Compact, lightweight hydraulic steering system for underwater applications

Abstract

A compact and lightweight hydraulic manipulation system for underwater applications is disclosed. The system includes a hybrid underwater manipulator and a hydraulic control architecture, the hydraulic control architecture including at least two compensation bladders. The system is used for general underwater work, such as underwater sampling, high-pressure welding and marine aquaculture.

Description

Technical Field

[0001] A compact, lightweight hydraulically manipulated device, system and method for underwater applications are disclosed. The device, system and method are used for underwater sampling, high pressure welding and marine aquaculture, among other underwater uses. Background Art

[0002] Underwater sampling is essential for scientific research of marine organisms. Typically, at least four experienced divers are required to conduct underwater sampling in shallow water within the euphotic zone, especially in tropical waters where biodiversity is remarkably active. Considering the high risk and physical limitations of divers in highly dynamic underwater environments, remotely operated vehicles (ROVs) and autonomous underwater vehicles are actively developed and widely used for underwater exploration and intervention. However, underwater operations, which mainly focus on underwater intervention through physical action, remain highly challenging, and most commercial underwater manipulators either have primitive flexibility (1-2 degrees of freedom) or heavy / bulky heavy applications. The rigidity of conventional rigid manipulators also makes it difficult to handle flexible and fragile aquatic specimens. In addition, high-pressure hydraulic actuation and control systems are often heavier and bulkier than the manipulator itself, and such systems are usually driven under water pressure at the working depth. In this case, a fairly large underwater platform is required to provide underwater maneuverability for such manipulators and the corresponding drive systems. The progressively increasing size, weight, cost, and maintainability have greatly limited their wider application.

[0003] Compared to traditional rigid-body robots, flexible robots offer new methods for underwater manipulation. In particular, the inherent adaptability and water-resistance of flexible actuators are ideal for grasping soft and elastic objects underwater, inheriting cable-driven and biomimetic approaches. Fluid elastomer actuators also work well in underwater applications, with continuum structures, hybrid structures, and even modular structures tested in high-pressure environments equivalent to 2,300 meters of depth. 3D-printed flexible manipulators have also proven very successful in deep-sea operations, with testing depths exceeding 2,200 meters, offering better compactness and inherent compliance than rigid manipulators in delicate underwater sampling. Summary of the invention

[0004] A simplified overview of the present invention is provided below to provide a basic understanding of some aspects of the present invention. This overview is not a broad overview of the present invention. It is neither intended to identify key or important elements of the present invention nor to describe the scope of the present invention. On the contrary, the sole purpose of this overview is to introduce some concepts of the present invention in a simplified form as a preface to a more detailed description below.

[0005] Underwater manipulators are essential for underwater robotic sampling and other service operations. Conventional rigid underwater manipulators usually require large size and weight, which hinders their widespread application. Pioneering underwater soft robots overcome this by providing flexible and lightweight arms and grippers, but still require a large number of drive and control components to resist water pressure and achieve the required dynamic performance. In this work, we propose a novel underwater manipulator design and control method that takes advantage of the unique characteristics of soft robots, adopts a hybrid structure (rigid frame + soft actuators) to improve stiffness and force output, adopts a unified actuator design that allows a compact hydraulic drive system to drive all actuators, and a new fully customizable soft bladder design, which improves performance in multiple areas: (1) The force output of the actuators is decoupled from the working depth, allowing a wide working range; (2) All actuators are connected to the main hydraulic pipeline, eliminating the need for actuator-specific control circuits, resulting in a very compact actuator system that is particularly suitable for high dexterity situations; (3) The dynamic response is significantly improved compared to the counter system without a bladder. A prototype of a flexible manipulator with 4 degrees of freedom, dual bladders and a 15N load was developed; the entire system (including drive, control and battery) can be installed on a consumer-grade remotely operated submersible, and its depth-independent performance is verified by various laboratory and field test results under different climatic and hydrological conditions. The analytical model and verification of the proposed soft bladder design also provide guidance for other applications.

[0006] The invention discloses a device for underwater application, which is made of a hybrid underwater manipulator and a hydraulic control framework including at least two compensation soft bags.

[0007] To achieve the above and related ends, the present invention includes the features fully described below and particularly pointed out in the claims. The following description and drawings set forth in detail certain illustrative aspects and implementations of the present invention. However, these are only a few of the various ways in which the principles of the present invention can be employed. Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of the present invention when considered in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 An underwater manipulator with a bladder-based hydraulic system mounted on a consumer-grade ROV (Remotely Operated Vehicle) according to an embodiment is described.

