Soft Manipulator Research Articles

Pushing with Soft Robotic Arms via Deep Reinforcement Learning

Soft robots can adaptively interact with unstructured environments. However, nonlinear soft material properties challenge modeling and control. Learning‐based controllers that leverage efficient mechanical models are promising for solving complex interaction tasks. This article develops a closed‐loop pose/force controller for a dexterous soft manipulator enabling dynamic pushing tasks using deep reinforcement learning. Force tests investigate the mechanical properties of a soft robot module, resulting in orthogonal forces of N. Then, the policy is trained in simulation leveraging a dynamic Cosserat rod model of the soft robot. Domain randomization mitigate the sim‐to‐real gap while careful reward engineering induced pose and force control even without explicit force inputs. Despite the approximate simulation, the sim‐to‐real transfer achieved an average reaching distance of mm (), an average orientation error of rad () and applied pushing forces up to N. Such performance is reasonable for the intended assistive tasks of the manipulator. The experiments uncovered that the soft robot interacting with the environment exhibited torsional and counter‐balancing movements. Although not explicitly enforced, they emerged from the mechanical intelligence of the manipulator. The results demonstrate the potential of soft robotic manipulation via reinforcement learning.

Analytical Modeling of Cable-Driven Soft Manipulators with Constant Curvature: A Simulation Approach

Analytical Modeling of Cable-Driven Soft Manipulators with Constant Curvature: A Simulation Approach

Underwater fluid-driven soft dock for dynamic recovery of AUVs with improved pose tolerance

Recovering autonomous underwater vehicles (AUVs) by a mobile platform is a complex challenge. Traditional docking stations require strict pose deviation from the underwater vehicle. To overcome this issue, we propose a novel solution: an in-situ fluid-driven soft dock. This system comprises three soft robotic manipulators that possess guiding, capturing, and securing functionalities, respectively. We present the design, fabrication, and control methods for each manipulator considering the requirements of underwater docking operations. The theoretically allowable docking pose deviation range during AUV docking is analyzed. Additionally, the required force of soft grippers was evaluated through simulation and equivalent experimental tests. To enable dynamic recovery of AUVs while the soft robotic manipulators remain mobile on the experimental platform, we introduce a hydraulic control method based on attitude and pressure feedback. The test results indicate that the designed soft manipulators can operate and maintain good sealing performance under an external pressure of no less than 11 MPa. Moreover, the soft dock successfully recovered and secured AUVs of two different diameters in 20 dynamic tests, achieving a 90% success rate with a maximum AUV velocity of 1.6 m/s.

A Moving Horizon Estimation for a Class of Soft Continuum Manipulators

Abstract This work proposes a formulation for the moving horizon estimation technique considering both states and parameters obtained simultaneously with reduced estimation issues. The parameters can be either constant or time-varying, while states are related to complex output trajectories related to underactuated systems. Particularly, the proposed formulation considers additional equality constraints as a counterpart of the dynamics tube-model predictive control. Thus, it becomes less dependent on probabilistic information, such as probability density function and covariance of the process noise. In addition, the calibration of the method parameters has less sensitivity and driven by the tube constraints. The proposed approach can be applied in different systems; however, here it is detailed for a class of soft continuum manipulators with fluidic actuation through a variable flowrate and demonstrated with numerical simulations in planar motion. Results indicate robustness of the algorithm estimation in a challenging scenario arising from underactuation as well as in the presence of uncertainty and external disturbance, while simultaneously states and a vector of structural parameters are coherently estimated.

Combined classification and regression inverse kinematics modeling for overall shape prediction of soft manipulators

In helping astronauts perform tasks, soft manipulators offer the advantages of softness, lightness, low cost, and interaction safety to interact with. However, modeling their inverse kinematics through traditional methods is subject to truncation errors, complex derivations and inexpressibility equations, owing to geometric singularities. Furthermore, facing the future control requirement for multiple major feature points on manipulators, most neural network approaches that solve the drive space only by the inverse solution of the end task space, cannot obtain the inverse kinematics solutions of all feature points, due to their including redundancy (the dimension n of the drive space of feature points is larger than that of the end task space). Therefore, this paper proposes a method combining classification and regression based on multi-layer perceptron for inverse kinematic modeling, namely the IK-MLP-CR method, with the bending states classified and the whole manipulator shape predicted, greatly reducing the amount of training data and training time. Finally, the performance evaluation of the proposed model is validated through the simulation and ground microgravity experiments. Results show that using the IK-MLP-CR method results in a successful grasping rate of over 97%, and that the trend of the experimental bending conforms to physical motions under different pressures.

