Design and Experimental Validation of a 3D-Printed Hybrid Soft Robotic Gripper for Delicate Object Manipulation

Abstract

This work presents a novel soft gripper concept featuring integrated force feedback and a compact, resource-efficient geometry. The gripper is designed to provide a low-cost, adaptable, and precise solution for manipulating delicate and irregularly shaped objects. By embedding force feedback directly into the structure, the system reliably detects contact and enables controlled, gentle gripping of fragile items. The design was developed for collaborative and assistive robotic applications, where safety and human–robot interaction are prioritized. The prototype is fabricated using consumer-grade 3D-printed components and employs a simple cable-driven actuation system. The hybrid soft–rigid architecture combines compliant fingers with a rigid, sensorized thumb, preserving the adaptive grasping characteristics of soft robotics while simplifying sensing integration and construction. A motor-based control mechanism synchronizes finger motion through cable traction, ensuring reliable and repeatable performance. Experimental evaluations demonstrate secure, damage-free handling across diverse object types, highlighting the gripper’s potential in assistive robotics, cobot environments, biomedical contexts, and other domains requiring safe and delicate manipulation.

1. Introduction

Soft robotics has emerged as a promising alternative to traditional rigid manipulators [1], offering flexibility, safety, and adaptability in scenarios where conventional grippers often fail [2,3]. In particular, soft robotic grippers have gained attention for their potential in industrial automation [4], biomedical contexts [5,6], and service robotics. Their main advantage lies in handling fragile or irregular objects without complex sensing or high-precision control [7]. While these characteristics make soft grippers highly attractive, achieving reliable performance still requires balancing adaptability and dexterity with sufficient gripping force, repeatability, and durability [8]. To address these challenges, four major actuation strategies have been explored [3]: (i) shape-memory alloys (SMAs); (ii) electroactive polymers (EAPs); (iii) fluidic actuation, encompassing pneumatic and hydraulic systems; and (iv) tendon- or cable-driven mechanisms.

Although these actuation families have enabled a variety of soft robotic prototypes, each presents significant drawbacks [9]. Pneumatic systems offer high adaptability and large deformations, but they rely on external pumps and valves that hinder compact integration and portability [10]. SMAs allow compact integration with electrical input, yet their actuation speed is constrained by thermal cycles and their force output remains modest [11,12]. EAPs provide fast response and large deformation, making them attractive for biomedical micromanipulation, but their reliance on high driving voltages and limited durability under cyclic use restrict broader application [13]. These limitations highlight the need for alternative actuation strategies capable of combining structural simplicity with reliable performance.

In light of these trade-offs, tendon-driven systems emerge as a compelling alternative. By transmitting force through lightweight tendons and guiding pulleys, cable actuation offers accurate motion control, low mechanical complexity [14], and straightforward miniaturization into compact robotic platforms [8]. Despite these advantages, existing studies often remain at the proof-of-concept stage, with limited attention to repeatability, mechanical wear, or scalability.

In recent years, hybrid soft-rigid grippers have emerged as a promising design paradigm that combines the advantages of both compliant and rigid structures [15]. By integrating rigid components into soft gripper architectures, these hybrid systems can overcome inherent limitations of purely soft designs—such as low output force and limited controllability—while preserving the compliance, adaptability, and safe interaction characteristics essential for delicate manipulation [16]. This soft-rigid coupling approach enables simplified sensing integration, improved structural stability, and enhanced grasping performance without sacrificing the gentle, shock-absorbing behavior provided by compliant elements.

Following this rationale, the present work adopts a hybrid soft-rigid architecture in which a compliant cable-driven actuation system is combined with a rigid, sensorized thumb, achieving a balance between adaptability and functional simplicity.

This configuration retains the key advantages of soft grippers—safe interaction, compliance, and adaptability—while simplifying mechanical construction, enabling straightforward sensor integration, and providing a stable reference surface for force measurement. By leveraging recent advances in additive manufacturing, the proposed gripper demonstrates affordability, reproducibility across multiple prototyping cycles, and renewed potential for accessible and scalable soft robotic applications.

To validate these contributions, the remainder of this paper is structured as follows. Section 2 reviews the state of the art in soft robotic grippers, with particular emphasis on cable-driven mechanisms. Section 3 presents the design rationale of the proposed soft gripper, covering the fabrication process with particular attention to additive manufacturing, and the actuation hardware and control system. Section 4 reports the experimental setup and validation tests. Section 5 discusses the results in the context of current soft robotic gripper technologies. Finally, Section 6 concludes the paper, summarizing the key findings.