[0009] Figure 2 A comparison between a conventional underwater manipulation system and a CSB-based manipulation system according to an embodiment is described. Figure 2a is a conventional underwater manipulation system. The entire system is isolated from the ambient water, so an oil tank is required as a liquid source. At least one high-power hydraulic pump is required to counter the ambient water pressure. The high-power drive unit also needs to respond quickly, and each actuator requires a separate feedback loop for dexterous and accurate control. Figure 2 b describes an underwater manipulation system according to an embodiment. Only the electronic components are sealed. The liquid source comes directly from the ambient water, and both the CSBs and the actuators are exposed to the ambient water, ensuring the balance of the ambient water pressure. All actuators are controlled by compact and simple inlet and outlet control loops. Improved dynamic response can be achieved by utilizing the elasticity of two CSBs (compensating soft bladders).

[0010] Figure 3 A system overview according to an embodiment is described. Figure 3 a is a schematic diagram of the proposed underwater hydraulic system. Figure 3 b Describes the underwater hydraulic system with a manipulator.

[0011] Figure 4 The manipulator features according to the embodiment are described. On the left is a schematic diagram of the manipulator's motion, in the middle is an exposed view of the manipulator, and on the right is a dexterous hybrid underwater manipulator.

[0012] Figure 5 a describes the CSB molding process (inlet). The outlet CSB molding process is the same except there is no fiber reinforcement. Figure 5 b Describes the flexible actuator forming process. The gripper and wrist actuator follow the same procedure but with different lengths. Figure 5 c is a schematic diagram of two CSBs. The upper one is the inlet CSB (stretching), and the lower one is the outlet CSB (contracting). λe and λc are the principal stretching ratios of the inlet and outlet CSBs, respectively. Figure 5 d shows the CSB pressure and strain test results (inlet: top, outlet: bottom). The blue line is the analytical result and the red dot is the experimental result. Figure 5 d Graph showing the test results of CSB flow rate and pressure difference (inlet: top, outlet: bottom). ANA, analytical; EXP, experimental

[0013] Figure 6 A CSB test platform apparatus according to an embodiment.

[0014] Figure 7 Table 1 and Table 1 describe the CSB-based hydraulic system tracking performance test results according to an embodiment. Figure 7 a shows the Pa (step function) tracking performance. Figure 7 b shows the Pb (sinusoidal signal) tracking performance. Figure 7c-7f shows the Pc (sinusoidal signal with increasing frequency) tracking performance of all four system devices. The red shading is the tracking time before tracking failure (critical time).

[0015] Figure 8 The flexible actuator and the gripping force test are described. Figure 8 a and Figure 8 b Reports the stress and strain test results of the flexible actuators (wrist and gripper actuators). Figure 8 c and Figure 8 d Reports the force and pressure test results of the flexible actuators (wrist and gripper actuators). Figure 8 e Reports the hybrid gripper holding force test results (different inlet CSB pressures). Figure 8 f Describes the flexible actuator validation test setup (top: gripper actuator; bottom: wrist actuator). Figure 8 g shows the gripper test setup. The force sensor is mounted on one finger of the gripper; apart from mounting the force sensor on one finger, no further restraint is added and the rest of the gripper is left in its natural working state.

[0016] Fig. 9 Tests of the maneuvering system at different water depths are reported. Fig. 9 a Test results are reported. Tests were conducted on land, in water tanks and at sea. Fig. 9 b shows the clay box used to demonstrate the gripping force. The bottom section lists the clay boxes for all tested depths.

[0017] Fig.10 The hybrid manipulator system validation is described. Fig.10 a describes the 15 m water flume used for the speed tests. Fig.10 b Depiction of the manipulator dexterity and accuracy testing (pick and place) setup (onshore). Fig.10 c describes the pick and place process. Fig.10 d depicts a speed test of a single ROV versus the same ROV with an underwater maneuvering system.

[0018] Fig.11 A validation study of an underwater maneuvering system in three marine biodiversity hotspots with different hydrological and climatic conditions is shown. Fig.11 aCoral lagoon, Kandima Island, Maldives; Fig.11 bTaiwan, China, and Guishan Island; and Fig.11 cWest Coast, Hong Kong, China. DETAILED DESCRIPTION

[0019] The solutions of the prior art still compromise between compactness, lightness and dexterity, mainly because the hydraulic actuation and control components of existing flexible manipulators are still similar to those of rigid manipulators: (1) for flexibility, multiple actuators require dedicated pressure regulation feedback loops; (2) for working depth, the drive pressure needs to be matched to counter the ambient water pressure at the working depth; (3) for fast response, high-power drive units are often required. Some efforts have been made to address these limitations, from compact pneumatic systems with chemical reactions (not suitable for underwater applications) to calipers with ambient water pressure compensation for grasping objects with original dexterity. In particular, accumulators are manufactured to both counteract the ambient pressure and increase the flow of the drive stage. Springs are used in the accumulator to maintain the ambient pressure and provide increased flow. Therefore, in order to achieve compactness, lightness and fast response at the same time, innovations in the drive and control systems under development are needed.

