Pneumatic System Capable of Supplying Programmable Pressure States for Soft Robots

Abstract

Pneumatic soft robots are of great interest in varieties of potential applications due to their unique capabilities compared with rigid structures. As a part of the soft robotic system, the pneumatic system plays a very important role as all motion performance is ultimately related to the pressure control in air chambers. With the increasing flexibility and complexity of robotic tasks, diverse pneumatic robots driven by positive, negative, or even hybrid pressure are developed, and this comes with higher requirements of pneumatic system and air pressure control precision. In this study, we aim to propose a simplified pneumatic design capable of generating programmable pressure states ranging from negative to positive pressure in each air branch. Based on the design concept and system configuration, special inflation and deflation strategies and closed-loop feedback control strategy are proposed to achieve precise pressure control. Then, a prototype of the pneumatic system with six independent air supply branches is designed and fabricated. Experimental results show that the pneumatic system can achieve a wide range of pressure from −59 to 112 kPa. The speed of inflation and deflation is controllable. Finally, we demonstrate three robotic applications and design the related algorithms to verify the feasibility and practicability of the pneumatic system. Our proposed pneumatic design can satisfy the pressure control requirements of a variety of soft robots driven by both positive and negative pressure. It can be used as a universal pneumatic platform, which is inspiring for actuation and control in the soft robotic field.

Introduction

As an emerging branch in the field of robotics, soft robotic technology has aroused increasing concerns.¹ Compared with rigid structures, soft robots have the inherent compliance, more degrees of freedom, and stronger environmental adaptability. These unique capabilities have led to varieties of potential applications in the area of bionic robots,^2,3 search robots,⁴ assistive devices,⁵ manipulating devices,^6,7 and medical robots.^8,9 Diverse robots actuated by smart memory alloy,^10,11 pneumatic/hydraulic pressure,^12–14 tendons,^15,16 and smart deformable materials^17–20 have been developed. Among them, soft pneumatic robots have good performances in various practical applications owing to their flexibility, wider motion range under simple pneumatic inputs, rapid response, low cost, and high fabrication efficiency.

Generally, soft pneumatic robots have a soft main body as the actuator, which is designed with single or multiple internal air chambers that can be deformed by applying positive or negative pressure. The motion, such as bending and twist, can be achieved by controlling the pressure state of multiple chambers. All motion performance is ultimately related to the pressure states in the air chamber. According to the actuating pressure state of each air chamber, the pneumatic actuators can be divided into three categories: positive-pressure actuators,^21–24 negative-pressure actuators,^25–30 and hybrid-pressure actuators^31–34 (Supplementary Data S1). With increasing flexibility of soft robots and the complexity of robotic tasks, diverse pneumatic robots driven by positive, negative, or even hybrid pressure are developed, and this comes with higher requirement of pneumatic system and air pressure control precision.

As a part of the soft robots, the pneumatic system plays a very important role as all motion performance is ultimately related to the pressure control in air chambers. The elaborate pressure control can directly influence the robotic performance. To achieve effective control of air pressure, pneumatic systems have been studied. Sorted by methods of gas source generation, the current pneumatic systems for soft robots can be divided into two types: based on the principle of the syringe and the combination of pumps and solenoid valves. The principle of the syringe is the simplest concept applied to the soft robots actuating.^29,31,35,36 The pressure of the air chambers is controlled by the volume adjustment of syringes or air cylinders. The process of inflation and deflation is more continuous and smoother. However, the maximum pressure of the system is influenced by the volume of both the air chambers and the syringe or cylinder. As the capacity of gas is limited, this system is not suitable for the pneumatic-driven growing robot or the air chamber with poor airtightness. To solve such limitations, the pneumatic system based on the combination of pumps and solenoid valves has been developed. The pump is used as a gas source, and the solenoid valve is used to control the on/off state of the air pathway. However, most of the contemporary pneumatic systems are designed to supply only a single pressure state, positive or negative pressure.^{28,30,37–46} The air pressure in the deflation process occurs as a sharp fall because of the limitation of pneumatic design and the valve's discrete on/off property, which results in the difficulty to achieve the precise pressure control. Besides, several research studies refer to generate both positive and negative pressure by multiple diaphragm pumps, vacuum pumps, and solenoid valves.^32,34,47,48 Respectively, the diaphragm pump is used to generate positive pressure, while the vacuum pump is used to generate negative pressure, which leads to the complexity and cost of the system. With the increasing flexibility of soft robots and the complexity of robotic tasks, it is essential to explore a pneumatic technology capable of supplying diverse pressure states ranging from negative to positive pressure and achieving higher precision of air pressure control.