2. Related Work

In order to identify effective design strategies for tendon- or cable-actuated soft grippers, numerous representative works from the literature have been analyzed and reviewed, providing the basis for the subsequent analysis. Figure 1 illustrates a typical configuration in which a cable embedded in the soft body is fixed at one end and actuated at the other to generate contraction or bending motion [14]. Building on this principle, Manti et al. [17] developed a bioinspired soft gripper with a cable-driven actuation mechanism powered by motors, as shown in Figure 2a, demonstrating versatile grasping of diverse objects. Their dual-material approach (Dragon Skin 30 for flexibility and Smooth-Sil 950 for stiffness) optimized adaptability while maintaining stability, although the reliance on multiple custom silicones increased fabrication complexity.

More recently, Zhang et al. [18] presented a continuum manipulator actuated by cables and designed through topological optimization; its configuration is shown in Figure 2d. Their experimental validation confirmed the potential of such strategies for complex object manipulation, although the fabrication steps remained highly specialized. Similarly, the SIMBA platform [19], a tendon-driven modular continuum arm equipped with a reconfigurable soft gripper, demonstrated high adaptability and robustness in unstructured environments. Together, these works highlight how continuum manipulators and modular tendon-driven architectures can extend the versatility of soft robotics, but also reveal that specialized fabrication and system complexity often limit broader adoption.

Figure 2. Representative examples of cable-driven soft grippers reported in the literature. (a) Bioinspired gripper with dual-material fingers, integrated into a rigid frame for adaptable and stable grasping [17]. (b) SpiRob gripper 3D-printed in TPU, leveraging a logarithmic spiral design for scalable, adaptive wrapping grasps [20]. (c) Topology-optimized gripper 3D-printed in TPE, enabling lightweight structure and adaptable grasp [21]. (d) Cable-driven continuum manipulator with a topology-optimized TPU gripper, capable of adaptive grasps from a few grams to 500 g [18]. (e) Gripper with integrated triboelectric nanogenerators for self-powered sensing of pressure and bending [22]. (f) Multi-joint cable-driven soft gripper with viscoelastic joint modeling for improved control [23]. (g) Compact 3D-printed gripper with monolithic fingers and integrated actuation [24].

Figure 2. Representative examples of cable-driven soft grippers reported in the literature. (a) Bioinspired gripper with dual-material fingers, integrated into a rigid frame for adaptable and stable grasping [17]. (b) SpiRob gripper 3D-printed in TPU, leveraging a logarithmic spiral design for scalable, adaptive wrapping grasps [20]. (c) Topology-optimized gripper 3D-printed in TPE, enabling lightweight structure and adaptable grasp [21]. (d) Cable-driven continuum manipulator with a topology-optimized TPU gripper, capable of adaptive grasps from a few grams to 500 g [18]. (e) Gripper with integrated triboelectric nanogenerators for self-powered sensing of pressure and bending [22]. (f) Multi-joint cable-driven soft gripper with viscoelastic joint modeling for improved control [23]. (g) Compact 3D-printed gripper with monolithic fingers and integrated actuation [24].

Subsequent efforts have focused on extending performance and functionality in tendon-driven soft grippers. For example, triboelectric nanogenerators (TENGs) have been integrated into tendon-driven actuators [22], enabling self-powered sensing of pressure and bending to enhance interaction with objects, as illustrated in Figure 2e. In a complementary direction, Chen et al. [21] proposed a topology-optimized cable-driven gripper designed to maximize deflection at contact points while preserving structural stiffness. Manufactured via 3D printing, this prototype (depicted in Figure 2c) successfully grasped objects of varying shapes and withstood loads up to 1 kg, yet issues related to long-term durability and repeatability were not systematically addressed. Similarly, Hussain et al. [25] introduced a modular tendon-driven gripper with soft–rigid phalanges, where flexible joints were fabricated in thermoplastic polyurethane (TPU) using additive manufacturing. By tuning printing parameters such as infill density, they were able to regulate joint compliance. Their screw-theory model, validated experimentally, confirmed the potential of additive manufacturing for customizable tendon-driven grippers. In line with these advances, Spirob [20] demonstrated the feasibility of low-cost 3D-printed tendon-driven grippers tailored to specific tasks, reinforcing the role of additive manufacturing as a practical pathway for accessible and reproducible prototyping, as illustrated in Figure 2b.

