A tactile body and tactile system for soft manipulators

Abstract

The present invention discloses a tactile body and a tactile system for a soft manipulator. The tactile body is mainly composed of a plurality of liquid metal sensing bodies, each of which includes an elastic tube, liquid metal filled in the elastic tube, a flexible metal wire, a pressure pipe and a hydraulic measurement sensor. The tactile body is formed by weaving a plurality of liquid metal sensing bodies, thereby enhancing the sensitivity of the sensing body to touch, which is much stronger than a single non-wound sensing body. By mapping the tactile sense and resistance of the plurality of liquid metal sensing bodies, the size and position of the touch can be obtained, thereby solving the technical path problem of pressure touch. The tactile body is installed in the execution end of the soft manipulator, and cooperates with the signal acquisition module and the data processing module to obtain frequency characteristics, and the numerous characteristics are used as input samples of the neural network. After the network is fully trained, a variety of tactile indicators are obtained, and the tactile results are obtained by compressing the various tactile indicators.

Description

Technical Field

[0001] The present invention relates to the field of tactile technology for soft manipulators, and in particular to a tactile body and a tactile system for soft manipulators. Background Art

[0002] Global demand for robots continues to expand, with applications in various fields, including production and daily life. Robots, as the end effectors that enable robots to interact with their workpieces, are a key development focus in the robotics industry. Early robotic arms were mostly rigid grippers. Over the years, these grippers have evolved from single, distributed rigid grippers to various types of bionic, dexterous hands, integrating various tactile and visual sensors to meet the diverse demands of industrial and automated tasks.

[0003] However, as robotic arms are increasingly used for complex tasks such as grasping fragile objects and human-robot collaboration, rigid manipulators are struggling to meet these requirements. Soft manipulators are becoming increasingly common. Soft manipulators are lightweight and typically made of rubber, polymers, and multifunctional materials. Their nonlinear, viscoelastic, and hysteretic properties offer advantages such as high adaptability and safety.

[0004] At present, the realization of tactile perception for soft manipulators mostly relies on two methods: (1) Based on the principle of resistance or capacitance change of conductive materials under strain. These sensing units themselves have a certain degree of flexibility, but the elastic modulus is generally larger than that of silicone materials, which has a certain impact on the movement of soft manipulators. (2) Relying on embedded fiber optic sensors (arrays), the soft manipulator can sense the intelligent recognition of touch, position, shape, object composition, temperature, vibration, etc. Based on the above methods, the soft manipulator's perception of the tactile information of the object surface is still relatively difficult, and its implementation scheme is still relatively rough. Most tactile perception uses pressure sensing to obtain the pressure exerted on the manipulator, but it cannot perceive the roughness of the object surface and the local shape. Therefore, a tactile body and tactile system for soft manipulators are urgently needed to solve the above problems. Summary of the Invention

[0005] The present invention provides a tactile body and a tactile system for a soft manipulator, which can obtain the size and position of the tactile sensation, thereby obtaining a pressure tactile sensation, to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a tactile body for a soft manipulator, which is mainly composed of multiple liquid metal sensing bodies, wherein each of the liquid metal sensing bodies includes an elastic tube, liquid metal filled in the elastic tube, a flexible metal wire, a pressure pipe and a hydraulic measurement sensor. The flexible metal wire is connected in series with the liquid metal to form a series resistor, and a mapping relationship between touch and resistance is obtained by changing the series resistance. One end of the pressure pipe is inserted into the elastic tube and contacts the liquid metal, and the other end is connected to the hydraulic measurement sensor to transmit the change of liquid metal pressure to the hydraulic measurement sensor.

[0007] Preferably, both ends of the elastic tube are provided with sealing heads to prevent leakage of liquid metal.

[0008] Preferably, the tactile body is formed by weaving N longitudinal liquid metal sensing bodies and M transverse liquid metal sensing bodies in a warp and weft manner.

[0009] Preferably, the N longitudinal liquid metal inductors and the M transverse liquid metal inductors are woven in a winding manner at at least some of their intersections.