[0020] In this paper, a novel hybrid underwater manipulator architecture is proposed, characterized and demonstrated to have a unified flexible actuator design and soft bladder hydraulic actuation ( Figure 1 ), by using two compensating soft bladders (CSBs) in the hydraulic actuation system, flexibility, compactness, depth independence and fast response are achieved. The complete work of mechanical design, analytical modeling and prototype development of CSB and flexible actuators is presented, as well as various test results in controlled and open water environments at different sites in tropical areas with various hydrological conditions.

[0021] The main contribution is the multi-purpose CSB concept and the overall flexible manipulator drive and control method based on this concept, which achieves a unique combination of experimentally verified characteristics that are different from the state-of-the-art solutions: (1) Highly compact, with CSBs, the entire prototype arm has a payload of 15N and 4 degrees of freedom flexibility, weighs 2.5kg (including batteries and electronics), and can be installed on a leisure ROV platform; (2) Depth independence, with uniform force output performance throughout the entire working depth range, unaffected by changes in ambient water pressure; (3) Improved dynamic response, which is nearly twice as fast by using CSBs in the hydraulic circuit.

[0022] The results of the manipulator described in this paper are very promising for a flexible, compact, lightweight compliant manipulator with nearly consistent load and dynamic performance throughout the entire specified operating depth range. Following this approach, recreational-class ROVs can be equipped with flexible arms that are easy to carry and use, leading to a host of applications in diver assistance and underwater operations.

[0023] Concept Design and System Overview

[0024] Design requirements

[0025] For general underwater sampling and manipulation tasks, a flexible manipulator that is compact, easy to operate, has a large working depth range, and is highly responsive and capable of performing tasks is specifically designed.

[0026] The following design requirements describe these rather contradictory characteristics:

[0027] (1) Compact. The complete self-contained underwater system supporting the manipulator should be able to be installed on a consumer-grade compact ROV and be significantly more compact than existing underwater manipulators. Therefore, traditional hydraulic boxes can be eliminated, and compact pumps / valves are superior to traditional compressors / cylinders, with minimal drive and control hardware.

[0028] (2) Large working depth range. Since the compact hydraulic components used have a limited pressure range, the ambient water pressure becomes a key factor affecting the hydraulic performance. Measures can be taken to ensure that the operation and performance of the flexible manipulator are not affected by the working depth.

[0029] (3) Fast response. Fast response and tracking capabilities are critical to dynamic tasks of sampling and operation. Given the limited flow and response time of compact drive components, steps can be taken to improve dynamic response.

[0030] (4) Flexible arm. Considering the complex underwater environmental factors and the inherent compliance characteristics of flexible robots, underwater sampling tasks require flexible arms that are easy to carry and use.

[0031] Overview of underwater control system

[0032] The proposed underwater manipulator system is different from the traditional hydraulic underwater manipulator ( Figure 2 ) has been significantly modified to meet the design requirements. The main features are as follows:

[0033] (1) A CSB-based control strategy is adopted to reduce the use of pumps, valves, and control hardware. Compared with the traditional underwater manipulator that uses a separate control loop for each actuator, the hydraulic system described in this paper has only two control loops for all the actuators used in the manipulator. Since two CSBs are connected to all the actuators, they act like a pressure control unit. In addition, CSBs can protect the actuators from exceeding the maximum pressure threshold by presetting the working pressure range.

[0034] (2) Two CSBs are exposed to ambient water to balance the ambient water pressure. In this case, the working pressure of the pump is independent of the ambient water depth. Compared with the accumulator used by Phillips et al. (Phillips BT, Becker KP, Kurumaya S, et al. A dexterous, glove-based teleoperable low-power soft robotic arm for delicate deep-sea biological exploration. Sci Rep 2018; 8: 14779) that uses springs to maintain ambient pressure, the two CSBs naturally offset the ambient pressure and utilize the inherent compliance of the soft robot. In this case, the water depth factor can be removed from the development level.

[0035] (3) Fast dynamic response and tracking capability are achieved by utilizing the inherent compliance of the two CSBs. By setting the operating pressure range of the CSB, a pressure difference will be generated between the CSB and the actuator. In this case, water from the pump can be released at a greater flow rate. In addition, by adjusting the material and wall thickness using the analytical model proposed for the design and modeling of the CSB and actuator parts, the CSB can be customized for different applications.

[0036] (4) The underwater manipulator system adopts a modular actuator design. Since the pressure of the flexible actuator can be controlled by two CSBs and corresponding valves, no sensor is required for each actuator. In this case, when more degrees of freedom (DOFs) are required, only valves and actuators need to be added to the system. In addition, the modular design helps to keep the system compact when increasing the degrees of freedom.

