Tactile sensor and dexterous hand

Abstract

The application discloses a visual-tactile sensor and a dexterous hand. The visual-tactile sensor comprises a tactile module, a visual module and a connecting module. The tactile module comprises a shell, a transparent elastic body and a transparent support body. The shell is provided with a plug-in cavity and an opening communicating with the plug-in cavity. The transparent support body is arranged on the shell. The transparent elastic body is arranged on the transparent support body. The side of the transparent elastic body away from the transparent support body is provided with a preset pattern. The visual module is plug-in matched with the plug-in cavity through the opening. The connecting module is detachably connected to the shell or the visual module and covers the opening. The connecting module is used for connecting the dexterous hand. Thus, when the tactile module or the visual module is damaged, the dexterous hand does not need to be disassembled. The connecting module and the shell or the visual module are only disassembled, so that the disassembly of the components can be completed, and the maintenance time and the labor cost are greatly saved.

Description

Technical Field

[0001] This application relates to the field of robotics, and in particular to a visual-tactile sensor and a dexterous hand. Background Technology

[0002] In fields such as robotics, intelligent prosthetics, and human-computer interaction, the dexterous hand, as a key component for performing fine motor tasks, directly impacts the overall system performance due to its sensory capabilities. To endow dexterous hands with the ability to perceive physical information such as contact force, texture, and shape, visual-tactile sensors have been widely researched and applied. These sensors typically combine visual and tactile principles, using a camera to capture the deformation patterns of an elastic tactile layer under pressure, thereby retrieving the contact state and achieving high-resolution tactile perception.

[0003] However, in related technologies, the tactile components of visual-tactile sensors are often directly and rigidly fixed to the dexterous hand. When the visual-tactile sensor is damaged and needs to be replaced, the entire dexterous hand needs to be disassembled in a complicated manner. This is not only cumbersome and time-consuming, but may also cause mechanical damage or a decrease in accuracy due to repeated disassembly and assembly. Utility Model Content

[0004] This application provides a visual-tactile sensor and a dexterous hand, which can solve at least one of the above-mentioned technical problems.

[0005] In a first aspect, embodiments of this application provide a visual-tactile sensor, including:

[0006] A tactile module, comprising a housing, a transparent elastomer, and a transparent support, wherein the housing has a insertion cavity and an opening communicating with the insertion cavity, the transparent support is disposed on the housing, the transparent elastomer is disposed on the transparent support, and a preset pattern is provided on the side of the transparent elastomer facing away from the transparent support;

[0007] A vision module, wherein the vision module is inserted into the insertion cavity through the opening; and

[0008] A connection module is detachably connected to the housing or the vision module and covers the opening, and the connection module is used to connect a dexterous hand.

[0009] In some embodiments, the transparent support is disposed within the housing, and the side of the transparent support opposite to the transparent elastomer is provided with a support reflective surface, which is used to reflect the deformed preset pattern to the visual module.

[0010] In some embodiments, the transparent elastomer and the transparent support are stacked, and the transparent support and the housing cooperate to form a reflective cavity communicating with the insertion cavity. The cavity wall of the reflective cavity is provided with a reflective structure opposite to and spaced apart from the transparent support. The reflective structure is used to reflect the deformed preset pattern to the visual module.

[0011] In some embodiments, the reflective structure is a housing reflective surface or a reflective mirror.

[0012] In some embodiments, a stop structure is provided at the connection between the reflective cavity and the insertion cavity, and the vision module abuts between the stop structure and the connection module.

[0013] In some embodiments, the transparent support is integrally formed with the shell.

[0014] In some embodiments, the visual-tactile sensor further includes a first locking element threadedly connected to the housing and the vision module;

[0015] And / or, the visual-tactile sensor further includes a second locking element, which is threadedly connected to the vision module and the connection module.

[0016] In some embodiments, the connection module is provided with wire-passing holes that extend through both opposite sides of the connection module, and the wire-passing holes are used for the wires of the vision module to pass through.

[0017] In some embodiments, the connection module has multiple connectors on the side opposite to the vision module, the connectors being used to cooperate with a dexterous hand, and adjacent connectors being spaced apart.