Alongside these efforts on tendon-driven mechanisms, the development of 3D-printed pneumatic actuators has also shown remarkable progress. High-force soft actuators fabricated via fused deposition modeling (FDM) have demonstrated robust grasping with simple geometric designs [26], while pneumatic sensing chambers directly 3D-printed in TPU enabled multimodal deformation sensing [24]. Other studies have introduced 3D-printed soft vacuum actuators capable of linear motion and high force output, emphasizing the potential of vacuum-driven pneumatic systems for soft robotics [27]. More recently, Hiremath et al. [10] presented a 3D-printed pneumatic actuator based on a semi-oval groove geometry in TPU, combining finite element modeling and experimental validation to demonstrate high compliance and adaptability under load. Beyond these recent developments in 3D-printed pneumatics, other pneumatic gripper designs have also contributed to the development of finger architectures that directly inspired our final design. Hao et al. [28] presented a four-fingered pneumatic gripper in which the soft silicone fingers were anchored to a 3D-printed rigid base through dedicated connectors (Figure 3a), enabling robust integration while preserving compliance. In a different approach, Zhu et al. [29] developed a three-fingered pneumatic gripper using 3D-printed molds, where the silicone fingers were mounted onto a rigid support via a snap-lock connector (Figure 3b), facilitating quick assembly. These studies were especially valuable to our work because they highlight practical connector strategies, offering insights into how soft and rigid components can be effectively integrated.

Altogether, these pneumatic grippers illustrate how modular geometries and hybrid connections between soft fingers and rigid structures can expand the versatility of soft robotics. Yet, despite their valuable role as inspiration, their dependence on external air systems limits their portability and broader adoption. By contrast, tendon-driven approaches, particularly when combined with additive manufacturing, emerge as a more compact and scalable approach, directly motivating the design strategy adopted in this work.

3. Soft Gripper Design and Fabrication

This section presents the design and fabrication process of the proposed soft gripper, focusing on the finger architecture, material selection, and assembly strategy. The design emphasizes a hybrid structure that integrates rigid and compliant elements, with careful hinge configuration and material distribution to ensure both durability and repeatable motion. Additive manufacturing plays a central role in this approach, enabling rapid prototyping of functional components with tailored flexibility, strength, and mechanical integration. This strategy allows the development of a gripper that is adaptable to diverse object geometries while maintaining robustness for continuous operation.

3.1. Finger Architecture and Geometry

The finger architecture of the proposed gripper (Figure 4) was developed to maximize adaptability to diverse object shapes while ensuring mechanical reliability. The hybrid design strategically combines rigidity and compliance to achieve both stability and adaptability—two properties often competing in conventional designs.

A rigid thumb-like component provides a fixed, monolithic opposition surface that ensures reliable force distribution and stable pinching during grasping. In addition to structural stability, the rigid element reduces the number of actively controlled parts, resulting in a simpler and more robust mechanism compared with fully compliant configurations.

An additional feature of the rigid thumb component is the integration of a load cell to provide direct force feedback. The sensing element consists of a standard aluminum bar-type load cell, commonly used in weighing scales. This generic component was selected for its low cost, robustness, and ease of integration; however, the design accommodates any bar-type load cell or even a custom metal element instrumented with a strain gauge. The use of a rigid thumb structure simplifies sensing integration without compromising the soft grasping properties of the system: shock absorption and compliance are provided entirely by the flexible fingers, while the rigid opposing surface enables stable and repeatable force detection for threshold-based control. This decoupling of sensing from the compliant elements simplifies construction and signal acquisition while preserving the adaptive characteristics inherent to soft robotic grippers.

Complementing the rigid thumb, two compliant segmented fingers actuated by cable transmission provide the adaptability needed to conform to irregular object geometries. The key dimensions and specifications of the finger design are summarized in

Table 1. Compliant finger specifications.

Parameter	Value
Number of compliant fingers	2
Number of segments per finger	4
Cable guiding elements per finger	7

. This dual-structure approach leverages the strengths of both rigid and soft elements: precise, repeatable pinching motions are supported alongside encompassing grasps, broadening the range of achievable manipulation tasks.

Each flexible finger consists of four sequential segments connected by hinge regions of reduced thickness, enabling smooth and continuous bending under cable traction.

Concentrating deformation within these compliant joints is intended to reduce material fatigue in the rest of the structure and promote durability. The segmented geometry also improves adaptability, allowing the fingers to effectively wrap around irregular objects while generating sufficient restoring forces for reliable release. To further optimize performance, the phalanges were dimensioned with gradually decreasing size toward the distal end, reducing bending resistance in the final segments and promoting smoother overall motion (Figure 5). The detailed dimensions of the prototype finger are provided in Figure 6 and Table 1; these parameters balance compactness, flexibility, and structural robustness, though the design is inherently scalable for different applications. The overall assembly dimensions are illustrated in Figure 7 (side view) and Figure 8 (top view). This combination of segmental geometry and tapered design establishes a flexible yet robust architecture designed for precise, repeatable, and fatigue-tolerant motion.