[0010] Preferably, each of the liquid metal inductors comprises a plurality of winding sections, wherein the resistance of the non-winding section is constant, and the resistance of the winding section of the liquid metal inductor is constant. ρ is the resistivity of liquid metal, L is the length of each liquid metal segment, and S is the cross-sectional area of liquid metal;

[0011] The relationship between the resistance R′ of the winding section of a single liquid metal inductor after being squeezed by contact and the resistance R before contact and squeeze is R′=R*T, where, ΔS is the change in the cross-sectional area of the liquid metal.

[0012] Preferably, the relationship between the winding segment resistance matrix R′ of N*M dimensions and the deformation matrix T of N*M dimensions is:

[0013] R′1=R1*T;

[0014] R′2=R2*T;

[0015] Among them, the deformation matrix T is:

[0016]

[0017] Measure the values of R1, R2 and R'1, R'2 through the circuit and calculate the cross-sectional area deformation ratio The relationship between T and haptic size and position.

[0018] A tactile system for a tactile body of a soft manipulator includes a tactile body, a signal acquisition module, and a data processing module. The tactile body is installed in the execution end of the soft manipulator. The signal acquisition module collects signals generated by the tactile body at the execution end. The data processing module processes the signals to obtain frequency characteristics and uses the numerous characteristics as input samples of a neural network. After the network is fully trained, multiple tactile indicators are obtained, and the tactile results are obtained by compressing the individual tactile indicators.

[0019] Preferably, the execution end of the soft manipulator includes the fingers, palm and back of the hand of the soft manipulator, and the size of the tactile body is woven based on the size of the execution end.

[0020] Compared with the existing technology, the present invention has the following advantages: In the present invention, the touch sensing body is formed by weaving multiple liquid metal sensing bodies, which enhances the sensitivity of the sensing body to touch, making it much stronger than a single non-entangled sensing body. By mapping the tactile sensation of multiple liquid metal sensing bodies to resistance, the tactile size and position can be obtained, solving the technical path problem of pressure touch.

[0021] In the present invention, the tactile system is combined with the soft manipulator, so that the soft manipulator has the ability to perceive touch, and the functions of the signal acquisition module and the data processing module are utilized to take or process the signal to obtain frequency characteristics, and the numerous characteristics are used as input samples of the neural network. After the network is fully trained, a variety of tactile indicators are obtained. Among them, the vibration caused by the relative movement of the soft manipulator and the surface of the object can be obtained with the help of the rough surface contact mechanics theory, and the frequency spectrum of the hydraulic sensor signal can be analyzed to solve the technical path problem of roughness touch. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0023] In the attached figure:

[0024] Figure 1 It is a schematic structural diagram of the tactile body grid of the present invention;

[0025] Figure 2 This is a schematic structural diagram of a single liquid metal induction body of the present invention;

[0026] Figure 3 This is a schematic structural diagram of the resistance distribution of a single liquid metal inductor of the present invention;

[0027] Figure 4 This is a schematic diagram of the structure of a single liquid metal induction body braided in the present invention;

[0028] Figure 5This is a schematic diagram of the structure of the vertically crossed liquid metal induction body braided in the present invention;

[0029] Figure 6 Schematic diagram of the structure of the N*N dimensional resistor matrix of the present invention;

[0030] Figure 7 This is a schematic diagram of the structure of the integrated tactile body and the soft manipulator of the present invention;

[0031] Figure 8 Schematic diagram of the structure of the roughness touch on the soft robotic finger of the present invention;

[0032] Figure 9 It is a schematic diagram of the structure of the comprehensive acquisition and analysis of the tactile signal of the present invention;

[0033] Figure 10 Schematic diagram of the application of the touch sensing system of the present invention in aircraft skin inspection;

[0034] Numbers in the figure: 1. Elastic tube; 2. Liquid metal; 3. Flexible metal wire; 4. Pressure pipe; 5. Hydraulic measurement sensor. DETAILED DESCRIPTION