A comprehensive dynamic model and configuration control of a free-floating soft manipulator-spacecraft system

Space robots are inseparable components of many space missions. However, capabilities of space robots with rigid manipulators are restricted in confronting with unknown and confined environments and situations require high safety. To overcome these defects, the potential of bio-inspired soft robots for space applications has recently been addressed in the literature. Dynamic modeling and control of soft manipulators are challenging due to their infinite degrees of freedom. In this paper, a general comprehensive three-dimensional dynamic modeling framework is developed for a floating soft manipulator-spacecraft system employing the Euler-Lagrange method. This framework derives the coupled equations of motion adapting the generalized Jacobian approach. Consequently, the opportunity of exploiting model-based control methods will be provided. In addition, the proposed model avoids numerical singularities once the bending angles are near zero. Furthermore, the generalized Jacobian matrix is defined for any arbitrary point of the soft manipulator essential for the contact dynamics. Eventually, to verify the proposed dynamic modeling procedure, the dynamic response of a floating soft manipulator-spacecraft system released from an initial configuration is investigated. Then, a sliding mode controller is designed for the floating soft manipulator to track the desired shape and compared to the proportional-derivative control scheme. The obtained results follow the physical principles and fulfill the control objective.

Visuo-dynamic self-modelling of soft robotic systems.

Soft robots exhibit complex nonlinear dynamics with large degrees of freedom, making their modelling and control challenging. Typically, reduced-order models in time or space are used in addressing these challenges, but the resulting simplification limits soft robot control accuracy and restricts their range of motion. In this work, we introduce an end-to-end learning-based approach for fully dynamic modelling of any general robotic system that does not rely on predefined structures, learning dynamic models of the robot directly in the visual space. The generated models possess identical dimensionality to the observation space, resulting in models whose complexity is determined by the sensory system without explicitly decomposing the problem. To validate the effectiveness of our proposed method, we apply it to a fully soft robotic manipulator, and we demonstrate its applicability in controller development through an open-loop optimization-based controller. We achieve a wide range of dynamic control tasks including shape control, trajectory tracking and obstacle avoidance using a model derived from just 90 min of real-world data. Our work thus far provides the most comprehensive strategy for controlling a general soft robotic system, without constraints on the shape, properties, or dimensionality of the system.

Design, performance analysis and applications of pneumatic bellows actuator for building block soft robots

Currently, there are some new challenges for pneumatic soft robots brought by complex application environments and variable task objectives. To address these challenges, inspired by the building blocks, we develop a novel pneumatic soft actuator named pneumatic bellows actuator (PBA) that can generate bidirectional deformation, and propose a building block design concept for the soft robots based on the developed PBA. The PBA consists of three parts: a soft rubber layer (SRL), a fiber-reinforced layer (FRL) and some rigid joints. We design the SRL as a bellows-like shape so the PBA can generate bidirectional deformation, and weave an FRL as the coating of the SRL to prevent the significant radial expansion of the SRL that may affect the application of the PBA. The rigid joints include sealing glands and connectors. The sealing glands are used to seal the SRL, while the connectors having interfaces on their five surfaces are used to connect the PBAs through the customized plugs corresponding to these interfaces. Moreover, the experimental results indicate that the PBA can generate large deformation and driving force, while it has the advantages of good repeatability and approximate rate-independent under low frequency. Based on the building block design concept for the soft robots, we construct three PBA-based soft robots: a soft crawling robot, a soft gripper, and a soft manipulator, to demonstrate that we can easily use the PBAs to construct various soft robots.

Design and navigation method of a soft robot for single-port transvesical radical prostatectomy.