Cable routing was also improved to overcome common limitations of existing soft gripper architectures. Unlike the design reported in [20], where actuation cables are embedded within narrow internal channels, the present configuration employs seven independent guiding elements mounted externally on each segment (Figure 5). This approach avoids threading cables through confined passages, significantly simplifying assembly and enabling quick maintenance. The guiding pieces, fastened with miniature screws, ensure stable routing while remaining easily replaceable—an often-overlooked feature in comparable designs. Furthermore, unlike [17], where cables are embedded during silicone molding and any failure requires fabricating an entirely new finger, the proposed solution decouples the cable from the finger body. This separation allows rapid, low-cost maintenance without compromising structural integrity. The cable remains locally exposed yet securely constrained, minimizing friction, preventing misalignment, and ensuring stable actuation. A final guiding element at the fingertip locks the cable path, eliminating slippage and guaranteeing repeatable performance. Collectively, these refinements establish a routing system that is simpler, more reliable, and more sustainable than those used in conventional soft grippers. The combination of nylon cables and PLA guiding elements provides adequate wear resistance for the low actuation loads in this application. For extended operational lifetimes, the guiding elements could alternatively be fabricated from brass or low-friction engineering plastics.

In addition to these structural improvements, two system-level features further distinguish the proposed design. First, the actuation system is relatively compact and lightweight compared with other soft gripper prototypes, while remaining highly scalable. The same architecture can be readily adapted to larger configurations, extending its applicability from small assistive devices to more capable collaborative robotic platforms. Second, the cable-driven system is configured such that the motor actively opens the fingers, while closing relies on the passive elastic restoring force of the pre-curved fingers. This arrangement provides intrinsic safety through a force-limited design: since the motor only acts to open the fingers and never contributes to the gripping force, the maximum grip strength is inherently bounded by the elastic energy stored in the pre-curved TPU structure. Unlike systems where motor failure could cause uncontrolled clamping, any failure in this configuration results in a grip force that cannot exceed the passive elastic restoring force—a predictable and limited value determined by the material properties and geometry.

Two further refinements enhance both structural integrity and functional performance. First, the finger base incorporates a dedicated support piece (Figure 9), which ensures precise alignment with the structure, facilitates assembly, and increases robustness. Second, the fingertip adopts a trapezoidal profile instead of the rounded tips commonly used in soft grippers. During iterative prototyping, an earlier version with a rounded fingertip profile exhibited inconsistent contact and reduced stability during preliminary grasping tests, leading to its replacement with the trapezoidal geometry. The trapezoidal profile was selected empirically from several geometries tested during iterative prototyping, as it provided the most consistent contact behavior in preliminary trials (Figure 10). This choice aligns with previous work that explored trapezoidal fingertip geometries [30], though no analytical optimization was performed in the present study. As a result, the configuration achieves stable and repeatable actuation, establishing a solid foundation for the performance evaluation presented in Section 4.

3.2. Material Selection and Additive Manufacturing

The fabrication of the gripper combined two thermoplastics with complementary properties in order to balance compliance and structural rigidity (see

Table 2. Materials used in gripper fabrication.

Component	Material
Compliant fingers	TPU (thermoplastic polyurethane)
Rigid thumb	PLA (polylactic acid)
Cable guiding elements	PLA
Support structure	PLA
Actuation cables	nylon

). Thermoplastic polyurethane (TPU) was employed exclusively for the flexible fingers, while polylactic acid (PLA) was selected for the rigid finger, cable-guiding elements, and the connector structure. Specifically, TPU with 90A Shore hardness was used for the compliant fingers, and generic PLA filament (Bambu Lab, Shenzhen, China) was used for rigid components. The specific material grades are not critical for the core functionality; different material hardness values will produce different bending behaviors and may require minor geometry adjustments to achieve comparable performance. TPU was chosen due to its elastomeric nature, which provides high elasticity, resilience to fatigue, and the ability to withstand repeated bending without permanent deformation [31]. These characteristics make TPU particularly suited for soft robotic actuators, where compliant motion and long-term durability under cyclic loading are required. In contrast, PLA was used for the structural components because of its stiffness [32], dimensional stability, low cost, and ease of printing with fused deposition modeling (FDM). Despite its inherent brittleness compared to other polymers, PLA ensures geometric accuracy and robust interfaces, which are critical for stable cable transmission and mechanical integration. For the cable guiding elements specifically, PLA provides adequate wear resistance given the low actuation loads involved; however, for applications requiring extended operational lifetimes or higher duty cycles, alternative materials such as brass or low-friction engineering plastics (e.g., PTFE, acetal) could be substituted without modifying the overall design.