[0035] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0036] Example: Figure 1 As shown, a tactile body for a soft manipulator is mainly composed of multiple liquid metal sensing bodies;

[0037] Among them reference Figure 2 As shown, each liquid metal sensor includes an elastic tube 1, liquid metal 2 filled in the elastic tube, a flexible metal wire 3, a pressure pipe 4 and a hydraulic measurement sensor 5. The flexible metal wire 3 is connected in series with the liquid metal 2 to form a series resistor. When an external object contacts the liquid metal sensor, the elastic tube 1 is deformed, causing the series resistance to change. By measuring the resistance change, the mapping relationship between touch and resistance can be obtained. One end of the pressure pipe 4 is inserted into the elastic tube and contacts the liquid metal, and the other end is connected to the hydraulic measurement sensor 5 for transmitting the change in liquid metal pressure to the hydraulic measurement sensor. The hydraulic measurement sensor is used to measure the hydraulic value of the liquid metal.

[0038] Among them, liquid metals include gallium-aluminum alloys, gallium-bismuth alloys, gallium-tin alloys, gallium-indium alloys, bismuth-tin alloys, tin-bismuth alloys, mercury, etc. The elastic tube can be a hollow polystyrene (SEBS) elastic hose, a synthetic rubber (TPE) elastic tube, etc.; both ends of the elastic tube are provided with sealing heads to prevent liquid metal leakage; the flexible metal wire can be conducted by braiding non-ferrous metals such as Au, Ag, Cu and Al.

[0039] In the present invention, reference Figure 1 The tactile body is formed by weaving N longitudinal liquid metal sensors and M transverse liquid metal sensors in a warp and weft manner, wherein the N longitudinal liquid metal sensors and the M transverse liquid metal sensors are woven in a winding manner at least at some of the intersections. The two cross-entangled liquid metal sensors have a mutually constrained effect due to their cross-entangled parts. When they are squeezed by external force, their effective area under pressure is larger than the force-bearing area of a single liquid metal sensor. Therefore, their pressure sensitivity is also greater than that of a single liquid metal sensor. It has greater sensitivity and can perform pressure positioning. Through experimental measurement, the sensitivity of the braided sensor is more than ten times that of the unbraided sensor.

[0040] The relationship between touch sensation and resistance in a single liquid metal sensor is as follows:

[0041] refer to Figure 3 As shown, after a single inductor is partially woven, multiple local windings are generated, and its total resistance R can be regarded as a circuit composed of N cross points connected in series (refer to Figure 3 , including 5 resistors), can be simplified to: R 单根 =R1+R2+R3+R4+R5;

[0042] refer to Figure 3 Left figure, through experimental measurement, we can approximate Figure 3 In the left figure, the resistance of the non-wound section R2, R4 and R5 is considered as a constant. Its resistance value does not change much during the normal pressure process. R2=R4=R5=R 常数 ;

[0043] The resistance calculation formula of the winding section is: Where ρ is the resistivity of liquid metal, L is the length of each liquid metal segment, and S is the cross-sectional area of liquid metal;

[0044] When an external force contacts the liquid metal segment R1, its cross-sectional area and overall length at the contact deformation point change, causing its resistance to change. (Assuming that an external force occurs in segment R1, causing the resistance of segment R1 to change, the same principle applies to other segments). Its resistance after deformation can be expressed as:

[0045] Since the contact area in the sensor is often very small, the contact length is L 接触 , since the liquid metal caused by contact is transferred to other parts of R1 segment, ΔL is actually approximately:

[0046] Therefore, in each liquid metal segment, L, S and ρ are considered as constants, and the main factors affecting R are L 接触 and Among them, it is obvious The impact is greater:

[0047]

[0048] make:

[0049] The relationship between the resistance R′1 of the winding section after contact and compression and R1 before contact and compression is: R′1=R1*T;

[0050] Therefore, from the change of the resistance value of a single winding segment, it is possible to sense whether there is a touch and the degree of deformation of the sensing body, as well as the degree of change in the cross-sectional area. It is the main factor affecting the change of resistance R′1.