In this article, we focus on developing a universal pneumatic platform to satisfy diverse pressure requirements of a variety of robotic applications. An innovative design concept and pneumatic configuration capable of generating programmable pressure states ranging from negative to positive pressure are proposed. Special control strategies and the closed-loop feedback control algorithm are designed to achieve controllable inflation and deflation process and precise pressure adjustment. The pneumatic concept and system proposed in this study have the unique advantages of “positive and negative pressure integration” and “controllable inflation and deflation” characteristics. Detailed system verification and a variety of practical applications are conducted to illustrate the efficiency and practicability. Besides, the proposed design concept can be used as a universal pneumatic platform to satisfy the pressure control requirements of a variety of soft robots driven by both positive and negative pressure. With universal applicability and scalability, it is inspiring for actuation and control in the soft robotic field.

Materials and Methods

Design concept and pneumatic configuration

As shown in Figure 1A, the single-cavity actuators are classified into three categories: positive-pressure, negative-pressure, and hybrid-pressure actuators (Supplementary Data S1). Such actuators can be combined to form more complicated soft robots, such as robotic hand, origami robot, snake robot, and growing robot. Generally, each air chamber of the pneumatic soft robot is connected to one air supply branch using an air tube, which acts both as the inlet and outlet port.

FIG. 1.

Classification of typical single-cavity actuators and our designed pneumatic system. (A) Some typical pneumatic robotic single-cavity actuators and their pressure states. (B) Prototype of the pneumatic system. (C) Pneumatic configuration and connecting relationship between components of a single-channel air supply unit. Color images are available online.

To achieve a controllable pressure adjustment and smooth transition between positive and negative pressure in each air branch, we proposed a pneumatic concept and pneumatic configuration using only one pump source and the minimum number of solenoid valves.

The prototype of our pneumatic system and the schematic of one branch are shown in Figure 1B and C; a diaphragm pump is used as the only gas source to generate both the positive and negative pressure in a single-channel air supply unit. Three direct-acting solenoid valves are configured to control the velocity and pressure value in both inflation and deflation process (Supplementary Data S2). Specifically, the outlet port of the pump is connected to the outlet port (A) of a normally closed 3-way 2-state (3/2) solenoid valve (Valve1). The inlet port of the pump is connected to the outlet port (A) of a normally closed 2-way 2-state solenoid valve (Valve2), and the inlet port (P) of Valve2 is connected to the outlet port (A) of a normally closed 3-way 2-state (3/2) solenoid valve (Valve3). The exhaust ports (O) of Valve1 and Valve3 are connected to the atmosphere. The air inlets (P) of Valve1 and Valve3 are integrated into one air tube and connected to one air chamber of the soft actuator.

The pressure in the soft actuator chamber is measured by the pressure sensor. The control of the pump and valves and the pressure feedback form a closed-loop pressure control. The control algorithm can be programmed on the computer according to different tasks, and the pneumatic process is controlled through the generated control signal from the microcontroller and computer. Eventually, a smooth pressure adjustment is achieved in both the inflation and deflation process.

Strategies of pneumatic flow

Based on the pneumatic configuration, the strategy of pneumatic flow is specially designed to control the pressure value. There are three basic pressure states: inflation, pressure holding, and deflation. In addition to the basic inflation and deflation strategies, special slow pulsed deflation and slow pulsed inflation control strategies are designed for more precise pressure adjustment. The coordination of components in inflation and deflation process is shown in Figure 2. The specific strategies are illustrated as follows:

FIG. 2.

Schematics and states of each component in five different strategies. Color images are available online.

Inflation: Valve1 and Valve2 are in “on” state. Valve3 is in “off” state. When the pump works, the air flows from the environment to the inner chamber of the actuator and pressure gradually increases. The inflation speed is controlled by regulating the running speed of the pump.

Holding: The pump, Valve1, Valve2, and Valve3 are all in “off” state. Thus, the air tube is blocked, and the pressure of the external air chamber can be maintained.

Deflation: Valve2 and Valve3 are in “on” state. Valve1 is in “off” state. The air flows from the inner chamber to the atmosphere environment. The pressure of the actuator can decrease to the negative state when the pump works and absorbs the air from the inner chamber. The deflation speed is controlled by regulating the running speed of the pump.

Slow pulsed inflation: Due to the problem that the pump cannot maintain the pressure in the deflation process and according to the limitation of response speed and the switching behavior of valves, the slow pulsed inflation is specially designed to achieve a more precise inflation process and slower pressure increasing speed. The slow pulsed inflation strategy is a dynamic controlling process of the pump and valves. In the control sequence of one circle, Valve2 and Valve3 are in “off” state. Valve1 is in “on” state, and then this state delays for some time (Δt). During the delay time, the working state (off or on) and the flow rate of the pump can be controlled by the pulse width modulation (PWM) signal. Then Valve1 is in “off” state, and Valve2 is in “on” state. After one circle of the slow pulsed inflation, a small amount of air in the tube section between the Valve 1 and Valve 2 is forced into the actuator. As a result, the pressure gradually increases by repeating this control sequence of one circle. The increasing speed of pressure can be controlled by the delay time and the pump.