Additive manufacturing was central to the development of the gripper, as it enabled the direct printing of both compliant and rigid parts within the system [33]. In particular, the rigid finger, the cable-guiding elements, and the connector structure were all produced in PLA during the same process as the TPU fingers, eliminating the need for additional machining or external fittings. This integration reduced assembly complexity, minimized potential failure points, and facilitated cost-effective iteration of the design. An additional advantage of using low-cost 3D-printed materials is that extensive stress analysis in CAD becomes less critical: owing to the simplicity of the assembly, components can be rapidly reprinted and replaced in case of fracture or material fatigue, or re-designed for further iterations, thereby accelerating the development cycle. Printing parameters were adjusted for each material: the TPU fingers were oriented vertically to align filament deposition with the primary bending axis, improving fatigue resistance, while PLA components were fabricated with higher infill densities to enhance load-bearing capacity and dimensional stability. Unlike most existing soft grippers, where fingers are printed straight and bent under cable traction, the proposed design incorporates pre-curved fingers. In this configuration, cable actuation is primarily required to open the fingers during grasping, as successive design iterations demonstrated that applying force for finger opening is mechanically more efficient than for closing. Printing the fingers in a pre-curved state therefore simplifies motion control, reduces actuation effort, and enhances adaptability to objects of varying geometries. The pre-curvature angle was set to the maximum achievable initial flexion, since the gripping force is ultimately controlled by the cable actuator rather than the passive elastic restoring force alone. This approach maximizes the default closed position, ensuring that objects are securely held even at low cable tensions. Figure 11 shows the printing orientation used for finger fabrication.

This combined material and manufacturing strategy demonstrates how the elasticity and durability of TPU complement the stiffness and precision of PLA, resulting in a lightweight yet robust gripper. Beyond lowering assembly effort and fabrication cost, additive manufacturing also enables rapid customization and adaptation of geometry, confirming its value as a key enabler for soft robotic systems [34].

3.3. Actuation System

The actuation system of the proposed gripper (Figure 12) is based on a single stepper motor that drives two pulleys mounted on its shaft. These pulleys tension the cables routed through the guiding structures and connected to the compliant fingers, thereby enabling synchronized opening and closing motions. Compared with multi-motor configurations, this solution reduces system complexity, lowers energy consumption, and ensures balanced force distribution between the two flexible fingers.

A key feature of the proposed actuation strategy is that the compliant fingers are fabricated in a pre-curved, closed configuration. During 3D printing, the TPU fingers are produced in this collapsed state, which introduces internal elastic tension within the material. This stored elastic energy naturally biases the fingers toward the closed position. The cable, routed along the outer surface of the finger through the guiding elements, acts on each segment to generate an opening moment when tensioned by the motor. Figure 13 illustrates the force distribution along the finger profile: cable traction produces a pulling force at each guiding point, creating a leverage effect that counteracts the internal restoring force of the pre-curved geometry. When the motor releases cable tension, the elastic energy stored in the finger structure drives closure passively, enabling gentle and compliant grasping without active motor engagement.

This configuration contributes directly to the intrinsic safety of the system: since the motor only acts to open the fingers and never contributes to the gripping force, the maximum grip strength is inherently bounded by the elastic energy stored in the pre-curved TPU structure. In the event of power loss or system failure, the fingers return to their natural closed position, but the resulting grip force cannot exceed this predictable elastic limit—unlike motor-driven closure systems where failure could cause uncontrolled clamping. Combined with the integration of a load cell at the gripper base, the system achieves real-time force feedback for threshold-based control, preventing excessive force application and protecting delicate objects. Positioned within the rigid thumb structure, the load cell enables contact detection during grasping, enhancing grasp reliability. Unlike many soft grippers that depend on indirect displacement estimates or distributed sensing arrays, the present approach achieves threshold-based force feedback with a single, compact sensing element.

The gripping force regulation relies on a threshold-based control strategy. During operation, the load cell continuously monitors the contact force, and the controller automatically halts finger closure when a predefined threshold is reached. This threshold was empirically determined to ensure secure grasping without damaging delicate objects. Importantly, the maximum gripping force is inherently bounded by the passive closing mechanism: since closure is driven solely by the elastic restoring force of the pre-curved TPU fingers—rather than by active motor actuation—the force cannot exceed the mechanical limits imposed by the material properties and finger geometry. This eliminates the risk of over-gripping due to control errors or motor over-actuation, providing a predictable and repeatable force profile across all grasping cycles.

The overall actuation strategy therefore combines compactness with intrinsic safety. The single motor and dual-pulley transmission ensure lightweight, synchronized motion, while the pre-curved finger geometry and integrated load cell deliver predictable, safe, and repeatable performance in an economical and resource-efficient manner. Collectively, these refinements distinguish the actuation approach from conventional soft gripper designs by providing a scalable, low-cost, and inherently safe solution suitable for collaborative and assistive environments (see

Table 3. Actuation system specifications.

Parameter	Value
Motor type	Stepper motor
Transmission	Dual-pulley, cable-driven
Pulley diameter	5 mm
Cable diameter	1 mm
Actuation mode	Motor pulls the fingers open; passive elastic closure

and

Table 4. Load cell specifications.