[0018] Secondly, embodiments of this application provide a dexterous hand, which includes the visual-tactile sensor of any of the above embodiments, and the visual-tactile sensor is disposed on the dexterous hand.

[0019] The visual-tactile sensor and dexterous hand provided in this application embodiment include a tactile module, a vision module, and a connection module. The tactile module includes a housing, a transparent elastomer, and a transparent support. The housing has a insertion cavity and an opening communicating with the insertion cavity. The transparent support is disposed on the housing, and the transparent elastomer is disposed on the transparent support. A preset pattern is provided on the side of the transparent elastomer facing away from the transparent support. The vision module is inserted into the insertion cavity through the opening. The connection module is detachably connected to the housing or the vision module and covers the opening, and is used to connect the dexterous hand. Thus, compared to visual-tactile sensors in related technologies, the visual-tactile sensor provided in this application embodiment, by setting a detachable connection module, changes the connection method between the visual-tactile sensor and the dexterous hand from the traditional rigid fixation to a modular, quickly replaceable structure. This helps reduce the operational complexity of maintaining and replacing the visual-tactile sensor: when the tactile module or vision module is damaged, it is not necessary to disassemble the entire dexterous hand; only the connection module and the housing or vision module need to be disassembled to complete the component disassembly, significantly saving repair time and labor costs. Meanwhile, this design avoids potential mechanical damage or decreased assembly precision caused by repeated disassembly and reassembly of the dexterous hand's main structure, effectively ensuring the long-term operational stability and sensing accuracy of the dexterous hand. Furthermore, the housing features a insertion cavity and an opening. The vision module is inserted into the insertion cavity through the opening and then sealed by the connecting module. This ensures the stability of the internal optical path of the visual-tactile sensor and effectively prevents external contaminants such as dust and liquids from entering, improving the reliability of the visual-tactile sensor in complex environments. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the structure of the visual-tactile sensor provided in an embodiment of this application.

[0022] Figure 2 for Figure 1 A cross-sectional schematic diagram of the Zhongshi tactile sensor.

[0023] Figure 3 for Figure 1 A schematic diagram of the exploded structure of the Zhongshi tactile sensor.

[0024] Explanation of icon numbers:

[0025] 10. Visual-tactile sensor; 100. Tactile module; 102. Opening; 103. Reflective cavity; 104. Insertion cavity; 110. Transparent support; 121. Wear-resistant layer; 122. Transparent elastomer; 123. Reflector; 130. Housing; 140. Stop structure; 200. Vision module; 210. Camera mounting bracket; 220. Camera module; 300. Connection module; 301. Cable guide hole; 310. Connector; 400. First locking element; 500. Second locking element;

[0026] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0028] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0031] Please see Figure 1 , Figure 2 and Figure 3This application provides a visual-tactile sensor 10, which includes a tactile module 100, a vision module 200, and a connection module 300. The tactile module 100 includes a housing 130, a transparent elastomer 122, and a transparent support 110. The housing 130 has a insertion cavity 104 and an opening 102 communicating with the insertion cavity 104. The transparent support 110 is disposed on the housing 130, and the transparent elastomer 122 is disposed on the transparent support 110, providing support for the transparent elastomer 122. A preset pattern is provided on the side of the transparent elastomer 122 facing away from the transparent support 110. The vision module 200 is inserted into the insertion cavity 104 through the opening 102. The connection module 300 is detachably connected to the housing 130 or the vision module 200 and covers the opening 102, and the connection module 300 is used to connect a dexterous hand. Thus, compared to visual-tactile sensors in related technologies, the visual-tactile sensor 10 provided in this embodiment, by setting a detachable connection module 300, changes the connection method between the visual-tactile sensor 10 and the dexterous hand from the traditional rigid fixation to a modular and quickly replaceable structure. This helps reduce the operational complexity of maintaining and replacing the visual-tactile sensor 10: when the tactile module 100 or the vision module 200 is damaged, it is not necessary to disassemble the entire dexterous hand; only the connection module 300 needs to be disassembled from the housing 130 or the vision module 200 to complete the disassembly of the component, significantly saving maintenance time and labor costs. At the same time, it avoids the mechanical damage or decreased assembly accuracy that may be caused by repeated disassembly and reassembly of the main structure of the dexterous hand, effectively ensuring the long-term operational stability and sensing accuracy of the dexterous hand.