Slow pulsed deflation: Similarly, the slow pulsed deflation is specially designed to achieve a more precise deflation process and slower pressure decreasing speed. In the control sequence of one circle, Valve1 is in “off” state. Valve2 is in “off” state, and then this state delays for some time (Δt). During the delay time, the working state (off or on) and the flow rate of the pump can be controlled by the PWM signal. After the delay time, the air pressure in the tube section between the pump and Valve2 is lower than the pressure in the actuator. Then Valve3 is in “on” state. Sequentially, Valve2 is in “on” state. Valve3 is in “off” state. After one circle of the slow pulsed deflation, a small pressure drop is achieved. As a result, the pressure gradually decreases by repeating this control sequence of one circle. The decreasing speed of pressure can be controlled by the delay time and the pump.

Principle of closed-loop pressure control

According to the pneumatic flow strategies above, the closed-loop feedback control can be realized by monitoring the pressure through the air pressure sensor. The whole control principle is shown in Figure 3. The target air pressure of the actuator and the acceptable error range are set through the computer, which also sends the control signal to the microcontroller.

FIG. 3.

Flowchart showing the process of closed-loop feedback pressure control strategy. Color images are available online.

In every calculating cycle, the microcontroller receives the signal, compares the target air pressure ( $P_{t a r g e t}$ ) with the actual pressure ( $P_{t}$ ) measured by the sensor, and computes the error ( $e_{t} = P_{t a r g e t} - P_{t}$ ). If the error is beyond the acceptable error range, the pump and valves will be controlled using the pneumatic flow strategies according to the error. First, the error is required to determine either inflation or deflation is needed. If the error is greater than zero, then if the error of last calculating cycle ( $e_{t - 1}$ ) is less than zero, the chamber has been over-deflated. With the over-deflated condition, the chamber will get slow pulsed inflation. Otherwise, the air pump and the solenoid valve will be controlled to inflate directly according to the error. The process of deflation is similar. Eventually, the actual inner pressure is measured and dynamically regulated to reach the target pressure within the allowable error range.

Verification of the Pneumatic System

After prototype fabrication, we evaluate the quantitative characteristics of the pneumatic system. The experiments include pressure and flow characteristics experiment, closed-loop pressure control experiment, step response of the pneumatic controller experiment, and controllable positive and negative pressure cycling experiment. Since the six pneumatic units in the prototype system have the same components and are independently controlled, we measured a single unit to illustrate the performance.

The schematic and experimental setup are shown in Figure 4 and Supplementary Data S3. A syringe is fixed on a linear motion platform and connected with a single pneumatic unit to simulate the chamber of the soft actuator. The computer controls the pneumatic system and the movement of the platform and changes the volume of the syringe to simulate the volume change of the soft actuator. The air pressure in all experiments is measured by the pressure sensor of the system itself. During the pressure and flow characteristics experiment, the flowmeter (MF4003-2-R-BV-A; Siargo, Ltd.) is used between the pneumatic system and syringe to measure the gas flow.

FIG. 4.

Schematic and experimental setup. (A) The experimental schematic of pressure and flow characteristics experiment. (B) The experimental setup of pressure and flow characteristics experiment. Color images are available online.

Pressure and flow characteristics

The pressure and flow characteristics experiment is to evaluate the maximum positive and negative pressure that can be generated by the pneumatic system. The rate of flow is measured simultaneously. The inflation and deflation processes are preprogrammed. The system starts at the atmospheric pressure, and the diaphragm pump works at full speed. The volume of the syringe is set at 5 mL first. The pressure and gas flow curves of the system during the experiment are recorded. Similarly, the volume of the syringe air chamber is changed to 15 mL, and then the inflation or deflation process is repeated.

Figure 5A shows that the pressure rises from the atmospheric pressure to the maximum positive pressure. When the syringe chamber volume is 5 and 15 mL, the positive pressure limit of the system is both stable at 112 kPa and the steady state error is ±2 kPa for long-term stability. The time for pressure to rise from atmospheric pressure to maximum positive pressure is 1.9 and 3.3 s, respectively.

FIG. 5.

Experimental data for the pressure and flow characteristics experiment. (A) The relationship of pressure with time when generating the maximum positive pressure. (B) The relationship of pressure with time when generating the maximum negative pressure. (C) The relationship of flow with time when generating the corresponding positive pressure. (D) The relationship of flow with time when generating the corresponding negative pressure. Color images are available online.

Figure 5B shows the pressure decreases from the atmospheric pressure to the maximum negative pressure. When the syringe chamber volume is 5 mL, the negative pressure limit of the system is stable at −59 kPa. The time for pressure to fall from atmospheric pressure to maximum negative pressure is 3.2 s. When the syringe chamber volume is 15 mL, the negative pressure limit of the system is stable at −57 kPa. The time for pressure to fall from atmospheric pressure to maximum negative pressure is 5.0 s. The steady state error is ±1 kPa for long-term stability. After the pump is turned off, the pressure can maintain the maximum pressure.