Parameter	Value
Type	Aluminum bar (strain gauge)
Measuring range	5 kg
Precision	1 g
Integration location	Rigid thumb

).

It should be noted that the motor placement in the current prototype was selected primarily for demonstration purposes. In practical deployments, the cable actuation can be driven from remote locations by extending the guiding system. This flexibility allows the actuation unit to be positioned away from the gripper, reducing weight at the end effector and further improving scalability and adaptability to different robotic platforms.

4. Experimental Validation and Testing

The following experiments were designed to validate the proof-of-concept prototype under controlled laboratory conditions, focusing on qualitative demonstration of functionality rather than exhaustive performance benchmarking. The experimental validation of the proposed soft gripper focused on assessing its ability to securely grasp and release objects, its adaptability to varying shapes and sizes, and the effectiveness of the actuation and cable-guiding system under operational conditions. The tests were designed to demonstrate the gripper’s performance in practical scenarios and to evaluate the performance of the trapezoidal fingertip geometry and integrated force-feedback system.

A load cell integrated at the base of the gripper provided real-time feedback during grasping. As described in Section 3.3, the primary function of this sensor is to enable closed-loop control of the grasping cycle rather than precise force measurement. The current implementation serves as a demonstrator concept for sensor integration; the specific load cell was selected for accessibility and ease of integration rather than measurement precision. In a more evolved prototype, a properly calibrated sensor with appropriate specifications could be substituted to enable quantitative force measurement. Two control thresholds were defined: an upper threshold corresponding to secure grip closure, and a lower threshold corresponding to the fully open state. The load cell signal was sampled at 60 Hz, with a 5-sample moving average filter for noise reduction and a spurious reading filter to reject outliers. The open state threshold was set at 1500 a.u. and the grip closure threshold at 2750 a.u.; these values were determined empirically by observing the sensor response when grasping the most delicate test object (the egg). When the sensor output reaches the upper threshold during finger closure, the controller automatically halts the closing motion, limiting the force applied to the object. Conversely, when the sensor output falls below the lower threshold during opening, the system detects that the fingers have fully released the object.

Figure 14 illustrates this control principle with a representative example of two consecutive grasping cycles. Each cycle shows the progressive increase in sensor response as the fingers close, automatic stopping upon reaching the grip threshold (green zone), and subsequent decrease until the open state is detected (blue zone). The consistent behavior across cycles demonstrates the repeatability of the threshold-based control approach. This same strategy was applied throughout all grasping tests described below.

The gripper was tested using a set of representative objects as shown in Figure 15, including a lime, a mandarin, a tomato, an egg, a spice jar, and a spool of thread. These items were selected as representative objects commonly found in household environments, covering a range of fragility, surface properties, and sizes. While predominantly regular in geometry (spherical and cylindrical), these objects reflect typical manipulation targets for assistive applications. Round and soft objects (lime, mandarin, tomato) evaluated the gripper’s ability to adapt to gentle deformation, while the egg tested sensitivity and precision in handling delicate items. Cylindrical objects (spice jar, spool of thread) assessed the gripper’s capability to maintain stable contact across different geometries. The trapezoidal fingertip design contributed to consistent and secure contact with each object.

Throughout the experiments, the gripper demonstrated repeatable grasp and release cycles, maintaining stability and adaptability across all objects. The real-time feedback from the load cell enabled controlled application of force, preventing slippage or damage while maintaining reliable performance. These results confirm that the proposed hybrid design successfully combines key advantages of rigid systems—such as direct force feedback and structural stability—with the inherent compliance and adaptability of soft robotic elements, enabling safe and reliable handling of objects with varying geometries and fragility.

5. Discussion

This section provides a comprehensive discussion of the proposed cable-driven soft gripper. The analysis begins with an overview of the implementation, followed by a detailed examination of its key advantages and inherent limitations. A comparative evaluation with existing soft gripper designs is then presented to highlight the unique contributions of the system in terms of actuation and cable integration strategies. Finally, the discussion addresses the broader implications and potential real-world applications of the developed gripper, emphasizing its practicality, cost-effectiveness, and adaptability for diverse manipulation tasks.

5.1. Implementation Overview

The developed soft gripper follows a lightweight, low-cost, and modular design philosophy consistent with current trends in soft robotics. The lightweight character results from single-motor actuation, 3D-printed plastic components (TPU and PLA), and cable-driven transmission—avoiding the mass of pneumatic systems or multi-motor configurations. The modular and scalable nature is enabled by additive manufacturing: finger segments can be added or removed without complete redesign, the architecture is not constrained to a specific geometry, individual components can be reprinted independently, and the cable-driven actuation scales naturally with different finger sizes. The simplicity of the 3D-printed components makes geometry modifications straightforward and inexpensive; the key contributions of this design lie in demonstrating successful force-feedback sensor integration and establishing a functional form factor that can be readily adapted to different application requirements. The system was designed to demonstrate how accessible fabrication methods and simplified actuation can achieve functional and reliable manipulation without relying on complex control architectures. It employs a three-finger configuration, combining one rigid, sensorized finger with two compliant fingers to enable adaptive grasping of objects with varied shapes and fragility.