[0037] Proposed CSB-based hydraulic control architecture

[0038] like Figure 3The proposed hydraulic control architecture has multiple advantages, including system compactness, wide working depth range and dynamic performance. The hydraulic system consists of eight valves (2.5W each), two pumps (7.2W each), two CSBs, two pressure sensors and a control unit. The maximum power consumption is 34.4W, which is less than half of the power required by a standard underwater manipulator. In order to be waterproof and compact, all electronic components are placed in four acrylic tubes. The diameter of the four tubes is 100mm and the length ranges from 140 to 250mm. Overall, the whole system weighs 2.5kg and can be packed into a 400x 500x 250mm box. The weight in water is about zero. The two pumps are connected to the two CSBs respectively to balance the ambient water pressure when the CSBs are exposed to the ambient water. The three actuators on the joint correspond to six valves, and the remaining two valves are connected to the two actuators on the gripper and move simultaneously with the two actuators. All actuators are controlled by two CSBs and corresponding valves. Since the large pressure difference between the actuator and the two CSBs will result in a higher flow rate, the dynamic performance of the system is improved by presetting the two CSBs to a preset pressure range. In this embodiment, due to the limitations of the 3D printed parts, the maximum pressure of the CSBs is set to 15psi.

[0039] When the hydraulic system is activated, the two CSBs are first pumped to the preset pressure range (inlet positive pressure and outlet negative pressure). Determining the pressure range allows all actuators to operate within the safe working pressure range and can significantly shorten the response time. When the two CSBs are exposed to water, they also act like water depth pressure compensators. So no matter what the depth is, the two pumps only need to provide the pressure difference. When the command signal is received and the actuator needs to extend or retract, the control unit opens the corresponding valve and the water is first released from (inlet) / to (outlet) the two CSBs. When the pressure in the two CSBs exceeds the preset range, the two pumps will be activated again to keep the pressure range within the preset value.

[0040] Design of hybrid manipulator

[0041] A hybrid design of the underwater manipulator was conducted. On the one hand, flexible robots have inherent compliance advantages over traditional rigid robots in complex underwater environments. On the other hand, better precision and larger payloads can be achieved by applying rigid parts in the design.

[0042] Underwater manipulator( Figure 4) consists of two parts: a gripper and a wrist. The gripper has two fingers, both covered with soft tissue to increase friction. The two fingers are mounted on the gripper base, and two gripper actuators are fixed between the fingers and the base. The gripper closes when the two actuators are stretched, with the two actuators moving simultaneously. The maximum opening angle b is 85°. The wrist consists of three wrist actuators, a ball joint, and two 3D printed plates. The upper and lower plates are connected by a ball joint, and three flexible actuators are placed in a triangle along the edge of the plate. When one actuator is stretched, the other two actuators contract accordingly, which results in a bending motion together with the contracted actuator. The maximum bending angle a is 36°.

[0043] Design and modeling of CSB and actuators

[0044] Another component of the hydraulic system, the two CSBs play an incomparable role in the performance of the entire system. In this case, the development of the two CSBs is separated from the hydraulic system to elaborate their functions in a more detailed way. However, compared with CSBs, flexible actuators have the same importance and similar manufacturing processes. In this case, the development of the actuators is put together with CSBs for better understanding.

[0045] Design and manufacture of CSB and actuators

[0046] The purpose of the CSB proposed in this embodiment is to reduce the volume and complexity of the system while optimizing the output of the pump so that the control response time can be significantly reduced. In order to achieve these goals, at least one of the following three conditions must be met: (1) The ambient water depth pressure can be passively balanced to remove the water depth factor from the development stage. (2) The pressure difference between the CSB and the actuator is large enough to achieve a fast response, but not so large as to cause damage to the actuator. (3) When the pressure increases, the CSB needs to have sufficient volume change for the actuator. Taking advantage of the inherent compliance and waterproofness of the flexible robot, by exposing the two CSBs to water, the ambient pressure no longer needs to be considered during the development stage. In order to further study the relationship between the performance and parameters of the CSB, an analytical model was established in the design and modeling of the CSB and actuator parts. For ease of study, the two CSBs were designed to deform (stretch and shrink) only on one axis.

[0047] There are two CSBs for the hydraulic system, one at the inlet (stretch) and the other at the outlet (retract). They are both the same shape, but the stretch one has a fiber reinforcement layer, while the retract one is pure silicone. The molding process of the inlet CSB is as follows Figure 5As shown in a. The inner layer is molded first (dragon skin 30; Smooth-On, Inc.). After laying the fiber reinforcement, the outer layer is molded (dragon skin 30; Smooth-On, Inc.). When it solidifies, the 3D printed mold core is removed and the fixture is added. For the export CSB, since there is no fiber reinforcement, it is only molded once (dragon skin 30; Smooth-On, Inc.).