The flow rate changes with time in the corresponding inflation and deflation process are shown in Figure 5C and D. Results show that the system has a maximum flow rate when the pump starts to work. As the pressure continues to rise from atmospheric pressure to the maximum positive pressure (or decrease from atmospheric pressure to the maximum negative pressure), the flow rate becomes smaller. Eventually, the flow rate turns to nearly zero when the pump turns off. The flow curve of the system is intermittent due to the working principle of the diaphragm pump.

Results show that although the inflation and deflation speed changes with the volume of the external chamber, the maximum positive and negative pressure of the system are not affected by changes in the volume of the external chamber. The experimental parameters basically agree with the nominal values of the pump.

Closed-loop pressure control

The closed-loop pressure control is essential in the robotic application. The pneumatic system is required to generate stable pressure and can resist the pressure disturbance caused by external factors. This experiment is to evaluate the pressure holding and stabilizing capabilities of the system under closed-loop pressure feedback.

The system is started under atmospheric pressure when the volume of the syringe is fixed at 5 mL. The target pressure is set on the computer. By changing the volume of the syringe, the air pressure disturbance is generated. The measured pressure is deviated from the target pressure. After a short time, the measured pressure returns to the target pressure. The error between the measured pressure and the target pressure is calculated after the system becomes stable.

As shown in Figure 6A and B, the target pressure is set to 40 kPa and −40 kPa, respectively. After the system is started, the air pressure reaches the target pressure in a short time and stabilizes near 40 or −40 kPa. The error between the measured air pressure and the target air pressure is within ±3 kPa. By changing the volume of the syringe chamber, the measured pressure is deviated from the target pressure. The air pressure returns to the target pressure after a short time. When the air pressure is stable, the error between the measured pressure and the target pressure is within ±3 kPa.

FIG. 6.

Experimental data for the closed-loop pressure control experiment. (A) The relationship of pressure with time when generating positive pressure. (B) The relationship of pressure with time when generating negative pressure. Color images are available online.

The system can realize controllable inflation and deflation process. The air pressure is stabilized near the target pressure and can quickly return to the target pressure when pressure disturbance is generated.

Step response of the pneumatic controller

As shown in Figure 7, the experiment is conducted to investigate the performance of our proposed pneumatic system with respect to abrupt changes in target pressure. The volume of the syringe is fixed at 5 mL. The target pressure is abruptly changed ranging from positive to negative, and the experiment is executed for 4 s in each pressure condition. It is verified that our system is very effective in the rapid response of the system air pressure to the target pressure step change and the air pressure stabilization.

FIG. 7.

Experimental data for the step response for the pneumatic controller experiment. Color images are available online.

Controllable positive and negative pressure cycling

Based on the design concept of the pneumatic system, it can freely convert between positive and negative pressure. Moreover, the speed of inflation and deflation can be controlled.

This experiment is designed to evaluate the controllability and stability of the system when the pressure changes under both positive and negative pressure states with different speeds. In this experiment, the volume of the syringe is fixed at 5 mL. The range of positive pressure in the experiment is set between 0 and 80 kPa. The range of negative pressure is set between 0 and −40 kPa. The experimental result is shown in Figure 8. Three different speed states are set in the experiment. Each speed state is repeated two cycles.

FIG. 8.

Experimental data and corresponding speed states for the controllable positive and negative pressure cycling experiment. Color images are available online.

In the first speed state, the pump runs at 35% full speed during the inflation and deflation process. In the second speed state, the pump runs at 35% full speed to reach 80 kPa. After that, the pump is in “off” state, and the delay time is set at 150 ms to decrease the pressure under the positive state. Then the pump runs at 35% full speed to reach −40 kPa. After that, the pump is in “off” state, and the delay time is set at 150 ms to increase the pressure under the negative state. In the third speed state, the process is similar to the process in the second speed state. The difference is that the pump runs at 27% full speed, and the delay time is set at 200 ms. It is verified that our system can generate different speed of inflation and deflation with various control strategies.

Compared with contemporary pneumatic systems summarized in Table 1, our proposed system has the following innovative characteristics: (1) the wide pressure range of each single air channel can satisfy requirements of a variety of robotic applications. Only one pump is applied as the gas source in each air supply branch, which is applicable and economical. Besides, our system still applies to the pneumatic-driven growing robot as the air can be continuously supplied by the pump. (2) The controllable and smooth transition ranging from negative to positive pressure states. (3) The closed-loop pressure control under both positive and negative pressure condition.

Table 1.