[0032] In addition, the housing 130 is provided with a insertion cavity 104 and an opening 102. After the vision module 200 is inserted into the insertion cavity 104 through the opening 102, it is covered and sealed by the connection module 300. This not only ensures the stability of the internal optical path of the visual-touch sensor 10, but also effectively prevents external contaminants such as dust and liquid from entering, thereby improving the working reliability of the visual-touch sensor 10 in complex environments.

[0033] The vision module 200 is installed inside the insertion cavity 104. The vision module 200 includes a miniature camera, a light source assembly, and an image processing circuit, enabling it to stably acquire high-contrast, high-resolution deformation patterns. The light source assembly provides illumination for the camera, reducing shadow interference and improving image quality. When the transparent elastomer 122 comes into contact with an external object, the preset pattern deforms, and the camera on the vision module 200 captures the deformation pattern of the preset pattern in real time.

[0034] Please see Figure 2 and Figure 3In some embodiments, the connecting module 300 is detachably connected to the housing 130. For example, the connecting module 300 and the tactile module 100 can be connected by screws or snap-fit, covering the opening 102 of the tactile module 100, serving both a sealing and structural support function. The connecting module 300 has a standardized interface on its outer side for quick docking with the fingertips of a dexterous hand. This interface can be a snap-fit, magnetic, threaded connection, or quick-change flange, supporting rapid installation and disassembly by hand or with tool assistance. The connecting module 300 itself is made of rigid material, possessing sufficient mechanical strength to withstand loads during daily operation, while ensuring the stability of the vision module 200 during operation.

[0035] In some implementations, the connection module 300 and the vision module 200 are detachably connected. For example, the connection module 300 and the vision module 200 can be connected by screws or clips, covering the opening 102 of the tactile module 100, serving a dual purpose of sealing and structural support. The outer side of the connection module 300 has a standardized interface for quick docking with the fingertips of a dexterous hand. This interface can be a snap-fit, magnetic, threaded connection, or quick-change flange, supporting rapid installation and disassembly by hand or with tool assistance. The connection module 300 itself is made of rigid material, possessing sufficient mechanical strength to withstand the loads during daily operation, while ensuring the stability of the vision module 200 during operation.

[0036] In some embodiments, the inner wall contour of the insertion cavity 104 matches the shape of the vision module 200 to achieve precise insertion and positioning of the vision module 200. One end of the insertion cavity 104 has an opening 102, which serves as an assembly channel, allowing the vision module 200 to be inserted axially from the outside into the insertion cavity 104 to complete the assembly with the tactile module 100.

[0037] The inner wall surface of the insertion cavity 104 is precision machined, with good dimensional tolerance control and surface finish to ensure smooth and unobstructed insertion of the vision module 200. In other embodiments, the inner wall of the insertion cavity 104 is provided with a guide ramp, a limiting step, or an elastic snap-fit ​​structure to guide the vision module 200 into accurate positioning and to provide axial or circumferential mechanical limiting after full insertion, preventing the vision module 200 from loosening or shifting due to vibration or impact during use.

[0038] In some embodiments, a transparent support 110 is disposed within the housing 130. The transparent support 110 is made of a high-transmittance material, such as optical-grade silicone, polymethyl methacrylate, or polycarbonate, to reduce light loss during internal propagation. A transparent elastomer 122 covers the upper surface of the transparent support 110, and the two are tightly bonded together through bonding, hot pressing, or integral molding processes to ensure the accuracy and consistency of the transmission of the preset deformation pattern under stress.

[0039] In some embodiments, the transparent support 110 is a transparent solid structure. The outer surface of this transparent solid structure includes a side surface adjacent to the insertion cavity 104, a bottom surface contacting the inner surface of the housing 103, and an upper surface connected to the transparent elastomer 122. The bottom surface of the transparent support 110 in contact with the inner surface of the housing 103 is a reflective surface; that is, the reflective surface is approximately located on the side of the transparent support 110 away from the transparent elastomer 122. For ease of explanation, this reflective surface is named the support reflective surface. The support reflective surface is used to reflect the deformed preset pattern to the vision module 200. The support reflective surface undergoes special treatment to possess high reflectivity characteristics and can be formed by methods such as aluminum plating, silver plating, dielectric film, or multilayer optical thin film. The main function of the support reflective surface is to project the deformation of the preset pattern caused by contact force on the surface of the transparent elastomer 122 into the imaging field of view of the vision module 200 through specular reflection.

