Design and Implementation of a Soft-Rigid Hybrid Gripper with Bionic Ligaments and Joint Capsule

Abstract

Dealing with grasping tasks in unstructured environments, existing soft grippers often exhibit a lack of static stability, while rigid-soft hybrid grippers display limited compliance due to the fixed connections at the joints. To address the challenge of balancing static stability and flexible adaptability, this study designs and implements a bioinspired hybrid gripper combining soft and rigid elements. The gripper draws inspiration from the collateral ligaments and joint capsule structures of human fingers. It employs a tendon-driven mechanism that ensures high static stability while enabling a large range of flexion movements and some degree of deflection, mimicking the dynamic bending of a human finger. Experimental results demonstrate that the hybrid fingers excel in terms of static stability, working range, and output force. Notably, under conditions of extensor tendon pretension, the fingers exhibit finer motion toward the fingertips. The dual-finger gripper performs exceptionally well in various grasping tasks, stably grasping objects of different shapes and weights, such as the Evolved Grasp Analysis Dataset and common daily items. This study offers a novel and straightforward design approach for the development of bioinspired fingers and high-performance robots, holding broad application prospects.

Introduction

The original intention behind designing humanoid robotic hands likely stems from the desire to use motorized prosthetic hands to restore the flexibility of lost hands.^1–4 Relying on their continuously evolving flexibility and load-bearing capacity, robotic hands have the potential to enter our daily lives in various forms. For instance, humanoid robotic hands can be used as part of exoskeletons to help workers carry out labor-intensive tasks, such as heavy lifting, more safely. With the advancement of artificial intelligence technologies, household robots equipped with humanoid robotic hands could better perform kitchen tasks, cleaning, and other chores.^5,6 To achieve the above goals, robotic hands need to have higher adaptability and safety. Therefore, research on soft robotic arms and rigid flexible coupled robotic arms has been widely carried out.

The flexibility of the human hand primarily derives from its unique structural composition, including the intricate shapes of bones, the coordinated actions of muscles and tendons, and the support of ligaments. Robotic fingers with a combination of rigidity and softness typically exhibit adaptive discrete or continuous kinematics and are characterized by their lightweight design.^7,8 This is beneficial for performing surface fitting and grasping tasks. A common driving mechanism employs a tendon-pulley transmission system. When a pulling force is applied to the tendon, it creates an antagonistic interaction between the tendon and the torsion springs, thereby replicating the flexion and extension movements of human fingers.⁹ By incorporating rigid frames, polymer wires, and flexible sheets, the rigid-soft finger eliminates the need for traditional revolute joints with bearings or bushings. Utilizing tendon-driven methods, the fingers achieve excellent fingertip output force and repeatability.¹⁰ Similarly, from the perspective of rigid-soft coupling, another hybrid finger with multiple modes and poses, coupled by a soft actuator and a rigid actuator in parallel, has been developed. This design achieves a wide graspable object range (ranging from 0.1 g potato chips to 27 kg dumbbells), addressing the issue of limited output force range.¹¹ Unlike current designs that can only bend in the longitudinal direction, He et al.¹² have proposed a finger based on the topological structure of the human finger, which can bend simultaneously in both the longitudinal and transverse directions, demonstrating enhanced adaptability. By increasing the complexity of the metacarpophalangeal joint, the degrees of freedom can be significantly enhanced. Zhu et al.¹³ proposed a hybrid gripper design with eight independent muscles, combining lightweight bellows soft actuators, which can achieve in-hand operations including passing a soft towel between two grippers.

For soft fingers, enhancing variable stiffness without compromising compliance and agility is a critical issue. Adjustable stiffness materials are widely used to modify the overall or local stiffness of soft fingers, such as jamming particles^14–16 and jamming layers.^17,18 By combining tendon-driven mechanisms with particle jamming, the limitations of vacuum interference can be effectively overcome, thereby further enhancing rigidity.¹⁹ Another robotic finger composed of soft actuators and an integrated layer jamming unit can freely deform under low stiffness and maintain gripping stability at high rigidity during high acceleration.¹⁸ To a certain extent, the compliance of these fingers is reduced due to the friction between high-modulus particles or layers. In addition, they cannot dynamically deform in their rigid state, resulting in decreased dexterity. To address this issue, Yang et al.²⁰ demonstrated a hybrid jamming mechanism using particles and copier paper to adjust the deformation and stiffness of joints and phalanges. In a similar manner, by combining air-tendon hybrid actuation and a bone-like structure, it provides robust lateral performance while maintaining bending compliance and independently adjusting lateral stiffness.²¹ In addition, there are methods to achieve adjustable stiffness by incorporating thermo-responsive polymers,^22–25 but this increases the complexity of control.