Comparison of Different Pneumatic Systems

Description	Characteristics of one air branch
Description	Reference^28,30,34,35	Reference^37–46	Reference^28,30	Reference^32,34,47,48	Our work
Pressure states	Negative and positive	Positive	Negative	Negative and positive	Negative and positive
Air source	Syringe	One air pump	One vacuum pump	One air pump One vacuum pump	One air pump
Capacity of gas	Limited	Unlimited	Unlimited	Unlimited	Unlimited
Number of basic valves	0	1 or more	1	4 or more	3
Additional driving unit	Moto and linear-movement system	No	No	No	No
Factors affecting maximum or minimum pressure	Syringe capacity, volume of air chambers, Airtightness	Only pump	Only pump	Only pump	Only pump
Controllability of inflation process	√	√	X	√	√
Controllability of deflation process	√	X	√	√	√
Smooth positive and negative transition	√	–	–	X	√
Programmability	√	√	√	√	√
Applicable	Positive, negative, and hybrid pressure driven robots (except for growing robots)	Positive driven robots	Negative driven robots	Positive, negative, and hybrid pressure driven robots	Positive, negative, and hybrid pressure driven robots

Based on the pneumatic configuration of the single-channel air supply unit, multiple independent branches can be conveniently united for controlling more air chambers of multi-degrees-of-freedom (multi-DoF) soft robots. It should be pointed out that the pneumatic configuration in this article will be applicable for most pneumatic soft robots as a universal actuation platform, which has a significant improvement in the robotic control field.

Applications of the Actuation System

The concept of “positive and negative pressure integration” and “controllable inflation and deflation” proposed in this article can be applicable in many robotic applications with various pressure requirements. Herein, typical soft pneumatic actuators and robots are designed to illustrate the efficiency and practicability.

Actuator with one pleated chamber

A soft pneumatic actuator with one pleated chamber is designed. The pleated structure can respond to both positive and negative pressure at the same time. As shown in Figure 9A, a prototype made of exoflex 0050 is fabricated using the molding method. The fabricated actuator has a length of 50 mm and a diameter of 20 mm. The experimental setup is shown in Figure 9B. The soft actuator is connected to one air supply branch of the pneumatic system. The pleated air chamber of the actuator elongates under the positive pressure and contracts under the negative pressure. Thus, the actuator has an apparent pressure response effect when the pressure changes between positive state and negative state.

FIG. 9.

Application experiment using pleated actuator. (A) CAD model and prototype of the soft pleated actuator. (B) Experimental setup for inflation and deflation of the soft pleated actuator. (C) Pressure curve and the corresponding state of the soft actuator. CAD, Computer Aided Design. Color images are available online.

An algorithm is designed to generate positive and negative pressure in three cycles. The pressure range in the application experiment is set from −40 to 20 kPa. When the system works, the dynamic curve of the pressure and the corresponding movement of the actuator are recorded.

As shown in Figure 9C, the pneumatic system can realize the smooth transition of positive and negative pressure. The soft silicone body can be extended during inflation process and compressed during deflation process (Supplementary Video S1). The dynamic curve of the air pressure is corresponding to the movement of the actuator.

Flexible parallel platform

Based on the one-DoF soft pleated actuator, a flexible parallel platform with more actuator integration is designed. As shown in Figure 10A, three one-DoF soft pleated actuators are fabricated using silicone rubber, and two disks are printed by a high-resolution printer. After assembling the components, three actuators are fixed in parallel between two disks (Fig. 10B). Actuators are equally spaced at 120° along with the cylinder and independently driven. The parallel platform has multiple degrees of freedom, which can realize the axial extension, contraction, and bending in three dimensions.

FIG. 10.

Prototype of the flexible parallel platform with three pleated actuators. (A) The printed disks and three fabricated pleated actuators. (B) The assembled flexible parallel platform. (C) Experimental setup of the parallel platform. (D) Demonstration of motion state control of the flexible parallel platform with three pleated actuators. Color images are available online.

Figure 10C shows the experimental setup, where the flexible parallel platform is fixed vertically and is connected with three air supply branches of the pneumatic system. A joystick is used to control the parallel platform. An algorithm is designed to receive the signal from the joystick and control the pneumatic system to generate the corresponding pressure value of each actuator.

As shown in Figure 10D and Supplementary Video S2, the platform is controlled by the joystick. When the lower disk is fixed and three soft actuators are independently driven, the upper disk can flexibly adjust the heading direction in the three-dimensional space by applying various pressure combinations. Different pressure states in the actuator are highlighted using different colors (positive pressure: red, negative pressure: blue, atmospheric pressure: white).

When an actuator is pressurized, it expands while the others maintain the initial length. Consequently, the platform bends in the direction opposite of this actuator. As the pressure in the three actuators can be controlled independently, the bending motion in any direction is achieved. When the pneumatic pressure in three actuators is equally increased, the platform can stretch in the axial direction. When the pneumatic pressures in three actuators are equally decreased to the negative state, the platform can contract in the axial direction.