[0040] When an external object comes into contact with the transparent elastomer 122, the preset pattern undergoes localized stretching, compression, or shear deformation along with the transparent elastomer 122. Due to the good optical uniformity of the transparent support 110, the pattern outline in the deformed area can clearly penetrate the transparent support 110 and be reflected by the support reflective surface of the transparent support 110. The reflected image path is captured by the camera of the vision module 200 after passing through the transparent support 110.

[0041] In some embodiments, the transparent elastomer 122 and the transparent support 110 are stacked together so that the transparent elastomer 122 can be fixed to the housing 130 by the transparent support 110.

[0042] The transparent support 110 and the housing 130 cooperate to form a reflective cavity 103 communicating with the insertion cavity 104, wherein the insertion cavity 104 extends from the opening 102 of the housing 130 to the vicinity of the transparent support 110. The transparent support 110 and the inner wall of the housing 130 enclose a reflective cavity 103 communicating with the insertion cavity 104. A portion of the boundary of the reflective cavity 103 is formed by the surface of the transparent support 110, and another portion of the boundary of the reflective cavity 103 is formed by the inner surface of the housing 130.

[0043] The cavity walls of the reflective cavity 103, which are opposite to and spaced apart from the transparent support 110, are equipped with reflective structures. These reflective structures reflect the deformed preset pattern to the vision module 200. Specifically, the reflective structures can be implemented using optical processing methods such as high-reflectivity metal coatings, multilayer dielectric films, microprism arrays, or white diffuse reflection coatings. The function of the reflective structures is to guide the deformation information of the preset pattern generated on the surface of the transparent elastomer 122 due to force into the imaging field of view of the vision module 200.

[0044] When an external object comes into contact with the transparent elastomer 122, the preset pattern undergoes stretching, compression, or shear deformation due to localized pressure. The optical modulation signals caused by these deformations pass through the transparent support 110 in a transmission manner and enter the reflection cavity 103. Subsequently, the reflection structure on the cavity wall of the reflection cavity 103 captures the signal and reflects it to the vision module 200.

[0045] In some embodiments, the reflective structure is a housing reflective surface. The inner wall of the housing 130 in the corresponding region of the reflective cavity 103 is optically treated to form a housing reflective surface with high reflectivity. This surface can be achieved by vacuum evaporation, sputtering, or electrochemical deposition processes to coat it with aluminum, silver, or a dielectric multilayer film. The housing reflective surface directly serves as the boundary of the reflective cavity 103, eliminating the need for additional optical components and simplifying the overall structure.

[0046] In some embodiments, the reflective structure is a reflector 123, which is fixedly installed inside the reflective cavity 103. The reflector 123 is positioned at a precise distance from the transparent support 110. The reflector 123 may be made of glass, silicon wafer, or polymer substrate, and its surface is coated with a highly reflective film. The reflector 123 is connected to the housing 130 by snap-fit, adhesive, or precision bracket to ensure that the optical axis remains aligned even under vibration or shock conditions.

[0047] In some embodiments, the transparent support 110 is integrally formed with the housing 130, which helps to reduce the assembly process of the transparent support 110 and the housing 130, eliminating the positioning, bonding, fastening or sealing steps required in traditional split designs. This not only simplifies the production process, but also significantly reduces the risk of mechanical loosening due to assembly errors.

[0048] In some embodiments, the preset pattern can be integrated onto the surface of the transparent elastomer 122 in various ways. For example, the preset pattern can be directly printed onto the exposed surface of the transparent elastomer 122 using a micron-precision printing process. The ink used has the characteristics of high contrast, aging resistance, and good adhesion to the elastomer material, ensuring that the pattern does not peel off or become blurred during repeated deformation. Another example is by setting the preset pattern on the transparent elastomer 122 as an independent pattern layer. This pattern layer consists of a flexible transparent film substrate and a graphic structure embedded or coated thereon, and is bonded to the transparent elastomer 122 through hot pressing, co-curing, or optical adhesive bonding processes to form a stable composite structure.