In previous studies, tendon-driven mechanisms have been utilized to drive joints, allowing for a reduction in the size and weight of the driving mechanism.²⁶ In addition, employing springs to passively generate tensile torque simplifies the mechanism and reduces the number of required actuators.²⁷ Internally embedded springs facilitate the interconnection of different joints, which to some extent constrains the relative positions between phalanges. Comparing rigid-flexible coupled fingers with soft fingers, it can be observed that research on soft fingers has been focused on enhancing rigidity. In contrast, rigid-soft hybrid structures, due to their fixed connecting structures, limit the relative positions between phalanges to some extent. Zhu et al.²⁸ explored the anisotropic variable joint stiffness of ligament joints and the feasible force space of fingertips expanded by the mesh extensor mechanism and designed and developed a multilayer humanoid robot finger. However, how to simplify the connection structure and coordinate the overall static stability and compliance is worth further exploration. Therefore, this study aims to connect phalanges using a flexible elastic elements, thereby endowing them with good compliance and considerable static stability.

Inspired by the biological structure of human fingers, particularly the collateral ligaments and joint capsules, we propose a novel design for a tendon-driven robotic gripper. This gripper integrates soft-rigid coupled actuators with soft elastic joint structures to better emulate the structure and function of human fingers. The proposed design ensures superior static stability through rigid-soft finger joints while allowing for a wide range of bending and slight deflection along the coronal axis, thus mimicking the dynamic bending and grasping capabilities of human fingers. This study delves into the structural design, performance, and grasping efficiency of a dual-finger gripper composed of such tendon-driven actuators. We explore the operational principles of the hybrid fingers and experimentally validate their static stability, range of motion, and output force. Furthermore, the dexterous grasping performance is demonstrated using the Evolved Grasp Analysis Dataset (EGAD) and common daily items. The ultimate goal is to advance the field of robotic grippers by providing a design that closely replicates the complex functions of human fingers, thereby broadening the potential applications of robotic hands. This article has contributions in three aspects. (1) The collateral ligaments with a higher elastic modulus ensure high static stability while allowing a large range of flexion movements and a certain degree of deflection. (2) By using flexible elastic joint capsule to connect the phalanges, the finger has good compliance. (3) Through extensor tendon preloading, finger movements are able to perform precise operations at the fingertip.

Design and Implementation

Structural design

The joints between adjacent phalanges, such as the proximal phalanx and the middle phalanx, contain soft tissues connected externally by collateral ligaments and enveloped in a joint capsule, as shown in Figure 1a. During finger movements, the collateral ligaments prevent excessive lateral motion of the joints, help maintain proper alignment, and prevent over-twisting and strain. The joint capsule is a thin film of connective tissue that wraps around the joint, sealing and securing it. It provides support to the overall structure, protects internal soft tissues from injury, and restricts unnecessary joint movement.

FIG. 1.

The structure design of the human finger and soft-rigid bionic finger. (a) Schematic diagram of the rigid-soft coupling structure of the human finger. (b) The structure diagram of the soft-rigid hybrid bionic finger. The bone-like structure is inspired by the phalanx. The tuft and base of the phalanx are simplified to a column. (c) The prototype of the flexible joints with no ligaments, parallel ligaments, or crossed ligaments. (d) The prototype of the proposed soft-rigid finger. (e) The state of joints being stretched.

Inspired by the collateral ligaments and joint capsules at human finger joints, we propose a tendon-driven actuator with a rigid-soft hybrid joint structure. This design includes tendon-driven elements that combine both rigid and soft structures. Key components, such as the joints between the proximal and middle phalanges, consist of soft tissues connected externally by collateral ligaments and enveloped by a joint capsule (Fig. 1b). By combining rigid bones with flexible collateral ligaments and joint capsules, this design closely mimics the human finger structure, allowing the finger to maintain considerable stability while possessing high flexibility, as the joints are not rigidly fixed.