Currently, the rigid instruments used for laparoscopic radical resection of prostate cancer not only have the risk of damage to tissues, blood vessels, and nerves, but their limited freedom will also cause surgical blind areas. Soft robots are expected to solve these issues due to inherent flexibility, compliance, and safe interaction with tissues and organs. In addition, to achieve high surgical accuracy and provide precise guidance for surgeons, the navigation method should be studied for the soft robot. A soft robot system for single-port transvesical radical prostatectomy (STRP) is developed, and a navigation method combining fiber Bragg gratings and electromagnetic tracking is proposed for the soft robot. To validate the soft robot design and the effectiveness of the navigation method, different groups of experiments are conducted. The proposed navigation method can achieve accurate location and shape sensing of the soft manipulator. The experiments show that the maximum tip sensing error is 2.691mm, which is 5.38 of the robot length for static configurations, and that the average tip sensing error is 1.966 mm, which corresponds to 3.93 of the robot length for dynamic scenarios. Additionally, phantom tests demonstrate that the designed soft robot can enter the prostate through navigation guidance in a master-slave control mode and cover the entire prostate space. The designed soft robot system, due to its soft structure, good flexibility, and accurate navigation, is expected to improve surgical safety and precision, thereby exhibiting significant potential for STRP.

Spring-IMU Fusion-Based Proprioception for Feedback Control of Soft Manipulators

Spring-IMU Fusion-Based Proprioception for Feedback Control of Soft Manipulators

Planar soft space robotic manipulators: Dynamic modeling and control

Space robots have proven their ability to accomplish many space missions impossible or dangerous for the human. However, the conventional space robots suffer from the low adaptability, heavy structures, and the low safety. In this regard, the bio-inspired soft robots, composed of the soft material, have been emerged to overcome these shortcomings and to expand the space missions’ diversity. The potential of soft robots for space applications has recently been investigated in the literature. Although the dynamic modeling and control of soft robotic arms are complicated due to their infinite degrees of freedom, soft manipulators are anticipated to provide the adaptive capture of targets, confined space operations, and the long-distance handling of space facilities. In this regard, a general novel dynamic modeling algorithm is presented here for soft space robotic manipulators (SSRM) in a specified orientation based on the generalized Jacobian method. The proposed algorithm could be employed for different numbers of soft manipulator constant-curvature elements and avoids numerical singularities using the Taylor expansion as well. In addition, the generalized Jacobian matrix is defined to compute the velocity of any arbitrary point on the soft manipulator for contact dynamics objectives. Accuracy of the proposed dynamic model has been verified via comparison to the simulation results of the existing fixed-base soft manipulators. Moreover, performance of the proposed model has been more investigated through two numerical simulation case studies. The first case verifies that the dynamic responses of the floating SSRM follows the physical principles. While the second one addresses the configuration tracking problem of the floating SSRM exposing a computed torque control scheme. The obtained results indicate the effectiveness of the presented dynamic modeling algorithm and the proposed control scheme.

Reconfigurable Transparent Variable‐Stiffness Soft Robot for Underwater Operations

In the realm of underwater exploration and operations, soft robots exhibit considerable application potential due to their capacity for agile and complex deformations as well as their inherent compliance. These characteristics grant them excellent environmental adaptability and reduce damage to delicate samples and organisms. Nonetheless, existing underwater soft robots are primarily designed to mimic the movements of marine organisms or for specific functions, and little attention is paid to the camouflage ability. To address these challenges, in this study, a reconfigurable transparent soft robot with variable stiffness for underwater operations is presented. The design and fabrication methodology of the robot module is presented, followed by the kinematic analysis and stiffness characterization. Leveraging the proposed robot module, a soft manipulator and a soft gripper are designed for underwater operations. The soft manipulator excels in underwater pipeline detection and obstacle avoidance, while the soft gripper showcases considerable load‐bearing capacity (about 71 times its weight), making it suitable for tasks such as retrieving aquatic biological samples or garbage fishing. The proposed reconfigurable soft robot demonstrates robust camouflage capabilities, high environmental adaptability, and notable versatility, suggesting potential solutions to existing challenges in the field of underwater exploration.