[0049] To protect the preset pattern from friction, scratches, or chemical corrosion during long-term use, the tactile module 100 also includes a wear-resistant layer 121. The wear-resistant layer 121 is layered with the elastomer 122, covering the outermost surface of the transparent elastomer 122, and directly bears external contact loads. It has high wear resistance and durability, thus improving the service life of the tactile module 100. When the wear-resistant layer 121 is touched or pressed, it compresses the transparent elastomer 122, causing deformation of the preset pattern.

[0050] In some embodiments, a stop structure 140 is provided at the connection between the reflective cavity 103 and the insertion cavity 104. That is, a stop structure 140 is provided in the connection area between the transparent support 110 and the housing 130. The stop structure 140 can be an annular boss or a stepped surface, and its position is located in the transition area between the reflective cavity 103 and the insertion cavity 104. The main function of the stop structure 140 is to provide axial positioning and mechanical limitation for the vision module 200, ensuring that the vision module 200 maintains a precise installation position inside the sensor.

[0051] The vision module 200 is installed in the insertion cavity 104 and abuts against the stop structure 140 and the connecting module 300. After the connecting module 300 is inserted into the opening 102 of the insertion cavity 104, a clamping force is applied inward through a detachable connection, so that the vision module 200 is firmly clamped in the axial gap formed by the stop structure 140 and the connecting module 300.

[0052] The height of the stop structure 140 is precisely calculated to match the overall thickness and optical focal length requirements of the vision module 200. When the vision module 200 is pressed into place, the optical axis of the lens of the vision module 200 intersects with the reflector 123, thereby ensuring that the acquired deformed image is clear and distortion-free.

[0053] Please see Figure 3In some embodiments, the visual-tactile sensor 10 further includes a first locking member 400, which is threadedly connected to the housing 130 and the vision module 200. The first locking member 400 mechanically fixes itself to both the housing 130 and the vision module 200 via a threaded connection. The first locking member 400 is typically a fastener or screw with external threads. Both the housing 130 and the vision module 200 may have internal threads, allowing the first locking member 400 to lock the housing 130 and the vision module 200 in place when tightened.

[0054] During assembly, the vision module 200 is first placed into the insertion cavity 104, and then the first locking member 400 is sequentially engaged with the corresponding threaded sections on the housing 130 and the vision module 200. As the first locking member 400 is gradually tightened, it helps to reduce the loosening or displacement of the housing 130 and the vision module 200 due to vibration, impact or repeated pressing during use.

[0055] The material of the first locking element 400 can be stainless steel, aluminum alloy or high-strength engineering plastic, depending on the different requirements of the usage environment for corrosion resistance, weight or insulation performance.

[0056] In some embodiments, there are multiple first locking elements 400, which are arranged sequentially at intervals along the circumference of the housing 130. Each first locking element 400 is simultaneously fixed to the housing 130 and the vision module 200 by a threaded connection, forming a circumferential fastening structure. This layout effectively distributes the mechanical load borne by the connection area, avoids local stress concentration, and significantly improves the stability and reliability of the overall connection.

[0057] Multiple first locking elements 400 are distributed at equal angles in the circumferential direction, for example, three, four, or six, the specific number depending on the sensor size, internal space, and force requirements. Each first locking element 400 is independently screwed into a corresponding threaded hole, which is respectively opened on the outer circumferential surface of the housing 130 and the vision module 200. When all the first locking elements 400 are tightened simultaneously, the vision module 200 is locked in a predetermined position inside the housing 130, ensuring that the vision module 200 and the tactile component 120 maintain a precise relative posture and preventing tilting or displacement caused by single-point force.

[0058] In some embodiments, the visual-tactile sensor 10 further includes a second locking member 500, which is threadedly connected to the vision module 200 and the connection module 300. The second locking member 500 mechanically fixes itself to both the vision module 200 and the connection module 300 via a threaded connection. The second locking member 500 is typically a screw or bolt. Threaded mating portions are respectively located on the side wall of the housing of the vision module 200 and in the corresponding mounting holes of the connection module 300, ensuring that the vision module 200 and the connection module 300 are securely integrated after the second locking member 500 is tightened.