Actuation is achieved through a single stepper motor driving a dual-pulley system, which synchronously tensions the cables routed along the compliant fingers. A prismatic load cell integrated into the rigid thumb provides direct force feedback, enabling real-time detection of contact during grasping. This sensing configuration supports gentle and repeatable manipulation while preserving the mechanical simplicity of the system.

The actuation cables are externally routed, reducing friction losses and simplifying inspection and maintenance. The overall system architecture prioritizes mechanical transparency, modularity, and ease of reproduction, enabling straightforward adaptation to different experimental or educational platforms. Together, these design choices highlight the gripper’s potential value as an accessible, resource-efficient example of soft robotic design.

5.2. Key Advantages

The proposed soft gripper aims to introduce several advantages that align with the principles of soft robotics and accessible design:

• Low-cost and accessible fabrication: The use of consumer-grade 3D-printed materials allows the entire system to be produced at minimal cost. This approach supports rapid prototyping, easy reproduction, and wide accessibility for research, educational, and experimental purposes.
• Modular and easily maintainable structure: Externally routed cables and independently mounted guiding elements simplify assembly, inspection, and replacement. Each structural part can be reprinted or modified without specialized tools, promoting open-ended experimentation and customization.
• Simplified actuation and control: The single-motor, dual-pulley configuration enables synchronized finger motion without complex kinematic coordination. This mechanical simplicity reduces control effort, power consumption, and system weight while maintaining consistent grasping behavior.
• Integrated sensing and intrinsic safety: The inclusion of a standard prismatic load cell provides direct force feedback, enabling consistent, repeatable, and gentle gripping. The force-limited actuation design—where the motor only opens the fingers and closing relies solely on passive elastic restoring force—provides intrinsic safety by bounding the maximum grip force to a predictable value determined by the pre-curved TPU geometry. This makes the system well suited for human-interactive or assistive robotic contexts.
• Scalability and adaptability: The lightweight and compact design can in principle be readily scaled or modified for different soft robotic platforms. The motor and cable routing can be positioned remotely, enabling integration into varied research setups or mobile manipulators.

Overall, the gripper suggests that effective, reliable, and safe manipulation can be achieved through a design that prioritizes simplicity, economy, and functional integration—core principles of modern soft robotics.

5.3. Limitations and Challenges

Despite its advantages in simplicity and cost efficiency, the proposed gripper presents several technical limitations inherent to its compact, cable-driven architecture.

• Limited dexterity: The single-motor actuation restricts independent finger motion, limiting the ability to perform complex or asymmetric grasps. Introducing differential or selective coupling mechanisms could enhance dexterity in future iterations.
• Cable and material fatigue: The cable-driven mechanism, while efficient and lightweight, is subject to wear and tension-related fatigue over prolonged use. Similarly, the compliant TPU material may experience mechanical degradation after extensive cyclic loading. The long-term durability and number of cycles to failure were not characterized in the present study. However, the flexible finger components are designed as consumable parts intended for periodic replacement. The modular architecture ensures that worn fingers can be quickly reprinted on any consumer-grade 3D printer capable of processing flexible filaments, minimizing downtime and maintenance cost.
• Force measurement dependency: The single integrated load cell provides global force feedback through the rigid thumb. While effective for contact detection and safe gripping, any misalignment or structural decoupling between the sensor and finger assembly may influence measurement accuracy.
• Limited performance characterization: The present work focused on proof-of-concept validation. Comprehensive performance metrics—including maximum gripping force, load capacity, energy consumption, actuation speed, and long-term reliability—were not systematically measured and remain areas for future investigation. Quantitative grasp success rates were not reported because isolated gripper testing—without integration into a complete robotic arm system—would not accurately reflect real-world performance. Grasp success depends critically on the positioning control loop provided by the manipulator, which is outside the scope of this gripper-focused study.
• Object geometry scope: The experimental validation employed objects with predominantly regular geometries (spherical and cylindrical). While these represent typical household items relevant to assistive applications, the gripper’s performance with highly irregular, non-convex, or asymmetric objects was not evaluated. We note that most household objects present an optimal grasping orientation with cylindrical or curved contact surfaces; successful manipulation of irregular objects would depend primarily on gripper alignment by the robotic arm rather than the gripper itself. Extending the validation to geometrically complex objects remains an area for future work.

These limitations reflect the trade-offs inherent to lightweight, low-cost soft robotic systems. They do not hinder basic functionality but instead define the practical operating boundaries of the current design, guiding future refinements in durability, sensing robustness, and dexterous control.