In this design, the three phalanges are simplified to have the same structure. This bone-like structure is inspired by the phalanges, as illustrated in Figure 1b. Initially, the tuft and base of the phalanx were simplified into pulley joints, allowing abduction and adduction along the coronal axis. However, practical tests revealed that this structure significantly limits the range of rotation. Consequently, the tuft and base were optimized into two columns. The check-rein ligaments and collateral ligaments were unified into a single ligament, resulting in two design schemes: the parallel ligament design and the crossed ligament design, named parallel ligaments and crossed ligaments, respectively.

Figures 1c, d show the prototype of the proposed soft-rigid joints and fingers. The size values of phalanges and the fabrication process of the entire finger can be found in Section S1 of the Supplementary Data, and the material properties are listed in Supplementary Table S1 in Supplementary Data. Ligaments made from Smooth-Sil 945 are encapsulated within a joint capsule composed of Ecoflex 0050. The joint capsule secures the ligaments, preventing them from deforming arbitrarily. There is no fixed connection between adjacent phalanges, and significant gaps can easily appear under external forces (Fig. 1e). Based on the position and function of tendons (extensor tendon and flexor tendon) in the human finger joints (Fig. 2a), our design is relatively simple. There are two main tendons along the fingers, namely the flexor tendon and the extensor tendon. Therefore, we used a nonelastic thread with a diameter of 1.0 mm as the tendon and tied it to the phalanges through several auxiliary holes, as shown in Figure 2b.

FIG. 2.

The structural design of tendons. (a) Schematic diagram of ligaments and tendons in human finger joints. (b) Schematic diagram of tendon position in rigid-soft hybrid actuator.

Working principle

As shown in Figure 3a, the four phalanges can be simplified into a linkage structure, with ligaments acting as elastic elements that can store or release energy in the form of an elastic loop. The material of the extensor tendon and flexor tendon is nonelastic thread (1 mm, OUFANLUO), which is represented by the green solid lines in Figure 2. The driving tendons on the inner side of the finger act on the distal phalanx, while the driving tendons on the dorsal side only act on the middle phalanx. Two operational states of the soft-rigid finger are proposed here. As illustrated in Figure 2b, the solid black line and dashed-dotted red line frame represent the initial and final bending process states. The extensor tendons are in a free state, tightening only the flexor tendons, allowing the linkage to achieve linkage bending. In this operational state, the hybrid finger can achieve a large range of bending, with the ligaments on one side in a relaxation state and on the other in a straining state. The elastic elements in the straining state store energy and assist the finger in returning to its original position. When the extensor tendons are in a pretensioned state, the movement range of the proximal phalanx and metacarpal phalanx is significantly restricted, resulting in pronounced bending at the fingertip, as shown in Figure 2c.

FIG. 3.

Schematic diagram of the working principle. (a) The initial state of the proposed soft-rigid finger. The working principle of fingers driven by tendons is illustrated when the extensor tendons are in the (b) free state or (c) pre-tension state.

Characterization

Stability tests

To evaluate the static stability of the hybrid finger, tests were conducted with a camera (240 fps, Osmo Action, DJI) under three typical working conditions: (1) external impact, (2) gravity, and (3) initial bending state. A soft-rigid finger without ligaments was used for comparison.

We first conducted external tensile force tests by pulling the end of a single joint using a force gauge, with the force directions divided into vertical downward and horizontal, as shown in (Fig. 4a, b). The results indicate that when pulled horizontally or vertically, the deflection angle of the hybrid finger increases linearly with the tensile force. The growth rate of the hybrid finger with parallel ligaments and crossed ligaments is similar, with the deflection angle of the single joint reaching approximately 20° at a horizontal tensile force of 2 N. In contrast, the hybrid finger without ligaments exhibits a deflection angle of approximately 33° under the tensile force of 1N. When facing the same external influences, ligaments can significantly reduce and weaken these effects compared to hybrid fingers with only a joint capsule.

FIG. 4.

Stability tests of hybrid fingers with or without ligaments under external pull impact, gravity, and initial bending angle. Deformation of hybrid fingers under (a) a horizontal tensile force or (b) a vertical tensile force. (c) Deformation of hybrid fingers under gravity parallel to their symmetry plane. (d) Vibration dissipation process of hybrid fingers at initial bends of 90 degrees inward and 50 degrees lateral.