[0059] The material of the second locking member 500 can be stainless steel, titanium alloy, or high-strength engineering plastic, selected according to the specific requirements of corrosion resistance, weight, and insulation performance in the application scenario. In addition, the head of the second locking member 500 can be designed as a countersunk head, a round head, or an internal hexagonal head, so that it is flush with or embedded in the outer surface of the connecting module 300, avoiding the protruding structure from affecting the overall shape of the visual tactile sensor 10 or interfering with other components.

[0060] In some embodiments, the connecting module 300 is provided with a wire passage hole 301, which extends through both opposite sides of the connecting module 300 and is used for the wires of the vision module 200 to pass through. The wire passage hole 301 extends through the two opposite sides of the connecting module 300 in the thickness direction. The main function of the wire passage hole 301 is to provide a passage for the wires of the vision module 200, enabling the camera, light source, and image processing circuitry inside the vision module 200 to achieve electrical connection with an external control system or power supply. The position of the wire passage hole 301 is carefully arranged, typically located in the edge area or non-stressed area of ​​the connecting module 300, to avoid weakening the structural strength or interfering with the sealing performance of the mechanical interface.

[0061] The inner wall surface of the wire passage hole 301 is smooth, without burrs or sharp edges, to prevent damage to the wire due to friction during assembly or use. In some embodiments, a flexible protective sleeve or insulating liner, made of silicone, polyurethane, or heat-shrinkable material, is also embedded inside the wire passage hole 301 to further protect the wire insulation layer and buffer stress caused by vibration or bending. The protective sleeve and the wire passage hole 301 can be fixed by an interference fit or a snap-fit ​​structure to ensure that it will not fall off or shift during long-term use.

[0062] The connection module 300 transmits signals and power between the internal vision module 200 and external devices through the wire passage hole 301, while maintaining the compactness and sealing of the overall structure. After the sensor is assembled, the wires are led out from the vision module 200, pass through the receiving space 101, extend through the wire passage hole 301 to the outside of the connection module 300, and connect to the main control circuit of the robot's dexterous hand. This design avoids the reliability risks caused by exposed and tangled wires or temporary openings, and improves the neatness of the wiring and the system integration.

[0063] In addition, a sealing component, such as a waterproof connector, potting compound, or elastic sealing plug, can be provided at the outlet end of the wire hole 301 to effectively prevent external liquids, dust, or moisture from entering the sensor along the wire path while ensuring that the wire passes through smoothly.

[0064] In some embodiments, the connection module 300 has multiple connectors 310 on the side opposite to the vision module 200. These connectors 310 are used to engage with the dexterous hand. The connectors 310 are used to mechanically connect to the mounting interface of the dexterous hand. Each connector 310 has a standardized geometry and dimensions, such as threaded holes, snap-fit ​​bosses, magnetic units, or quick-change flange interfaces, to match the end effector structure of different brands or models of dexterous hands.

[0065] Adjacent connectors 310 are spaced apart and arranged at intervals along the circumference of the connection module 300 or in specific functional areas, forming a multi-point distributed connection layout. The spaced arrangement of multiple connectors 310 not only enhances the mechanical stability of the connection but also improves the torsional and shear resistance of the visual-tactile sensor 10 during dynamic operation. Through multi-point force application, the connection module 300 can evenly transmit external loads, avoiding loosening or structural fatigue caused by localized stress concentration. Simultaneously, this layout provides users with flexible installation options: depending on the interface type of the target dexterous hand, some or all of the connectors 310 can be selected for adaptation without requiring structural modifications to the visual-tactile sensor 10.

[0066] In some embodiments, the vision module 200 is detachably connected to the housing 130 and the connection module 300. The vision module 200 is detachably connected between the housing 130 and the connection module 300 in a manner such as screw connection or snap-fit ​​connection, forming the core functional unit inside the sensor. This connection method is achieved through a mechanical limiting and locking structure, eliminating the need for adhesives or permanent fixation, and allowing for independent replacement or maintenance of the vision module 200 without damaging other components.