[0048] In order to reduce system complexity and realize modular actuator design, the five flexible actuators used in the gripper and wrist have the same structure but different lengths to meet different requirements. In particular, the two actuators used in the gripper are shorter to achieve a larger opening angle, while the three actuators used in the wrist are longer to achieve a larger bending angle. The molding process is as follows ( Figure 5 b): The soft chamber is first formed using a 3D printed mold (dragon skin 10; Smooth-On, Inc.). Once the silicon is cured, a fiber reinforcement layer is added to limit radial extension. The two sides of the soft chamber are then sealed with a 3D printed fixture. Combined with the design of the hybrid manipulator, the actuator length and inner chamber radius can be modified as needed without redesigning the overall structure of the soft actuator.

[0049] CSB analytical modeling

[0050] In order to simplify the analytical model, all materials used in CSB are considered incompressible and the NeoHookean hyperelastic material model is used. The strain energy density function is given by Eq. (1).

[0051]

[0052] Where μ is the initial shear modulus and I1 is the first input variable of the three principal stretch ratios λ1, λ2, and λ3:

[0053]

[0054] The principal stress is given by equation (3).

[0055]

[0056] where p is the Lagrange multiplier. For incompressible materials, λ1λ2λ3=1. For the inlet CSB( Figure 5 c), since there is a fiber reinforcement layer, the circumferential strain is considered to be negligible, and we get λ 2e = 1. Then we can get by equation (4).

[0057]

[0058] for Figure 5 The outlet CSB shown in c has two circular 3D printed parts at both ends, so the circumferential strain at these two points is zero, and λ 2e =1. The principal circumferential stretch at the center of the CSB is k (k is a constant, and 0>k≤1). It can be obtained by formula (5).

[0059]

[0060] For simplicity, although the inlet CSB is composed of different materials with different initial shear moduli, they are considered to be a uniform material with an effective initial shear modulus. For the inlet CSB, the radial stress is considered to be balanced through the thickness (i.e., s3 = 0). Considering the similar derivation process in the study of Polygerinos et al., Equation (3) is further simplified to

[0061]

[0062] For the export CSB, substituting equations (1), (2), and (5) into equation (3), we can obtain

[0063]

[0064]

[0065]

[0066] For inlet CSB, balance is achieved at both ends, giving equation (10).

[0067] P e A e1 =s 1e A e2 , (10)

[0068] Among them, P e is the pressure difference between the inside and outside of CSB, A e1 is the cross-sectional area of ​​CSB, A e2 is the cross-sectional area of ​​silicon. Assuming the initial length of CSB is l e , radius r e , wall thickness t e , Substituting formula (3) into formula (10) we get

[0069]

[0070] For the outlet CSB, when k = 1, the pressure difference P c The force generated is equal to the force generated by the axial stress. min When the radial stress is given by the pressure difference P c To balance. Then through equation (12) and equation (13) we can get

[0071] P c A c1 =s 1c A c2 (12)

[0072] P c =s 3c . (13)

[0073] The initial conditions of CSB are length l c , radius r c and wall thickness t c Combining equations (7) and (9) with equations (12) and (13), we can obtain

[0074]

[0075]

[0076] Substituting equation (15) into equation (14), we can obtain

[0077]

[0078] Assume that the pressure inside the actuator is P c And water passes through CSB and the actuator is incompressible, giving the Dary-Weisbach equation as Equation (17).

[0079]

[0080] Where ΔP is the pressure difference between the CSB and the actuator, L is the pipe length, ρ is the water density, v is the average flow velocity, and D is the pipe inner diameter. For simplicity, although there are different fittings and pipes of different diameters, they are considered to be a single pipe with an effective length of f D is the Darcy friction factor for laminar flow, where R e is the Reynolds number. When the flow is turbulent, the Swamee-Jain equation applies

[0081]

[0082] Where ε is the absolute roughness. Substituting equation (18) into equation (17), we get

[0083]

[0084] CSB verification experiment

[0085] The working performance of CSB is based on at least one of the following two characteristics: (1) the CSB has a large enough deformation to store enough water for the actuator. (2) the pressure difference between the CSB and the actuator is large enough to generate a large flow rate and increase the response time. In this case, the pressure-strain relationship of CSB is first tested. Then, the relationship between flow rate and pressure difference is tested on the proposed CSB-based hydraulic system. Finally, the performance of the entire hydraulic system is evaluated compared with the condition without CSB.

