Abstract In the realm of modern robot technology, the limitations of rigid actuators in industrial robot practical teaching have become increasingly evident, particularly in addressing diverse grasping needs for fragile, deformable, and various-sized objects. This paper presents our research on designing a flexible actuator experimental device for industrial robots, inspired by human finger mechanics and integrated with pneumatic driving technology. We detail the structural design, finite element analysis, fabrication processes, and performance testing of the device. The results demonstrate that the proposed flexible actuator enhances the adaptability of industrial robots in educational settings, aligning with the evolving demands of robot technology and engineering practice teaching.
Keywords: industrial robot; flexible actuator; pneumatic drive; finite element analysis; robot technology; practice teaching
1. Introduction
The advancement of robot technology has been a cornerstone in transforming manufacturing industries toward intelligence and high precision. As a critical component of industrial robots, end-effectors play a pivotal role in tasks such as grasping, handling, and assembly. However, traditional rigid actuators, widely used in current practice teaching, suffer from inherent limitations in adapting to objects with irregular shapes, fragile textures, or varying sizes . This gap motivates the development of flexible actuators, which mimic biological flexibility and offer gentler interaction with objects.
In educational contexts, integrating flexible actuator technology into robot technology curricula is essential for cultivating students’ capabilities in innovative design and interdisciplinary problem-solving. Our study aims to bridge the shortage of flexible actuator devices in practice teaching by proposing a pneumatically driven flexible actuator inspired by human finger structures. Through systematic design, simulation, and experimentation, we validate its performance and demonstrate its utility in enhancing the versatility of industrial robot training.
2. Working Principle and Design of the Flexible Actuator
2.1 Selection of Driving Mechanism
In robot technology, actuator driving mechanisms vary, including hydraulic, electric, and pneumatic systems. For educational purposes, pneumatic driving is chosen for its simplicity, safety, and compatibility with existing industrial robot setups . We adopt a high-speed pneumatic network actuator structure, which features rapid response, bidirectional bending capability, and straightforward fabrication—key advantages for practice teaching environments .
The working principle of the pneumatic network actuator relies on fluid pressure within a closed chamber. When pressurized air is introduced, the extendable top layer of the actuator expands, while the non-extendable bottom layer remains fixed, causing the actuator to bend . This mechanism mimics the articulation of human fingers, enabling precise control over bending direction and force.
2.2 Structural Design Inspired by Human Fingers
Human fingers exhibit exceptional adaptability in grasping, with the index finger being the most frequently used for fine manipulation. Our design parameters are derived from anthropometric data: the total length of the actuator is set to 120 mm (5 mm extended from the average adult index finger length), and the width is 20 mm to balance space efficiency and grasping stability .
To replicate the functionality of finger joints without complex compartmentalization, we design a variable-chamber structure where the chamber height decreases from the base to the tip. This design enhances energy transmission, ensuring greater bending at the proximal end and smoother contact at the tip . Additionally, a 15° wedge angle is incorporated at the top to optimize bidirectional bending performance, as validated by finite element analysis .
Table 1. Key Structural Parameters of the Flexible Actuator
| Parameter | Value | Rationale |
|---|---|---|
| Total Length | 120 mm | Based on human index finger dimensions |
| Width | 20 mm | Balances space and grasping area |
| Chamber Height Gradient | Decreasing from base to tip | Enhances bending consistency and tip force |
| Wedge Angle | 15° | Optimized for bidirectional bending |
2.3 Experimental Setup Design
The flexible actuator system comprises the actuator itself and a connecting mechanism. The connecting device is designed to interface with robot flanges, featuring gas channels for pneumatic control and a four-actuator configuration (spaced 90° apart) to ensure stable grasping of various objects .
The grasping process involves controlled air pressure input: when pressurized air flows into the actuators, they bend toward the object, conforming to its shape and securing it through gentle wrapping. This design minimizes damage to fragile items while accommodating different sizes, a critical improvement over rigid actuators in robot technology education .
3. Finite Element Analysis and Material Selection
3.1 Hyperelastic Material Modeling
For accurate simulation, we select Dragonskin 30, a hyperelastic silicone rubber, due to its high elasticity and durability. Its mechanical behavior is modeled using the Yeoh hyperelastic model, which describes the strain energy function W as:\(W = \sum_{i=1}^{n} C_{i}(I_{1} – 3)^{i}\) where \(C_{i}\) are material constants (shear moduli), and \(I_{1}\) is the first invariant of the strain tensor, calculated as:\(I_{1} = \lambda_{1}^{2} + \lambda_{2}^{2} + \lambda_{3}^{2}\) Here, \(\lambda_{j}\) represent the principal stretch ratios. The nominal principal stress \(\sigma_{j}\) is derived from the strain energy function and Lagrangian multiplier P:\(\sigma_{j} = \frac{\partial W}{\partial \lambda_{j}} – \frac{P}{\lambda_{j}}\) For Dragonskin 30, the material coefficients are set as \(C_{1} = 0.11\) and \(C_{2} = 0.02\), capturing both compression and tension phases effectively .