Worm-like robot

Taking inspiration from the characteristics of inchworm locomotion, a soft worm-like robot that has the capability of climbing in the tubular environment is developed. Generally, the elementary structure of the worm-like robot is composed of one extension module and two expanding clampers, and the extension module is equipped between the two clampers (Fig. 11A, B). The clamper expands and the extension module elongates under the positive pressure.

FIG. 11.

CAD model, prototype, and phantom experiment of the worm-like robot. (A) The elementary structure of the worm-like robot. (B) The prototype of the worm-like robot. (C) Phantom experiment of worm-like robot using a PMMA pipe. PMMA, polymethyl methacrylate. Color images are available online.

As shown in Figure 11C, the phantom experiments are conducted using a polymethyl methacrylate (PMMA) pipe. We design an algorithm to control the pressure sequence of moving forward. There are six phases in a complete motion sequence. Respectively, the internal pressure of air chambers is continuously monitored and used as the parameter to trigger the next phase. Phase (1): the end clamper keeps the expansion state, and the head clamper is relaxed to the normal state; Phase (2): the end clamper keeps the expansion state, and the extension module is pressurized to generate an elongation stride; Phase (3): the head clamper expands to anchor the inner wall; Phase (4): the head clamper keeps the expansion state, and the end clamper is relaxed; Phase (5): the extension module returns to the normal state; Phase (6): the head clamper keeps the expansion state, and the end clamper expands to anchor the inner wall. By repeating the six phases of the pneumatic sequence, the worm-like robot can move forward stably and efficiently in the tubular environment (Supplementary Video S3).

Discussion

This study proposes a pneumatic concept capable of supplying positive and negative pressure for soft robots. Based on the design concept, a prototype system with six air supply branches is developed. The pressure range is a very important parameter. The maximum positive and negative pressures are determined by the diaphragm pump. Thus, a wider pressure range can be achieved by choosing a higher power pump. In this article, this system can achieve six independent air supply branches. If the robot has more air chambers and needs more air supply branches, the system can be extensible by integrating more pneumatic units.

The pneumatic strategy and closed-loop control of the pneumatic system are also very important. The controllable pneumatic process can be programmed according to different robotic applications. The test for positive and negative pressure cycling illustrates the capability of the controllable pneumatic process. It should be noticed that although the experiment is carried out with three different frequencies, more inflation and deflation speed can also be achieved. In addition to the proposed basic control strategies and special control strategies, combined control strategies can be applied to achieve more precise pressure adjustment and meet the application requirements. Further industrialization of the system and a more elaborate pressure control algorithm will be explored in the future.

The closed-loop pressure control of the pneumatic system is an essential part of the closed-loop control of soft actuators, as all motion performance of soft actuators is ultimately related to the pressure control in air chambers. To achieve fully closed-loop control of the soft actuator, more sensors need to be applied in the robotic system, such as position sensors, orientation sensors, and force sensors. The multimodal feedback information from these sensors forms the closed-loop control of the soft robotic system and is ultimately performed by the closed-loop pressure control. Further control strategies of the soft robotic system will be explored in the future.

In the application experiments, the deformation and movement of robots are controlled by changing the inner pressure of air chambers. As the soft actuator is not the focus of this article, the typical hybrid-pressure-driven actuator with pleated structure and worm-like robot are designed and evaluated to illustrate the efficiency of our pneumatic system. Apparently, it is also suitable for other soft actuators and robots.

Moreover, the components of the pneumatic configuration can be compatible with fluid. According to the principle analysis, the transmitted medium has scalability. When an additional water tank is used, and the transport media is replaced with water, the pneumatic system can be extended to a hydraulic drive system with the medium like water. As a result, the system can also be used in actuating hydraulic-driven robots.

Conclusion and Future Work

This article presents an innovative pneumatic system capable of supplying programmable pressure states that can be applicable in many robotic applications with various pressure requirements. The design concepts of “positive and negative pressure integration” and “controllable inflation and deflation” are achieved using a single pump source and the minimum number of solenoid valves. Based on the design concept and pneumatic configuration, the inflation and deflation strategies and the principle of realizing the closed-loop control are designed. After theoretical analysis, we design the pneumatic system and fabricate a prototype with six independent air supply branches. The characteristics of the pneumatic system are evaluated and tested. The results show that the pneumatic system could achieve a wider pressure scale ranging from negative pressure to positive pressure. With the pressure sensor, the pneumatic system can realize closed-loop pressure feedback and the controllable speed of inflation and deflation.

In addition, a variety of robotic applications, including a one-DoF pleated actuator, a multi-DoF parallel platform, and a worm-like robot, are used to verify the practicability of the system. According to the future specific robotic application, the pressure response can be programmed for a variety of applications, such as robotic hand, origami robot, snake robot, and pneumatic-driven growing robot. The proposed design concept can be used as a universal pneumatic platform to satisfy the robust pressure requirements ranging from negative to positive driven robots. It is inspiring for actuation in the soft robotic field.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The authors acknowledge supports from National Natural Science Foundation of China (82027807, 81771940) and Beijing Municipal Natural Science Foundation (7212202).