[0086] Experimental setup

[0087] Experimental setup Figure 6 As shown in the figure. The hydraulic path in the experiment is the same as the hydraulic system mentioned in the conceptual design and system overview section, except that there is only one actuator in the experimental setup. Three pressure sensors (TSCDRRN015PDUCV; Honeywell International Inc.) are used to record the pressure of the CSB and the actuator, respectively. Two flow sensors (FS2012-1001-LQ; Integrated Device Technology, Inc.) are placed in the inlet path and outlet path just after the valve for flow measurement of the CSB. The actuator and CSB used in the experiment are the same as those in the hydraulic system and the underwater manipulator. The actuator and the two CSBs are fixed to the platform at one end and can move freely at the other end. The length of the pipe connecting the CSB and the actuator to the pressure sensors is set so that the resistance of the bending pipe is negligible when the CSB or the actuator is deformed.

[0088] Pressure and strain verification

[0089] In this section, the relationship between pressure and strain was first tested. The principal stretch ratios (λe and λc) were measured for each 1 psi increase in pressure. This process was repeated five times and the average value was used for each point. Figure 5 d shows the test results. The two blue curves are the analytical models of equations (11) and (16), where the effective initial shear modulus of the inlet CSB is 65 psi and the effective initial shear modulus of the outlet CSB is 40 psi. Both models fit the experimental data well; the root mean square error (RMSE) of the inlet CSB is 0.012 and the root mean square error (RMSE) of the outlet CSB is 0.005, indicating that the two analytical models fit the experimental data well. By using the pre-established model, the CSB can be specifically designed according to the amount of water that needs to be stored.

[0090] Verification of the relationship between flow rate and pressure difference

[0091] Since the analysis models of the two CSBs are different, the relationship between the pressure difference and flow rate of the two CSBs was tested separately. The test steps for the inlet CSB are as follows: (1) Set the pressure of the inlet CSB to 13psi and start pumping water into the CSB. The inlet valve remains closed and the pressure inside the actuator is 0. (2) When the CSB reaches the preset pressure, turn off the inlet pump. Then, open the inlet valve, and water begins to flow from the CSB to the actuator due to the pressure difference. The flow rate and pressure of the CSB and the actuator were measured during this process. (3) When the pressure difference between the CSB and the actuator reaches 0, close the inlet valve. (4) The above steps are repeated 5 times. The outlet CSB follows similar steps, except that the outlet CSB is preset to negative pressure and the actuator is filled with water (pressure = 6psi), so the initial pressure difference is also 13psi, and there is enough water inside the actuator for the outlet CSB to suck out. The test results are shown in the figure below. Figure 5 e; the effective lengths of the inlet CSB and outlet CSB are 0.5m and 1m respectively. Both analytical and experimental results show that the flow velocity of both CSBs increases rapidly with the increase of the pressure difference. However, when the flow becomes turbulent, the increase in flow velocity decreases significantly with the increase of the pressure difference.

[0092] Tracking performance evaluation

[0093] Four different system configurations were tested to evaluate the tracking performance of the system:

[0094] Sa: One wrist actuator, one inlet pump, no CSB.

[0095] Sb: one wrist actuator, one inlet CSB, one inlet pump.

[0096] Sc: A hydraulically controlled wrist actuator with inlet and outlet CSB and their corresponding pumps.

[0097] Sd:Sc with shorter gripper actuator.

[0098] Three different pressure signals are given for the four devices:

[0099] Pa: is a step function with a time interval of 6s.

[0100] Pb: is a 0.25Hz sinusoidal signal.

[0101] Pc: is a sinusoidal waveform with increasing frequency.

[0102] The results are as follows Figure 7 And as shown in Table 1.

[0103] For Pa tracking ( Figure 7a and Table 1), the rise time of Sb and Sc is half that of Sa, and the fall time of Sc is half that of Sa and Sb. In addition, the rise and fall time of Sd is significantly shorter (up to 687%) compared with the other three systems, indicating that under the same CSB device, the smaller the actuator, the faster the charge and discharge rate. In the Pb tracking test ( Figure 7 In b), the RMSE of Sc and Sd is much smaller than that of Sa and Sb, while Sd has the smallest RMSE (0.19psi). For the sake of comparison and easy understanding, when evaluating the tracking performance of Pc ( Figure 7 c-7f), we assume that the system is able to track a given signal when the RMSE is below 10% (0.4psi). Similarly, the threshold (critical time) is recorded as a reference when the RMSE is exceeded for the first time. The critical times of Sb and Sc are larger compared to Sa (Table 1). Sd has similar performance compared to Sc; one possibility is that although Sd requires less water for the same pressure change, the mechanical properties of the valve cannot provide sufficient switching frequency. In summary, it is demonstrated that the CSB-based hydraulic system can significantly improve the system response time compared to the system without CSB.