A structural optimization analysis of cable-driven soft manipulator

Cable-driven soft robots hold significant potential for surgical and industrial applications, yet their performance and maneuverability can be further enhanced through design optimization. By optimizing the design, factors such as bending angles, manipulator deformation, and overall functionality can be directly influenced, leading to improved interaction with the environment and more accurate task performance. This article presents a physics-based design optimization approach for cable-driven soft robotic manipulators, aiming to enhance bending performance through structural design enhancements. Four design criteria, namely, cross-sectional shape, material, gap shape, and gap size, are considered in the optimization process. Given the inherent nonlinearity of soft materials, finite element modeling techniques are employed to analyze the effects of modifying each design parameter on displacement and bending angle. The manipulator’s design is evaluated using ABAQUS/CAE, and an analysis of variance test is conducted to identify significant performance differences among the design parameters. The results reveal that material variation has the most substantial impact, followed by gap shape and gap size. Based on subsequent parameter optimization, Dragon Skin 10 is determined to be the optimal material for bending motion, while a trapezoidal gap shape is preferred. In addition, a genetic algorithm is utilized to select a maximum gap size of 8.87 mm. These findings provide valuable insights into key design principles for cable-driven soft manipulators, aiming to enhance flexibility and reduce actuation forces. By establishing a fundamental understanding of the relationship between morphology and motion capability, this methodology demonstrates an effective simulation-driven optimization approach that incorporates the nonlinear elastic behavior of materials to improve performance. Overall, this work establishes a framework for optimizing cable-driven architectures to suit various applications in the field of soft robotics.

Learning-Based Control for Soft Robot-Environment Interaction with Force/Position Tracking Capability.

Soft robotics promises to achieve safe and efficient interactions with the environment by exploiting its inherent compliance and designing control strategies. However, effective control for the soft robot-environment interaction has been a challenging task. The challenges arise from the nonlinearity and complexity of soft robot dynamics, especially in situations where the environment is unknown and uncertainties exist, making it difficult to establish analytical models. In this study, we propose a learning-based optimal control approach as an attempt to address these challenges, which is an optimized combination of a feedforward controller based on probabilistic model predictive control and a feedback controller based on nonparametric learning methods. The approach is purely data-driven, without prior knowledge of soft robot dynamics and environment structures, and can be easily updated online to adapt to unknown environments. A theoretical analysis of the approach is provided to ensure its stability and convergence. The proposed approach enabled a soft robotic manipulator to track target positions and forces when interacting with a manikin in different cases. Moreover, comparisons with other data-driven control methods show a better performance of our approach. Overall, this work provides a viable learning-based control approach for soft robot-environment interactions with force/position tracking capability.

Driving torque model of the bionic soft arm’s hyperelastic bellows

ABSTRACT The technology of soft continuum robots represents an advance in the field of robotics to benefit a wide range of industries such as healthcare, manufacturing or environmental exploration. Soft continuum robots can be driven pneumatically by bellows as the soft continuum manipulator presented in this article. A driving torque model of the bellows considering hyperelastic material properties, friction and restoring torques is derived, whose purpose is suitability for model-based control design and subsequent trajectory generation. Hence, the driving torque model must be real-time capable. This is realized by an iterative algorithm calculating the bellows’ torque transmission by assuming a two-dimensional no-slip membrane contact. Nonlinear strain behaviours and hysteresis effects of the bellows are considered by the Ogden material model. The driving torque model’s performance is validated experimentally by measuring the external torques of the bellows.

Design and experiment of a variable stiffness soft manipulator for non-destructive grasping

Design and experiment of a variable stiffness soft manipulator for non-destructive grasping

Adaptive Online Learning and Robust 3-D Shape Servoing of Continuum and Soft Robots in Unstructured Environments.

In this article, we present a novel and generic data-driven method to servo-control the 3-D shape of continuum and soft robots based on proprioceptive sensing feedback. Developments of 3-D shape perception and control technologies are crucial for continuum and soft robots to perform tasks autonomously in surgical interventions. However, owing to the nonlinear properties of continuum robots, one main difficulty lies in the modeling of them, especially for soft robots with variable stiffness. To address this problem, we propose a versatile learning-based adaptive shape controller by leveraging proprioception of 3-D configuration from fiber Bragg grating (FBG) sensors, which can online estimate the unknown model of continuum robot against unexpected disturbances and exhibit an adaptive behavior to the unmodeled system without priori data exploration. Based on a new composite adaptation algorithm, the asymptotic convergences of the closed-loop system with learning parameters have been proven by Lyapunov theory. To validate the proposed method, we present a comprehensive experimental study using two continuum and soft robots both integrated with multicore FBGs, including a robotic-assisted colonoscope and multisection extensible soft manipulators. The results demonstrate the feasibility, adaptability, and superiority of our controller in various unstructured environments, as well as phantom experiments.