[0051] The weaving method is as follows:

[0052] The first step is to refer to Figure 4 As shown, a single liquid metal induction body is woven around the weaving auxiliary rod using a weaving auxiliary rod, and then fixed with a clamp after weaving;

[0053] The second step is to weave the remaining parallel inductors in sequence. When weaving, sufficient space is reserved between each two inductors for the installation of the inductors in the vertical dimension.

[0054] Step 3: Reference Figure 5 As shown, to weave the vertically crossed liquid metal induction body, the braided body in the previous step must first be frozen and cooled to below the solidification temperature of the liquid metal and maintained for a certain period of time; then, after removing one braiding auxiliary rod, the vertically crossed liquid metal induction body can be weaved;

[0055] The fourth step is to weave the remaining vertically crossed liquid metal sensing bodies in sequence. Before each weaving, the woven body must be cooled and solidified, and then taken out for easy weaving, and finally the tactile body grid is obtained. Figure 1 shown.

[0056] The relationship between the N*M dimensional winding segment resistance matrix R′ and the N*M dimensional deformation matrix T is:

[0057] R′1=R1*T;

[0058] R′2=R2*T;

[0059] The calculation symbol * represents the Hadamard product; the deformation matrix T is:

[0060]

[0061] refer to Figure 6 As shown, the circle represents the cross-wound segment of two inductors, and the rest is the non-wound segment. Using the N*N dimension resistance matrix R1 and R2, after the weaving of multiple vertically crossed liquid metal inductors is completed, it can be obtained by measurement, where:

[0062]

[0063]

[0064] The resistance of the non-wound section is a constant R 常数 , the initial value of the winding resistance is R 缠绕 , can be obtained through experimental measurement.

[0065]

[0066]

[0067] get:

[0068]

[0069]

[0070] Substituting the relationship between the N*M dimensional winding segment resistance matrix R′ and the N*M dimensional deformation matrix T, the deformation matrix T is obtained as follows:

[0071]

[0072] In this embodiment, the values of R1, R2 and R′1, R′2 are measured through the circuit, and T=R -1 R′ obtains some values in the sparse matrix T; among them, only T 12 、T 14 、T 21 、T 23 、T 32 、T 34 、T 43 、T45 、T 52 、T 54 Not 0, the rest are zero;

[0073] After obtaining the above tactile deformation matrix, the formula Calculate the cross-sectional area deformation ratio The relationship between T and haptic size and position.

[0074] A tactile system for a tactile body of a soft manipulator includes a tactile body, a signal acquisition module, and a data processing module. The tactile body is installed in the execution end of the soft manipulator. The signal acquisition module collects the signal generated by the tactile body at the execution end. The data processing module processes the signal to obtain frequency characteristics and uses the numerous characteristics as input samples of a neural network. After the network is fully trained, multiple tactile indicators are obtained, and the tactile results are obtained by compressing each tactile indicator.

[0075] The execution end of the soft manipulator includes the fingers, palm and back of the hand. The network size can also be expanded or reduced as needed. For example, the minimum dimension is a 2*2 sensor weaving network, or it can be an N*M or N*N network. However, its design needs to meet T=R -1 Under the premise that the R′ formula is solvable.

[0076] In this embodiment, reference Figure 7 As shown, the tactile body can be designed and manufactured as an integral part of the fingers of the soft manipulator. In the fingers or palm of the soft manipulator, the protrusion of the winding section should be kept higher than the finger contact surface or buried in a very thin silicone layer to ensure high sensitivity.