[0104] Underwater control system experiment

[0105] According to the design concept, in addition to fast response time, the underwater manipulator system must also meet at least one of the following requirements: (1) be compact; (2) have a wide range of operating depths. In this case, the manipulator was installed in a consumer-grade ROV and tested at different depths. Similarly, a specially designed task was performed to test the manipulator's dexterity and accuracy.

[0106] Flexible actuator experiment

[0107] As shown in the experimental setup section, the five actuators used in the manipulator follow the same design for modularity and simplification. To prove the feasibility of this design, the stress-strain relationship and pressure-force relationship of the actuators were tested. Figure 8 f. Except for one end of the actuator that can move along the guide rod, the rest of the parts are fixed. The friction between the guide rod and the actuator can be ignored (<0.1N).

[0108] The results of stress and strain tests are shown in Figure 8 a and Figure 8b. Both actuators have small stretching in the initial stage. When the pressure reaches between 30kPa and 40kPa, the deformation of the wrist actuator and the gripper actuator becomes more significant. As the pressure continues to increase, the stretching slows down again. The actuators in the same group have little difference in the stress-strain relationship. For the wrist actuator, the average error is 0.022, and for the gripper actuator, the average error is 0.0138. The verification results of pressure and force are shown in Figure 8 c and Figure 8 As shown in Figure d, the fitting curve fits the measured data well. For the wrist actuator, the RMSE is 0.58 N, 0.48 N, and 0.26 N, respectively, and the overall RMSE is 0.81 N. For the gripper actuator, the RMSE is 0.49 N and 0.39 N, respectively, and the overall RMSE is 0.44 N.

[0109] Hybrid underwater gripper experiment

[0110] A bench test was first performed to evaluate the gripping force of the gripper. A force sensor ( Figure 8 g), and set the inlet CSB to three different pressure ranges to observe the effect of CSB on the clamping force. Figure 8 As shown in Figure 5, there is a positive correlation between the CSB working pressure and the clamping force, which means that the clamping force can be modified by adjusting the CSB working pressure range.

[0111] The maneuvering system was also tested at different water depths to demonstrate that the CSB-based hydraulic system is decoupled from the water depth. The use of a clay-filled 3D printed box for force measurement was recommended for at least one of the following three reasons: Fig. 9 b): (1) Ability to maintain shape and long-term recording after testing and leaving the water (in contrast, a spring scale will return to normal length after the clamp is released). (2) Can directly compare results between different locations and depths, facilitating visualization for later analysis. (3) As a passive device, it does not require additional power or wiring, making it easy to implement in multi-site testing. Mount the box on the tip of the clamp ( Fig. 9 a), measure the groove depth using the clamping force ( Fig. 9 b). To limit variables, all boxes were weighed between 10.1g and 10.2g, the clamping time was set to 15s, and the CSB operating pressure was set to 13psi to 14psi. This test was conducted in three different locations: onshore, in a water tank, and on the seafloor; the onshore test was used as a reference, using three boxes at each depth.

[0112] The results are as follows Fig. 9 a. The groove depth measured on land is 6.05 mm (blue dashed line), and the remaining points are evenly distributed along the straight line, with a root mean square error of 0.3 mm. The results show that the CSB-based hydraulic system is independent of the ambient water depth.

[0113] Hybrid underwater maneuvering system experiment

[0114] Experimental setup Fig.10 b. Two 3D printed rings are placed on either side with a thin stick in the middle. A person needs to use a controller to control the robot and pick up the two rings and place them through the middle stick ( Fig.10 c). To quantify the performance of the entire system, assume that the inner diameter of the ring is D and the diameter of the rod is 3mm. If the manipulator can complete the task, its accuracy is D-3mm. This task was performed both on land and underwater. In both cases, the manipulator was able to complete the task with an accuracy of 1mm. To evaluate the impact of the manipulation system on an underwater platform, the entire system was mounted on a consumer-grade ROV for speed testing. The ROV weighs about 3.8kg and has dimensions of 465x270x126mm. Fig.10 As shown in a, the ROV needs to pass through a 15-meter-long water tank. Fig.10 d, right), and another without a control system ( Fig.10 d, left). The ROV was rated at medium power because of the tension on the cable and the difficulty in controlling the direction in a narrow tunnel at high speeds. Both devices were tested three times. The results showed that when the manipulator system was on the ship, its average speed (0.28m / s) was 37.8% lower than when there was only the ROV (0.45m / s). A major reason for the speed loss is that the manipulator system significantly increases the cross-sectional area of ​​the ROV and destroys the streamlined shape of the ROV. Given the ocean current speed (0.08 to 0.25m / s), the results show that installing the manipulator will reduce the speed of the ROV, but it is still within the feasible range.