We then conducted tests on changes in the direction of gravity by allowing two fingers to freely deform under two typical gravity orientations. The results indicate that the hybrid finger with parallel ligaments can maintain its initial shape well under changes in the direction of gravity, whereas the finger without ligaments exhibits significant drooping. The vertical displacement of the fingertip was 23.1 mm (17.9% of its body length) and 16.4 mm (12.7% of its body length), respectively, as shown in Figure 4c.

Finally, we observed the performance of the hybrid finger in returning to its original state by bending it inward to 90° or laterally to 50°, as shown in Figure 4d and Supplementary Movie S1. The finger without ligaments approximately returns to its initial state within about 1.5 s, and the amplitude is also somewhat smaller compared to the finger without ligaments. During this process, the ligaments with lower elastic modulus can effectively dampen vibrations, showing better stability in response to external impacts.

Bending and backward tilting performance

This section evaluated the bending and backward tilting capabilities of the proposed actuators, as depicted in Figure 5. With the increase of the front tension, when the preload of the extensor tendon is 0 N or 5 N, the bending angles of the proposed actuator with parallel or crossed ligaments are similar, and both are significantly smaller than those without ligaments. Moreover, when the front tension reaches 10 N, the bending angle is between 60° and 70°. When the rear tension is applied and the preload of the flexor tension is 0 N, the backward tilt angle of the actuator can reach approximately−25° to−28°. This suggests that the actuator design and the interaction of different forces play a crucial role in determining the resulting angular displacements, which could have implications for the actuator’s performance. The details of bending and backward tilting performance can refer to Supplementary Figure S2, S3 and S4 in Supplementary Data and Supplementary Movie S1.

FIG. 5.

Bending and backward tilting performance of the soft-rigid finger with ligaments. (a) The process of finger flexion and backward tilt under the condition of rear preload. The bending degree of the proposed finger when (b) the preload is 0 N and the front tension is gradually applied; (c) the preload increases from 0 N to 5 N; (d) the preload is 5 N and the front tension is gradually applied.

Analysis of motion range

Figure 6 illustrates the working range of the hybrid finger under two operational states, actuated by the anterior tendons. By recording the continuous positions of three tracking points on the soft-rigid finger, as shown in Supplementary Figure S6 in Supplementary Data, the working range is depicted in Figure 6a when the extensor tendons are in a free state. At this time, the position coordinates of the fingertip are approximately (105.7, 63.8), and the bending angle is about 100°. Figure 6b shows the state when the initial pretension of the posterior tendons is 5 N, the bending angle of the fingertip is approximately 35°, and the position coordinates are about (11.9, 143.8). It can be seen that when the extensor tendons are pre-tensioned, the motion of the finger tends more towards the movement of the fingertip, while the movement of the lower part of the finger, including the metacarpal phalanx and the proximal phalanx, is somewhat restricted.

FIG. 6.

Workspace generated from the end trajectories of the distal phalanx, middle phalanx, and proximal phalanx. (a) Workspace when there is no posterior pretension. (b) Workspace when the posterior load is 5 N.

Output force

We conducted experiments to verify the active driving effects of anterior and extensor tendons on the soft-rigid finger. The experimental setup is shown in Figure 7a. The soft-rigid finger was horizontally fixed, with the fingertip in contact with a force sensor (LCM6, MANTON). The servo motor was controlled by a wireless control board, and the collected signals were acquired through the signal amplifier (FC400, Unipulse). Figure 7b, c present the comparison of output force results for fingers in horizontal and vertical direction. Here, the horizontal axis represents the delay time of the servo motor. Therefore, we compared the relationship between delay time (in the initial approximately 1250 ms) and applied load. For tendon drive, there is a positive correlation between the output force and the delay time. At a delay of 1000 ms, the horizontal output forces of the cross ligaments and parallel ligaments are 2.24 N and 1.87 N, while their vertical downward output forces reach 2.49 N and 2.10 N. This may be partly due to the influence of finger gravity. It can be observed that the output force of crossed ligaments is slightly greater than that of parallel ligaments, and both are significantly higher than that of no ligaments. The above indicates that the presence of ligaments has a positive impact on the output force and subsequent grasping tasks.

FIG. 7.

Output force test of the hybrid finger in two poses. (a) Experimental platforms to evaluate the output force of the finger and the architecture of the controller board. Comparison of output force results for fingers with and without ligaments in (b) horizontal and (c) vertical directions.