Supplementary Material

References

Rus

, Tolley

. Design, fabrication and control of soft robots. Nature, 2015; 521:467.

Wehner

, Truby

, Fitzgerald

, et al. An integrated design and fabrication strategy for entirely soft, autonomous robots. Nature, 2016; 536:451.

Xie

, Liu

, Kang

, et al. Fully 3D-printed modular pipe-climbing robot. IEEE Robot Autom, 2021; 6:462–469.

McMahan

, Chitrakaran

, Csencsits

, et al. Field trials and testing of the OctArm continuum manipulator. In: 2006 IEEE International Conference on Robotics and Automation (ICRA); Orlando: IEEE, 2006:2336–2341.

Polygerinos

, Wang

, Galloway

, et al. Soft robotic glove for combined assistance and at-home rehabilitation. Rob Autom Syst, 2015; 73:135–143.

Sun

, Song

, Liang

, et al. A miniature soft robotic manipulator based on novel fabrication methods. IEEE Robot Autom, 2016; 1:617–623.

Ranzani

, Cianchetti

, Gerboni

, et al. A soft modular manipulator for minimally invasive surgery: design and characterization of a single module. IEEE Trans Robot, 2016; 32:187–200.

Zhang

, Lu

. A review of recent advancements in soft and flexible robots for medical applications. Int J Med Robotics Comput Assist Surg, 2020; 16:5951–5975.

Ranzani

, Gerboni

, Cianchetti

, et al. A bioinspired soft manipulator for minimally invasive surgery. Bioinspir Biomim, 2015; 10:035008.

10.

Copaci

, Blanco

, Martin-Clemente

, et al. Flexible shape memory alloy actuators for soft robotics: modelling and control. Int J Adv Robot Syst, 2020; 17:1–15.

11.

Wang

, Ahn

. Shape memory alloy-based soft gripper with variable stiffness for compliant and effective grasping. Soft Robot, 2017; 4:379–389.

12.

Tolley

, Shepherd

, Mosadegh

, et al. A resilient, untethered soft robot. Soft Robot, 2014; 1:213–223.

13.

Onal

, Rus

. Autonomous undulatory serpentine locomotion utilizing body dynamics of a fluidic soft robot. Bioinspir Biomim, 2013; 8:026003.

14.

Marchese

, Onal

, Rus

. Autonomous soft robotic fish capable of escape maneuvers using fluidic elastomer actuators. Soft Robot, 2014; 1:75–87.

15.

Kang

, Lee

, In

, et al. Development of a polymer-based tendon-driven wearable robotic hand. In: 2016 IEEE International Conference on Robotics and Automation (ICRA); Stockholm: IEEE, 2016:3750–3755.

16.

Lee

, Crea

, Malcolm

, et al. Controlling negative and positive power at the ankle with a soft exosuit. In: 2016 IEEE International Conference on Robotics and Automation (ICRA); Stockholm: IEEE, 2016:3509–3515.

17.

Al-Rubaiai

, Pinto

, Qian

, et al. Soft actuators with stiffness and shape modulation using 3D-printed conductive polylactic acid material. Soft Robot, 2019; 6:318–322.

18.

Wang

, Chen

, Lau

HYK

, et al. Motion planning based on learning from demonstration for multiple-segment flexible soft robots actuated by electroactive polymers. IEEE Roboti Autom Lett, 2016; 1:391–398.

19.

Rizzello

, Serafino

, Naso

, et al. Towards sensorless soft robotics: self-sensing stiffness control of dielectric elastomer actuators. IEEE Trans Robot, 2019; 36:174–188.

20.

, Lum

, Mastrangeli

, et al. Small-scale soft-bodied robot with multimodal locomotion. Nature, 2018; 554:81.

21.

Holland

, Abah

, Velasco-Enriquez

, et al. The soft robotics toolkit: strategies for overcoming obstacles to the wide dissemination of soft-robotic hardware. IEEE Robot Autom Mag, 2017; 24:57–64.

22.

Hawkes

, Blumenschein

, Greer

, et al. A soft robot that navigates its environment through growth. Sci Robot, 2017; 2:eaan3028.

23.

Wang

, Ge

, Gu

. Programmable design of soft pneu-net actuators with oblique chambers can generate coupled bending and twisting motions. Sens Actuators A Phys, 2018; 271:131–138.

24.

Zhang

, Hu

, Yang

, et al. Design and modularization of multi-DoF soft robotic actuators. IEEE Robot Autom Lett, 2019; 4:2645–2652.

25.