[0115] This paper describes an underwater hybrid manipulator with a lightweight CSB-based hydraulic actuation control system. The hybrid manipulator is designed in such a way that the flexible actuator design is decoupled from the manipulator motion generation. Two CSBs are added to the hydraulic system to balance the ambient water pressure, reduce the system volume, and improve the dynamic response. An analytical model is built to explore the deformation, pressure, and flow characteristics of the CSBs, which are well matched with the experimental results. In addition, the tracking performance of the proposed hydraulic system is tested, and the CSBs can significantly improve the system performance compared to the condition without CSBs. The stress-strain and pressure-force relationships of the flexible actuators are tested, and different individuals show limited differences, which proves the feasibility of this modular design. Then, the gripping force and accuracy of the hybrid manipulation system are tested on land and underwater (see Appendix). The results show that the water depth has no significant effect on the manipulator performance, and the proposed underwater manipulation system can accurately handle underwater tasks even without precise control feedback. Finally, speed tests are conducted to evaluate the compactness of the system, and the results show that even a consumer-grade ROV can move at a reasonable speed (0.28m / s) using the underwater manipulation system described in this paper.

[0116] The devices, systems and methods herein allow that depth is no longer a factor in determining underwater operational performance by hybrid manipulators and CSB-based hydraulic systems, and even compact subsea platforms (<4kg) can manage complex underwater sampling tasks as heavy and expensive devices. Since large and expensive subsea platforms are no longer necessary units for shallow water sampling, the combined design concept of manipulators and hydraulic systems has become a benchmark for reducing system size and cost. In addition, the availability of compact commercial subsea platforms has greatly broadened the population in this field.

[0117] A prototype of the proposed soft robotic manipulator system has been field tested in three tropical and subtropical regions ( Fig.11 ). These areas are hotspots of marine biodiversity with the following hydrological and climatic conditions:

[0118] 1) Tropical waters of Hong Kong, China (pH 8.4) with low temperature (21-23 degrees Celsius) and low salinity (27-28 PSU for open seawater, salinity not suitable for indoor tank freshwater): Four test locations: a) indoor tank at shallow depth and still water; b) indoor flume; c) South Beach; d) West Pier Port Area.

[0119] 2) Taiwan, China, Guishan Island, subtropical waters (25m), active submarine volcano, harsh environment (pH = 6.1 strong acid (sulfur)), maximum temperature (33-36 degrees Celsius).

[0120] 3) Maldives, Kandima Island Coral Lagoon, shallow (2 meters) equatorial water (pH 8.5), moderate temperature (28-29 degrees Celsius) and high salinity (34 PSU).

[0121] Unless otherwise indicated in the examples and elsewhere in the specification and claims, all parts and percentages are by weight, all temperatures are in degrees Celsius, and pressures are at or near atmospheric.

[0122] For any graph or numerical range for a given feature, a graph or parameter from one range can be combined with another graph or parameter from a different range for the same feature to generate a numerical range.

[0123] Except in the operating examples or where otherwise indicated, all numbers, values ​​and / or expressions concerning quantities of ingredients, reaction conditions, and so forth used in the specification and claims are to be understood as being modified in all instances by the word "about".

[0124] Although the present invention is explained with respect to specific embodiments, it should be understood that various modifications thereof will be apparent to those skilled in the art after reading the specification. Therefore, it should be understood that the invention disclosed herein is intended to cover such modifications within the scope of the appended claims.

Claims

1. A device for underwater applications, comprising: Hybrid underwater manipulator; and A hydraulic control structure includes at least two compensation bladders, the hybrid underwater manipulator is connected to and actuated by the at least two compensation bladders, The hybrid underwater manipulator comprises a joint and a clamp, wherein the joint and the clamp are respectively connected to the at least two compensation bladders, and a preset pressure difference is generated between the at least two compensation bladders and the joint and the clamp.

2. The apparatus for underwater applications according to claim 1, wherein the hydraulic control architecture comprises an inlet compensation bladder and an outlet compensation bladder.

3. The device for underwater applications according to claim 1, wherein the weight of the device in water is zero.

4. The apparatus for underwater applications of claim 1, further comprising a sensor in each actuator.

5. The apparatus for underwater applications of claim 1, wherein the hybrid underwater manipulator comprises a manipulator arm.

6. The device for underwater applications according to claim 1, wherein the weight of the device on dry land is less than or equal to 2.5 kg.

7. The device for underwater applications of claim 1, wherein the weight of the device in water is approximately zero.

8. The device for underwater applications according to claim 1, wherein the device has two or more control circuits.

9. An underwater remotely operated submersible comprising: Device for underwater applications according to any one of claims 1 to 8.

Images