[0067] Specifically, the vision module 200 is housed in the insertion cavity 104 of the housing 130, and is provided with an axial positioning reference by the stop structure 140. After the connecting module 300 is inserted from the opening 102 end of the insertion cavity 104, the vision module 200 is pressed by means of threads, snaps, or locking elements, so that it is clamped between the stop structure 140 of the tactile module 100 and the inner end face of the connecting module 300, so as to form a stable bidirectional constraint, which prevents the vision module 200 from moving axially and restricts its circumferential rotation, ensuring that the imaging optical path remains aligned for a long time.

[0068] When the vision module 200 needs to be replaced or repaired, the operator only needs to loosen the connection structure between the connecting module 300 and the housing 130 (for example, by unscrewing the first locking member 400 or the second locking member 500) to remove the connecting module 300 as a whole, and then pull the vision module 200 out of the insertion cavity 104. The entire process does not require disassembling the dexterous hand body, nor does it require damaging the sealing structure of the housing 130, significantly reducing maintenance complexity and time costs.

[0069] In addition, the outer surface of the vision module 200 is provided with guide bosses or positioning pins, and the housing 130 and the corresponding position of the connecting module 300 are provided with matching guide grooves or positioning holes. These auxiliary structures guide the vision module 200 to be accurately positioned during assembly, improve the consistency of repeated installations, and avoid optical misalignment or poor contact caused by human error.

[0070] In some embodiments, the vision module 200 includes a camera fixture 210 and a camera module 220. The camera fixture 210 can cooperate with the first locking member 400 and the second locking member 500. The camera module 220, as the core imaging unit, integrates an image sensor, an optical lens, and an auxiliary light source, etc., for real-time capture of the deformation pattern generated by the tactile component 120 under pressure. The resolution, frame rate, and photosensitivity of the camera module 220 are selected according to the tactile perception accuracy requirements, and can support multiple imaging modes such as visible light, near-infrared, or structured light to adapt to different lighting conditions and surface characteristics.

[0071] The camera mount 210 is used to support and precisely fix the camera module 220 in position within the insertion cavity 104. The camera mount 210 is typically made of a rigid material, such as aluminum alloy, stainless steel, or high-strength engineering plastic. The camera mount 210 has an internal mounting groove structure that matches the shape of the camera module 220. The outer contour of the camera mount 210 forms a stable mating relationship with the inner wall of the housing 130 and the connecting module 300, achieving reliable positioning through the coordinated action of multiple locking elements.

[0072] In addition, the camera mounting bracket 210 is provided with a cable pass hole 301, which is used to guide the power cable and signal cable of the camera module 220 through the connection module 300 and to connect to the control system of the dexterous hand.

[0073] This application provides a dexterous hand, which includes a visual-tactile sensor 10 as described in any of the above embodiments. The visual-tactile sensor 10 is disposed on the dexterous hand. Specifically, the visual-tactile sensor 10 is installed on the fingertips of the dexterous hand to perceive physical information such as force, deformation, texture, and contour generated during contact with external objects in real time, providing high-resolution tactile feedback for the dexterous hand to perform fine tasks such as grasping, manipulating, and recognizing.

[0074] A dexterous hand typically consists of multiple joints, drive units, transmission mechanisms, and a structural skeleton, possessing multi-degree-of-freedom motion capabilities and able to simulate the complex movements of a human hand. The visual-tactile sensor 10 achieves quick and stable mechanical docking with the dexterous hand via its connection module 300. The connection module 300 has a standardized interface that allows it to connect to the dexterous hand's mounting base via snap-fit, magnetic attraction, threads, or quick-change flanges, ensuring the sensor remains stable during dynamic operation and supporting assembly and disassembly by hand or with simple tools.

[0075] Because the visual-tactile sensor 10 adopts a modular design, when it needs to be replaced due to wear, aging, or functional upgrades, there is no need to completely disassemble the dexterous hand. Operators only need to disconnect the connecting module 300 from the dexterous hand to remove the entire visual-tactile sensor 10 and replace it with a new or a different specification of visual-tactile sensor 10. This process significantly shortens maintenance time, lowers the technical threshold, and effectively avoids potential damage to the internal precision mechanisms of the dexterous hand caused by repeated disassembly and reassembly.