Grasping performance

To further demonstrate the performance of the finger, we assembled these hybrid fingers into a two-finger gripper using several frameworks. The grasping performance of the two-finger gripper was evaluated by grasping a series of objects from the EGAD.²⁹ Compared with other robotic grasping datasets, the objects in EGAD have diverse geometries and varying levels of complexity, ranging from simple to difficult to grasp. Figure 8a shows the 49 3D-printed evaluation objects selected from over 2000 diverse objects in EGAD. These objects provide a range of geometries from simple to complex (left to right) and graspability from easy to difficult (bottom to top). The proposed dual-finger gripper was able to achieve stable grasping for all 49 targets. Figure 8b showcases the grasping results for the 24 objects with higher grasp difficulty, and the others are shown in Supplementary Figure S7 of the Supplementary Data. Notably, we successfully grasped objects by contacting only smaller surfaces or protrusions, as highlighted by the images within the red dashed borders.

FIG. 8.

Forty-nine 3D-printed evaluation objects in EGAD and the performance of the proposed dual finger gripper in grasping partial objects (G0-G6, F0-F6, E0-E6, D0-D2). EGAD, Evolved Grasp Analysis Dataset.

We found that for a dual-finger gripper, the coordinated operation of the dual-finger gripper can be considered (Supplementary Movie S3). As shown in Figure 9a, the E4 object is grasped through the bilateral grasping mode, where both fingers work simultaneously. If only one finger is operating, that is, in unilateral grasping mode, the object (D4) can be pressed against the other side to achieve stable grasping as well (Fig. 9b). Next, we further installed the dual-finger gripper and the wireless control system on a robotic arm to evaluate the stability of grasping performance through multiple grasping. The details of the process can refer to Supplementary Movie S4. It can be seen from Figure 9c, d that in the bilateral grasping mode, the dual-finger gripper can achieve a greater number of consecutive successful grasps than in the unilateral grasping mode. Moreover, as the shape of the target object becomes more complex, the number of consecutive grasps decreases accordingly. In the unilateral grasping mode, the success of grasping depends more on the close fit between the object and the finger. Therefore, the target object with a flatter surface and a slightly larger size shows a better grasping effect.

FIG. 9.

The grasping performance of the dual-finger gripper through (a) bilateral grasping mode and (b) unilateral grasping mode. Multiple grasping tests process and results of the dual-finger gripper through (c) bilateral grasping mode and (d) unilateral grasping mode.

We conducted grasping tests on common daily items (Fig. 10). The first explored performance metric was the range of objects that could be grasped. By flexibly coordinating the anterior and extensor tendons, the range of motion at the fingertips can be selectively highlighted. Without the extensor tendons working, the 401.4 g (including servos) dual-finger gripper can safely grasp small items such as a ring shim (0.4 g), a piece of potato chip (1.6 g), a bolt (2.3 g), or a coin (13.0 g), as well as medium-sized items such as a badminton (3.5 g), a table tennis ball (2.8 g), a tennis ball (63.3 g), an orange (28.6 g), a roll of plastic bags (79 g), or even a chip container (146.9 g). Correspondingly, when posterior preload is applied, the gripper first opens to an appropriate angle and then maintains the grasp on different objects by continuously applying the posterior preload. This includes larger objects with approximately cylindrical or spherical shapes, such as a box of cookies (200 g), a roll of sponge tape (68 g), a balloon (2.7 g), or a wide-rimmed bowl (214.3 g). In addition, we tested the maximum weight that the dual-finger gripper can withstand, which is approximately 718 g, as shown in the bottom right corner of Figure 10 and Supplementary Figure S8 in Supplementary Data.

FIG. 10.

Dexterous grasping of common objects in daily life.

Conclusions

This study proposes a design for a hybrid bionic gripper based on tendon-driven mechanisms, inspired by the biological structure of human fingers, particularly the collateral ligaments and joint capsules. Through this design, the gripper maintains high static stability while offering a large range of flexion-extension motion and certain deflection capabilities, thereby better mimicking the dynamic bending of a human hand. In addition, the hybrid structure of the fingers has proven to be highly stable under gravity and external stimuli. The key to this design is the combination of rigid finger joints with flexible joint structures, utilizing tendon-driven mechanisms to achieve complex and precise finger movements.