[0077] The above-mentioned soft manipulator is used to sense the roughness of the surface of the object being measured:

[0078] The soft manipulator with the tactile body contacts and slides with the object to be measured. This scenario can be regarded as the friction between a highly elastic body and a rigid rough surface. Figure 8 As shown, in the friction of high-elastic body, the surface roughness of the high-elastic body plays only a small role, and the friction force mainly depends on the roughness of the rigid surface;

[0079] When a soft manipulator with a tactile body generates friction with the object being measured, the friction and the resulting vibration mainly depend on the state of the surface of the object being measured. The time it takes for the soft finger to travel a distance r of the microscopic feature size on the surface of the object being measured at a speed v is:

[0080]

[0081] During the sliding process, the order of magnitude of the characteristic frequency is:

[0082]

[0083] Based on the above, we can know that at a constant speed v, the vibration of the sensing body causes the pressure of the liquid metal inside the sensing body to change. By analyzing the spectrum of the liquid metal pressure value, the roughness of the surface of the touched object can be obtained;

[0084] Among them, reference Figure 9 As shown, two types of perception can be used, and signal acquisition software and hardware can be used to perform signal preprocessing and obtain frequency characteristics. Many features can be used as input samples of the neural network. After the network is fully trained, a variety of tactile indicators can be obtained, which can be used as effective criteria for the surface state of the object being measured.

[0085] In addition, the tactile body can be used in the evaluation of the surface condition of industrial products, the intelligent perception of medical robots, the flexible grasping of smart agriculture, etc. Figure 10 As shown in the figure, this is an application case of state perception in aircraft skin. It can identify the state of the skin and its attachments, such as damage cracks, paint peeling, icing and other faults, and can be used as one of the effective tools for intelligent detection and maintenance.

[0086] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A tactile body for a soft robot arm, characterized by: The tactile sensor is primarily composed of multiple liquid metal sensing bodies, each of which includes an elastic tube, liquid metal filled within the elastic tube, a flexible metal wire, a pressure pipe, and a hydraulic pressure measurement sensor. The flexible metal wire is connected in series with the liquid metal to form a series resistor. The mapping relationship between tactile sensation and resistance is obtained by changing the series resistance. One end of the pressure pipe is inserted into the elastic tube to contact the liquid metal, and the other end is connected to the hydraulic pressure measurement sensor to transmit changes in the liquid metal pressure to the hydraulic pressure measurement sensor. The touch sensing body is formed by weaving N longitudinal liquid metal sensing bodies and M transverse liquid metal sensing bodies in a warp and weft manner; The N longitudinal liquid metal inductors and the M transverse liquid metal inductors are woven in a winding manner at least partially at their intersections; Each of the liquid metal inductors comprises a plurality of winding sections, wherein the resistance of the non-winding section is constant and the resistance of the winding section of the liquid metal inductor is constant. ρ is the resistivity of liquid metal, L is the length of each liquid metal segment, and S is the cross-sectional area of liquid metal; The relationship between the resistance R′ of the winding section of a single liquid metal inductor after being squeezed by contact and the resistance R before contact and squeeze is R′=R*T, where, ΔS is the change in the cross-sectional area of the liquid metal. The relationship between the N*M dimensional winding segment resistance matrix R′ and the N*M dimensional deformation matrix T is: R1′=R1*T; R2′=R2*T; Among them, the deformation matrix T is: Measure the values of R1, R2 and R1', R2' through the circuit and calculate the cross-sectional area deformation ratio The relationship between T and haptic size and position.

2. The tactile body for a soft robot according to claim 1, characterized in that: Both ends of the elastic tube are provided with sealing heads to prevent leakage of liquid metal.

3. A tactile system for a tactile body of a soft manipulator according to any one of claims 1 to 2, characterized in that: The system includes a tactile body, a signal acquisition module and a data processing module. The tactile body is installed in the execution end of the soft manipulator. The signal acquisition module collects the signal generated by the tactile body at the execution end. The data processing module processes the signal, obtains the frequency characteristics, and uses the numerous characteristics as input samples of the neural network. After the network is fully trained, a variety of tactile indicators are obtained, and the tactile results are obtained by compressing each tactile indicator.

4. The tactile body and tactile system for a soft robotic arm according to claim 3, characterized in that: The execution end of the soft manipulator includes the fingers, palm and back of the hand of the soft manipulator, and the size of the tactile body is woven based on the size of the execution end.

Images