5.4. Comparison with Existing Soft Gripper Designs

To better understand the strengths and trade-offs of the proposed hybrid gripper, it is compared with other common cable-driven soft gripper approaches. Two aspects are considered: actuation type (single-motor vs. multi-motor) and cable integration (external vs. internal routing).

Table 5. Comparison of single-motor and multi-motor gripper designs.

Design/Feature	Independent Finger Control	Mechanical Complexity	Cost	Ease of Assembly/ Maintenance
Proposed Single-Motor Gripper	Low	Low	Low	High
Multi-Motor Gripper	High	High	Medium-High	Medium

provides a qualitative comparison of single- versus multi-motor designs across key engineering criteria, while

Table 6. Comparison of cable integration approaches in soft grippers.

Design/Feature	Ease of Assembly	Maintenance	Modularity	Manufacturing Complexity
Proposed Gripper	High	Low	High	Low
Traditional Cable-Driven Soft Grippers	Medium	High	Low	Medium-High

compares external versus internal cable integration approaches. These comparisons are based on general design principles and observations from the literature rather than direct experimental measurements against specific systems.

Regarding actuation, the gripper relies on a single stepper motor, in contrast to multi-motor systems which provide high independent finger control. While multi-motor systems allow more complex asymmetric finger movements, they also increase mechanical complexity, cost, and maintenance requirements. The single-motor configuration of the proposed design achieves sufficient finger control for most manipulation tasks, as summarized in the qualitative comparison.

Regarding cable integration, the proposed gripper uses externally routed conduits and 3D-printed components, which are intended to provide a low-cost, modular solution that simplifies assembly, facilitates cable inspection, and enables easy replacement or iteration of parts. In contrast, traditional soft grippers with internally routed cables often require more complex manufacturing and maintenance. These differences are summarized in the comparison Table 6, highlighting the anticipated practical advantages of the proposed approach.

By combining a rigid finger with two flexible fingers and integrating a load cell for real-time force feedback, the hybrid design aims to achieve stable, repeatable, and safe operation, while maintaining adaptability to a wide range of object shapes and sizes. Overall, the proposed system is intended to provide a practical, low-complexity, and cost-effective solution for indoor manipulation tasks and other controlled laboratory applications, seeking to achieve a balance of performance, simplicity, and ease of maintenance while avoiding common drawbacks of traditional cable-driven soft grippers.

5.5. Real-World Application Advantages

The proposed hybrid soft gripper is particularly suited for research, educational, and assistive robotics contexts where safety, adaptability, and cost efficiency are key requirements. Its lightweight and compact architecture make it ideal for experimental setups, small robotic arms, and mobile manipulation platforms that require gentle and repeatable object handling.

In collaborative or assistive applications, the gripper’s intrinsic safety and inherently force-limited actuation—where the maximum grip force is bounded by the passive elastic properties of the pre-curved fingers rather than motor torque—make it well suited for direct human interaction and the manipulation of delicate objects. It can support tasks such as laboratory assistance, sample handling, or service robotics scenarios that involve lightweight components and require reliable, damage-free grasping.

6. Conclusions

The proposed hybrid soft gripper demonstrates that functional and reliable manipulation can be achieved through a design philosophy centered on simplicity, accessibility, and intrinsic safety. By combining a rigid, sensorized thumb with two compliant fingers actuated through a single cable-driven mechanism, the system achieves stable, repeatable grasping without complex control or high-cost components.

The integration of a standard prismatic load cell provides direct and repeatable force feedback for threshold-based control, enabling gentle interaction with delicate objects while preserving the economic and lightweight character of the design. The use of 3D-printed materials and modular construction further enhances reproducibility, ease of maintenance, and adaptability to different experimental or assistive platforms.

This work highlights how accessible fabrication techniques and minimalist actuation strategies can effectively support the core goals of soft robotics—safe interaction, compliance, and adaptability—without reliance on advanced manufacturing or control systems. The resulting design offers a practical and affordable foundation for research, education, and collaborative robotic applications where simplicity, safety, and repeatability are prioritized.

Future developments will focus on improving long-term durability, refining the integration of sensing elements, and exploring scalable control strategies to extend the gripper’s capabilities while maintaining its resource-efficient nature. A key next step is the integration of the gripper onto a robotic arm platform, which will enable quantitative comparative testing against rigid gripper alternatives under realistic manipulation conditions—including the external positioning control loop that is essential for meaningful performance evaluation. Additionally, extending the experimental validation to include objects with irregular, non-convex geometries would provide a more comprehensive assessment of the gripper’s adaptability. Overall, this work contributes to the growing body of evidence that effective soft robotic manipulation can emerge from designs optimized for economy and functional integration rather than complexity.