Based on the fingers, a dual-finger gripper can be easily assembled. Experimental results show that the rigid-soft coupled actuators with flexible joint structures perform excellently in terms of working range, output force, and gripping performance. Particularly with extensor tendon preload, the finger movements tend to enable precise operations at the fingertips, further enhancing the gripper’s efficiency and flexibility. In experimental grasping tasks targeting the EGAD object set and common household items, the gripper demonstrated the ability to stably grasp objects of various shapes and weights. For example, it can grasp a bottle weighing up to 718 grams and a ring shim as light as 0.4 g, showcasing its wide applicability. Meanwhile, the gripper exhibited excellent high-grasp compliance (capable of accommodating objects up to 147% of the finger span size) and high stability under gravity and external stimuli. Furthermore, thanks to the flexible structural design, even with one finger working alone, objects can be stably secured within the inner side of the other finger, thus further enhancing the overall stability of the gripping system.

In summary, this study provides an innovative and straightforward design concept, offering a reference for the development of subsequent bionic fingers and high-performance robots. This design holds potential applications in several fields, including prosthetics, food processing, industrial sorting, and biological sample collection, demonstrating broad application prospects. With further research and optimization, this hybrid bionic gripper is expected to significantly improve the practical application capabilities of robots and prosthetics in various complex tasks.

Footnotes

Authors’ Contributions

T.H.: Conceptualization, validation and writing—original draft. Y.M.: Supervision. J.F.: Funding acquisition. S.L.: Methodology. J.W.: Project administration and writing—review and editing. All the authors approved the final article.

Data Availability Statement

All data that support the findings of this study are included in this article and its .

Author Disclosure Statement

All the authors have no conflict of interests.

Funding Information

This work was supported by the National Natural Science Foundation of China (No.12072233).

Supplementary Materials

References

, Zhang

, Xu

, et al. A soft neuroprosthetic hand providing simultaneous myoelectric control and tactile feedback. Nat Biomed Eng, 2023; 7(4):589–598; doi: 10.1038/s41551-021-00767-0

Zhou

, Tawk

, Alici

. A 3D printed soft robotic hand with embedded soft sensors for direct transition between hand gestures and improved grasping quality and diversity. IEEE Trans Neural Syst Rehabil Eng, 2022; 30:550–558; doi: 10.1109/TNSRE.2022.3156116

D’Anna

, Valle

, Mazzoni

, et al. A closed-loop hand prosthesis with simultaneous intraneural tactile and position feedback. Sci Robot, 2019; 4(27):eaau8892; doi: 10.1126/scirobotics.aau8892

, Zhang

, Chen

, et al. Soft robotics enables neuroprosthetic hand design. ACS Nano, 2023; 17(11):9661–9672; doi: 10.1021/acsnano.3c01474

Peter

, Florence

, Tedrake

. Dense object nets: Learning dense visual object descriptors by and for robotic manipulation. arXiv, 2018; doi: 10.48550/arXiv.1806.087561806.08756

, Zhao

, Finn

. Mobile ALOHA: Learning bimanual mobile manipulation with low-cost whole-body teleoperation. arXiv, 2024; doi: 10.48550/arXiv.2401.021172401.02117,

Zhou

, Zhang

, Gu

. A biomimetic soft-rigid hybrid finger with autonomous lateral stiffness enhancement. Advanced Intelligent Systems, 2022; 4(12):2200170; doi: 10.1002/aisy.202200170

, Cong

, Liu

, et al. Development of a novel robotic hand with soft materials and rigid structures. IR, 2021; 48(6):823–835; doi: 10.1108/IR-01-2021-0013

, Cong

, Liu

, et al. A practical model of hybrid robotic hands for grasping applications based on bioinspired form. J Intell Robot Syst, 2022; 105(4):73; doi: 10.1007/s10846-022-01569-5

10.

Kim

, Yoon

, Sim

. Fluid lubricated dexterous finger mechanism for human-like impact absorbing capability. IEEE Robot Autom Lett, 2019; 4(4):3971–3978; doi: 10.1109/LRA.2019.2929988

11.

Zhu

, Chai

, Yong

, et al. Bioinspired multimodal multipose hybrid fingers for wide-range force, compliant, and stable grasping. Soft Robot, 2023; 10(1):30–39; doi: 10.1089/soro.2021.0126

12.

, Lian

, Song

. Rigid-Soft coupled robotic gripper for adaptable grasping. J Bionic Eng, 2023; 20(6):2601–2618; doi: 10.1007/s42235-023-00405-2

13.