Design and Analysis of an MRI-Compatible Soft Needle Manipulator

Needle manipulation with the guidance of magnetic resonance imaging (MRI) plays a key role in minimally invasive procedures such as biopsy and ablation. However, the confined bore and strong magnetic field of the MR environment pose great challenges in developing a robotic system that fulfills the needle manipulation function. This paper presents the design and analysis of a soft needle manipulator (SoNIM) that can achieve needle manipulation in the MR environment. This pneumatically actuated manipulator consists of two bending actuators and one elongation actuator that are completely made of non-magnetic materials. These soft pneumatic actuators can generate flexible movements while maintaining a compact design, ensuring that the SoNIM is accommodated within the MRI bore. The kinematic modeling and closed-loop control of the SoNIM are investigated to achieve the position control of the needle tip. Experiments showed that the SoNIM was capable of directing the needle tip to reach the targets with a satisfactory accuracy of 2.9 ± 0.98 mm. Furthermore, the functionality and MRI compatibility of the SoNIM were validated in the clinical setting, demonstrating the capability of the SoNIM to perform needle manipulation in the MRI bore with negligible degradation to the image quality. With excellent MRI compatibility, compact design, and flexible movements, the SoNIM provides a promising solution for manipulating surgical needles in MRI-guided minimally invasive surgeries.

Soft modularized robotic arm for safe human–robot interaction based on visual and proprioceptive feedback

This study proposes a modularized soft robotic arm with integrated sensing of human touches for physical human–robot interactions. The proposed robotic arm is constructed by connecting multiple soft manipulator modules, each of which consists of three bellow-type soft actuators, pneumatic valves, and an on-board sensing and control circuit. By employing stereolithography three-dimensional (3D) printing technique, the bellow actuator is capable of incorporating embedded organogel channels in the thin wall of its body that are used for detecting human touches. The organogel thus serves as a soft interface for recognizing the intentions of the human operators, enabling the robot to interact with them while generating desired motions of the manipulator. In addition to the touch sensors, each manipulator module has compact, soft string sensors for detecting the displacements of the bellow actuators. When combined with an inertial measurement unit (IMU), the manipulator module has a capability of estimating its own pose or orientation internally. We also propose a localization method that allows us to estimate the location of the manipulator module and to acquire the 3D information of the target point in an uncontrolled environment. The proposed method uses only a single depth camera combined with a deep learning model and is thus much simpler than those of conventional motion capture systems that usually require multiple cameras in a controlled environment. Using the feedback information from the internal sensors and camera, we implemented closed-loop control algorithms to carry out tasks of reaching and grasping objects. The manipulator module shows structural robustness and the performance reliability over 5,000 cycles of repeated actuation. It shows a steady-state error and a standard deviation of 0.8 mm and 0.3 mm, respectively, using the proposed localization method and the string sensor data. We demonstrate an application example of human–robot interaction that uses human touches as triggers to pick up and manipulate target objects. The proposed soft robotic arm can be easily installed in a variety of human workspaces, since it has the ability to interact safely with humans, eliminating the need for strict control of the environments for visual perception. We believe that the proposed system has the potential to integrate soft robots into our daily lives.

A variable stiffness soft robotic manipulator based on antagonistic design of supercoiled polymer artificial muscles and shape memory alloys

This paper introduces a soft robotic manipulator with variable stiffness capability. A novel compact hybrid and antagonistic actuation method is proposed, inspired by the muscle structure of creatures/organs such as octopus and human wrist. The manipulator consists of three pairs of shape memory alloy (SMA) springs–supercoiled polymer (SCP) strings in an antagonistic configuration. Both SMA spring and SCP string are thermally actuated. When the two types of actuators are driven, they are tensioned to increase stiffness, and the variable stiffness feature of the manipulator is further amplified by the antagonistic configuration of the actuators. A kinematic modeling of the manipulator is conducted, extending the classic constant curvature approach to six-actuators case. The effectiveness of this hybrid antagonistic variable stiffness mechanism and in reducing hysteresis are demonstrated experimentally. The performance of the manipulator is further validated by integrating a three-finger gripper for grasping and an endoscope for monitoring at its end, showing the potential of the proposed design for application in minimally invasive surgery (MIS).