3.2 Tip Force Analysis
The tip force of the actuator is a critical performance metric in robot technology, determining its grasping capability. Using Abaqus, we simulate the tip force under different pressures, incorporating a damper as a load cell for accurate measurement . The results show a strong correlation between input pressure and tip force, with a near-linear increase in force as pressure rises from 10 kPa to 40 kPa .
Table 2. Tip Force Simulation vs. Experimental Data at 30 kPa
| Condition | Simulated Tip Force (N) | Experimental Tip Force (N) | Deviation (%) |
|---|---|---|---|
| 30 kPa Pressure | 0.58 | 0.60 | 3.3 |
3.3 Geometric Parameter Optimization
We analyze the effects of chamber width (16 mm, 20 mm, 24 mm) and chamber number (7, 9, 11) on bending angle and tip force. Results indicate that increasing chamber width enhances both bending angle and tip force due to larger air volume, while more chambers improve bending flexibility but reduce tip force due to increased leverage .
Based on these findings, we select a 20 mm chamber width and 11 chambers to balance performance, cost, and educational applicability .
4. Fabrication Process
The fabrication of the flexible actuator integrates 3D printing and lost-wax casting, ensuring precision and repeatability—key aspects for educational reproducibility in robot technology labs. The process involves:
- 3D Printing the Mold: Using a high-resolution 3D printer to create a mold that defines the actuator’s geometry.
- Material Mixing and Degassing: Combining Dragonskin 30 components in a 1:1 ratio and degassing them in a vacuum chamber to eliminate air bubbles .
- Casting and Curing: Pouring the degassed silicone into the mold, curing the top and bottom layers sequentially, and bonding them using heat (10-minute 烘烤 cycles) .
- Assembly: Attaching the actuators to a 3D-printed four-finger fixture with 90° spacing, ensuring airtight connections and easy replacement .
Table 3. Fabrication Equipment and Materials
| Equipment/Material | Role |
|---|---|
| 3D Printer | Mold fabrication |
| Vacuum Chamber | Degassing of silicone |
| Hot Air Dryer | Accelerating curing process |
| Dragonskin 30 Silicone | Primary actuator material |
| Polyjet Resin | Mold material |
5. Performance Testing
5.1 Experimental Setup
The testing platform includes a pneumatic control system with components such as pressure regulators, electromagnetic valves, and a data acquisition system via LabVIEW . Each actuator is independently controlled, allowing precise adjustment of air pressure and flow rate.
5.2 Bending Angle and Tip Force Validation
At 45 kPa, the actuator achieves nearly full bending, with experimental bending angles slightly deviating from simulations due to neglect of viscoelastic effects in the model . The tip force shows a strong linear relationship with pressure, with experimental values closely matching simulations (e.g., 0.60 N at 30 kPa vs. 0.58 N simulated) .
5.3 Grasping Experiments
Using a two-finger gripper, we test grasping forces on rigid rings of varying diameters (15–60 mm) at different pressure levels. Results show that higher pressure and more proximal grasping positions (closer to the base) yield greater forces, demonstrating the actuator’s adaptability to object size and shape .
Table 4. Grasping Force vs. Object Diameter at 40 kPa
| Object Diameter (mm) | Grasping Force at Position 1 (N) | Grasping Force at Position 6 (N) |
|---|---|---|
| 15 | 2.1 | 4.3 |
| 30 | 1.8 | 3.9 |
| 45 | 1.5 | 3.2 |
| 60 | 1.2 | 2.8 |
The three-finger gripper successfully grasps objects like cylindrical molds, oranges, and relays, highlighting its versatility in handling soft, fragile, and irregular items—scenarios where rigid actuators in traditional robot technology fall short .
6. Conclusion
This study presents a novel flexible actuator experimental device for industrial robots, addressing the limitations of rigid end-effectors in practice teaching. By integrating bionics principles, pneumatic driving, and advanced material science, we have developed a system that excels in grasping diverse objects, the results of the study show that the device enhances the educational value of robot technology courses by providing students with hands-on experience in flexible robot design, simulation, and testing.
Looking ahead, we aim to expand the device’s applications through interdisciplinary integration, such as combining it with sensor networks for force feedback or integrating it into automated robotic systems. This work contributes to the advancement of robot technology in education, fostering the next generation of engineers capable of driving innovation in flexible robotics and intelligent manufacturing.