[0076] The vision module 200 inside the visual-tactile sensor 10 is led out through the wire through the wire hole 301 on the connection module 300 and electrically connected to the control system inside the dexterous hand. The control system receives the deformation image data collected by the vision module 200, combines it with the algorithm to deduce the contact state information, and adjusts the grasping force, posture or trajectory accordingly to achieve closed-loop intelligent interactive control.

[0077] In some embodiments, the number of visual-tactile sensors 10 can be multiple, and they are preferably located at the fingertips of the dexterous hand. The fingertips are the core area where the dexterous hand contacts and interacts with external objects, and they play a major role in force transmission and information perception during fine operations such as grasping, pinching, and exploring. By arranging the visual-tactile sensors 10 at the fingertips of each finger, the deformation characteristics generated at the moment of contact can be captured directly and efficiently, thereby accurately retrieving key tactile information such as the magnitude and direction of the contact force, surface texture, and local geometry.

[0078] Each fingertip can integrate an independent visual-tactile sensor 10, which is detachably connected to the fingertip structure via its connection module 300. This layout ensures that during multi-finger collaborative operation, the system can simultaneously acquire high-resolution tactile data from multiple contact points, providing the dexterous hand with human-like distributed perception capabilities. Furthermore, due to the modular design of the visual-tactile sensor 10, when it malfunctions or requires an upgrade, replacement can be completed simply by disassembling the corresponding fingertip's connection module 300, without disassembling the entire dexterous hand, greatly improving maintenance efficiency and operational flexibility.

[0079] In the accompanying drawings of this embodiment, the same or similar reference numerals correspond to the same or similar components. In the description of this application, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting this application. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0080] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A visual-tactile sensor, characterized in that, include: A tactile module, comprising a housing, a transparent elastomer, and a transparent support, wherein the housing has a insertion cavity and an opening communicating with the insertion cavity, the transparent support is disposed on the housing, the transparent elastomer is disposed on the transparent support, and a preset pattern is provided on the side of the transparent elastomer facing away from the transparent support; The vision module is inserted into the insertion cavity through the opening; as well as A connection module is detachably connected to the housing or the vision module and covers the opening, and the connection module is used to connect a dexterous hand.

2. The visual-tactile sensor according to claim 1, characterized in that, The transparent support is disposed inside the housing, and a reflective surface is provided on the side of the transparent support away from the transparent elastomer. The reflective surface is used to reflect the deformed preset pattern to the vision module.

3. The visual-tactile sensor according to claim 1, characterized in that, The transparent elastomer and the transparent support are stacked together. The transparent support and the housing cooperate to form a reflective cavity that communicates with the insertion cavity. The cavity wall of the reflective cavity is provided with a reflective structure that is opposite to and spaced apart from the transparent support. The reflective structure is used to reflect the deformed preset pattern to the visual module.

4. The visual-tactile sensor according to claim 3, characterized in that, The reflective structure is either a shell reflective surface or a reflective mirror.

5. The visual-tactile sensor according to claim 4, characterized in that, A stop structure is provided at the connection between the reflective cavity and the insertion cavity, and the vision module abuts between the stop structure and the connection module.

6. The visual-tactile sensor according to any one of claims 1 to 5, characterized in that, The transparent support is integrally formed with the shell.

7. The visual-tactile sensor according to any one of claims 1 to 5, characterized in that, The visual-tactile sensor also includes a first locking member, which is threadedly connected to the housing and the vision module; And / or, the visual-tactile sensor further includes a second locking element, which is threadedly connected to the vision module and the connection module.

8. The visual-tactile sensor according to any one of claims 1 to 5, characterized in that, The connection module is provided with a wire through hole, which extends through both opposite sides of the connection module and is used for the wires of the vision module to pass through.

9. The visual-tactile sensor according to any one of claims 1 to 5, characterized in that, The connection module has multiple connectors on the side opposite to the vision module. The connectors are used to cooperate with the dexterous hand, and adjacent connectors are spaced apart.

10. A dexterous hand, characterized in that, include: The visual-tactile sensor as described in any one of claims 1 to 9, wherein the visual-tactile sensor is disposed on the dexterous hand.

Images