Zhou

, Chen

, Hu

, et al. Adaptive variable stiffness particle phalange for robust and durable robotic grasping. Soft Robot, 2020; 7:743–757.

26.

Eric

, Nicholas

, John

, et al. Universal robotic gripper based on the jamming of granular material. P Narl Acad Sci USA, 2010; 107:18809–18814.

27.

, Stampfli

, Xu

, et al. A vacuum-driven origami “magic-ball” soft gripper. In: 2019 IEEE International Conference on Robotics and Automation (ICRA); Montreal: IEEE, 2019:7401–7408.

28.

, Vogt

, Rus

, et al. Fluid-driven origami-inspired artificial muscles. P Narl Acad Sci USA, 2017; 13132–13137.

29.

Yang

, Verma

, Lossner

, et al. Negative-pressure soft linear actuator with a mechanical advantage. Adv Mater Technol, 2016; 2:160–164.

30.

Ainla

, Verma

, Yang

, et al. Soft, rotating pneumatic actuator. Soft Robot, 2017; 4:297–304.

31.

, Yang

, Yu

, et al. A crawling soft robot driven by pneumatic foldable actuators based on miura-ori. Actuators, 2020; 9:26–39.

32.

Fatahillah

, Oh

, Rodrigue

. A novel soft bending actuator using combined positive and negative pressures. Front Bioeng Biotechnol, 2020; 8:472–482.

33.

Yang

, Zhang

, Kan

, et al. Hybrid jamming for bioinspired soft robotic fingers. Soft Robot, 2020; 7:292–308.

34.

Hashem

, Stommel

, Cheng

, et al. Design and characterization of a bellows-driven soft pneumatic actuator. IEEE ASME Trans Mechatron, 2020, DOI: 10.1109/TMECH.2020.3037643.

35.

Fraś

, Czarnowski

, Maciaś

, et al. New STIFF-FLOP module construction idea for improved actuation and sensing. In: 2015 IEEE International Conference on Robotics and Automation (ICRA); Seattle: IEEE, 2015:2901–2906.

36.

Zou

, Lin

, Ji

, et al. A reconfigurable omnidirectional soft robot based on caterpillar locomotion. Soft Robot, 2018; 5:164–174.

37.

Zhang

, Fan

, Yang

, et al. Worm-like soft robot for complicated tubular environments. Soft Robot, 2019; 6, 399–413.

38.

Calderón

, Ugalde

, Zagal

, et al. Design, fabrication and control of a multi-material-multi-actuator soft robot inspired by burrowing worms. In: 2016 IEEE International Conference on Robotics and Biomimetics (ROBIO); Qingdao: IEEE,, 2016:31–38.

39.

Pylatiuk

, Kargov

, Gaiser

, et al. Design of a flexible fluidic actuation system for a hybrid elbow orthosis. In: 2009 IEEE International Conference on Rehabilitation Robotics (ICORR); Kyoto: IEEE,, 2009:167–171.

40.

Zhou

, Yi

, Chen

, et al. BCL-13: a 13-DOF soft robotic hand for dexterous grasping and in-hand manipulation. IEEE Robot Autom Lett, 2018; 3:3379–3386.

41.

Joshi

, Paik

. Pneumatic supply system parameter optimization for soft actuators. Soft Robot, 2021; 8:152–163.

42.

Lee

, Lee

, Shin

. An air recirculation system based on bioinspired soft re-air valve for highly efficient pneumatic actuation. Soft Robot, 2021; 8:564–576.

43.

Hong

, Park

, et al. Design of pneumatic origami muscle actuators (POMAs) for a soft robotic hand orthosis for grasping assistance. In: IEEE 2020 3rd IEEE International Conference on Soft Robotics (RoboSoft); New Haven: IEEE,, 2020:627–632.

44.

Yap

, Lim

, Nasrallah

, et al. A soft exoskeleton for hand assistive and rehabilitation application using pneumatic actuators with variable stiffness. In: 2015 IEEE International Conference on Robotics and Automation (ICRA); Seattle: IEEE, 2015:4967–4972.

45.

Zhao

, Jalving

, Huang

, et al. A helping hand: soft orthosis with integrated optical strain sensors and EMG control. IEEE Robot Autom Mag, 2016; 23:55–64.

46.

Chen

, Liu

, Wang

, et al. A bio-inspired self-propelling endoscopic device for inspecting the large intestine. Bioinspir Biomim, 2019; 14: 066013.

47.

Chen

, Hu

, Zou

, et al. Design and fabrication of a multi-motion mode soft crawling robot. J Bionic Eng, 2020; 17, 932–943.

48.

Yang

, Asbeck

Design and characterization of a modular hybrid continuum robotic manipulator. IEEE ASME Trans Mechatron, 2020, DOI: 10.1109/TMECH.2020.2993543.

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.

1.44 MB

6.50 MB

8.29 MB

17.07 MB

0.00 MB