Zhu

, Lu

, Zheng

, et al. A soft-rigid hybrid gripper with lateral compliance and dexterous in-hand manipulation. IEEE/ASME Trans Mechatron, 2023; 28(1):104–115; doi: 10.1109/TMECH.2022.3195985

14.

Wei

, Chen

, Ren

, et al. A novel, variable stiffness robotic gripper based on integrated soft actuating and particle jamming. Soft Robotics, 2016; 3(3):134–143; doi: 10.1089/soro.2016.0027

15.

, Chen

, Yang

, et al. Passive particle jamming and its stiffening of soft robotic grippers. IEEE Trans Robot, 2017; 33(2):446–455; doi: 10.1109/TRO.2016.2636899

16.

Zhou

, Chen

, Hu

, et al. Adaptive variable stiffness particle phalange for robust and durable robotic grasping. Soft Robot, 2020; 7(6):743–757; doi: 10.1089/soro.2019.0089

17.

Wang

, Zhang

, Li

, et al. Electrostatic layer jamming variable stiffness for soft robotics. IEEE/ASME Trans Mechatron, 2019; 24(2):424–433; doi: 10.1109/TMECH.2019.2893480

18.

Yan

, Xu

, Shi

, et al. A human-inspired soft finger with dual-mode morphing enabled by variable stiffness mechanism. Soft Robot, 2022; 9(2):399–411; doi: 10.1089/soro.2020.0153

19.

Jiang

, Chen

, Liu

, et al. Chain-like granular jamming: A novel stiffness-programmable mechanism for soft robotics. Soft Robot, 2019; 6(1):118–132; doi: 10.1089/soro.2018.0005

20.

Yang

, Zhang

, Kan

, et al. Hybrid jamming for bioinspired soft robotic fingers. Soft Robot, 2020; 7(3):292–308; doi: 10.1089/soro.2019.0093

21.

Lin

, Xiao

, Li

, et al. A bioinspired bidirectional stiffening soft actuator for multimodal, compliant, and robust grasp. arXiv, 2022; doi: 10.48550/arXiv.2211.11978and 2211.11978,

22.

Yang

, Chen

, Li

, et al. Bioinspired robotic fingers based on pneumatic actuator and 3D printing of smart material. Soft Robot, 2017; 4(2):147–162; doi: 10.1089soro.2016.0034

23.

Firouzeh

, Salerno

, Paik

. Stiffness control with shape memory polymer in underactuated robotic origamis. IEEE Trans Robot, 2017; 33(4):765–777; doi: 10.1109/TRO.2017.2692266

24.

Nasab

, Sabzehzar

, Tatari

, et al. A soft gripper with rigidity tunable elastomer strips as ligaments. Soft Robot, 2017; 4(4):411–420; doi: 10.1089/soro.2016.0039

25.

Alambeigi

, Seifabadi

, Armand

. A continuum manipulator with phase changing alloy. In: 2016 IEEE International Conference on Robotics and Automation (ICRA), Stockholm, Sweden. 2016, pp. 758–764; doi: 10.1109/ICRA.2016.7487204

26.

Chang

C-M

, Gerez

, Elangovan

, et al. On alternative uses of structural compliance for the development of adaptive robot grippers and hands. Front Neurorobot, 2019; 13:91; doi: 10.3389/fnbot.2019.00091

27.

McLaren

, Fitzgerald

, Gao

, et al. A Passive Closing, Tendon Driven, Adaptive Robot Hand for Ultra-Fast, Aerial Grasping and Perching. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Macau, China, 2019, pp. 5602–5607; doi: 10.1109/IROS40897.2019.8968076

28.

Zhu

, Wei

, Ren

, et al. An anthropomorphic robotic finger with innate human-finger-like biomechanical advantages Part I: Design, ligamentous joint, and extensor mechanism. IEEE Trans Robot, 2023; 39(1):485–504; doi: 10.1109/TRO.2022.3200006

29.

Morrison

, Corke

, Leitner

. EGAD! An Evolved Grasping Analysis Dataset for Diversity and Reproducibility in Robotic Manipulation. IEEE Robot Autom Lett, 2020; 5(3):4368–4375; doi: 10.1109/LRA.2020.2992195

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

1.12 MB

36.22 MB

36.67 MB

31.60 MB

52.12 MB

0.00 MB