindering the soft robots’ ability to sense and respond to their surroundings. Fiber Bragg grating (FBG)-based sensors are very popular among various fiber optic sensors, but their stiff nature makes it challenging to be used in soft robotics. In this study, a soft robotic gripper with a sinusoidally embedded stretchable FBG-based fiber optic sensor is demonstrated. Unlike a straight FBG embedding configuration, this unique sinusoidal configuration prevents sensor dislocation, supports stretchability and improves sensitivity by seven times when compared to a straight configuration. Furthermore, the sinusoidally embedded FBG facilitates the detection of various movements and events occurring at the soft robotic gripper, such as (de)actuation, object holding and external perturbation. The combination of a soft robot and stretchable fiber optic sensor is a novel approach to enable a soft robot to sense and response to its surroundings, as well as to provide its operation status to the controller.