Bionic hair tactile sensor and application

Abstract

The invention relates to the technical field of sensors, in particular to a bionic hair tactile sensor and application, and the sensor comprises a substrate and a bionic cilia array arranged on the substrate; the bionic cilia array comprises a plurality of flexible bionic cilia arranged in an array, and a marking layer is arranged at the bottom of the flexible bionic cilia; an optical system is integrated in the substrate, the optical system comprises an imaging module and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; and the imaging module is used for capturing deformation data of the marking layer and sending the deformation data to the deep learning algorithm module to realize decoupling and resolution of three-dimensional force. According to the invention, high-resolution multi-dimensional force detection and direction detection are realized through mark point change caused by cilium deformation, and the array design is expanded to a sliding sensing function and a multi-modal object identification function.

Description

Technical Field

[0001] This application relates to the technical field of sensors, and particularly to a bionic hair tactile sensor and its application. Background Art

[0002] As a key component for robots to perceive external mechanical information, the technology of tactile sensors has evolved from single mechanical parameter detection to multi-modal perception. In recent years, with the breakthroughs in flexible electronics, bionics, and optical technologies, tactile sensors face new technical requirements in terms of sensitivity, multi-dimensional perception ability, and environmental adaptability. Firstly, there is a refined demand for application scenarios; the demand for intelligent robots (such as surgical robots and collaborative robotic arms) to perceive micro-forces (<1 mN) and complex contact states (such as sliding and rolling) has increased significantly. For example, in minimally invasive surgery, it is necessary to detect the blood vessel contact force of 0.5 - 5 mN, and traditional sensors are difficult to balance sensitivity and range. Secondly, there is a contradiction in the real-time performance of dynamic perception; high spatio-temporal resolution (>100 Hz sampling rate) is required for sliding detection, while high-density array sensors (such as capacitive arrays) are difficult to achieve real-time response due to signal crosstalk and data processing delays. Thirdly, there is a lack of robustness in complex environments; there are complex environments such as strong light, oil stains, and electromagnetic interference in industrial and medical scenarios. Traditional optical sensors (such as those based on RGB cameras) are prone to failure, and magnetic sensors are significantly affected by the metal environment.

[0003] Existing tactile sensors based on the deformation of optical markers print high-contrast markers on the surface of an elastomer, and use a camera to capture the displacement change of the markers under the action of external forces, and then infer the magnitude and direction of the contact force. However, the existing optical marker deformation technology has problems such as insufficient sensitivity for micro-force detection, limited direction recognition dimension, and poor real-time performance of dynamic perception; the lower limit of micro-force detection of the existing optical marker deformation technology (such as GelSight) is about 10 mN, which cannot meet the requirements of high-precision scenarios such as surgical robots (requiring 0.5 - 5 mN level detection). Although the fiber optic cilia sensor can reach 1 mN, it is limited by the detection noise of light intensity and is difficult to further improve the resolution; moreover, existing technologies mostly rely on two-dimensional deformation analysis (such as the planar displacement of markers or the change of uniaxial magnetic field), and cannot decouple three-dimensional force components. For example, fiber optic sensors can only detect normal forces, while the vision-cilia fusion system (OmniTact) supports three-dimensional decoupling, but the direction error exceeds 10°, which cannot meet the requirements of precision operations; for dynamic perception, the bandwidth of fiber optic sensors is lower than 50 Hz, and the vision solution is affected by image processing delays (requiring GPU acceleration), resulting in a sliding detection response time exceeding 50 ms, making it difficult to achieve real-time closed-loop control. Summary of the Invention

[0004] The embodiments of the present application provide a bionic hair tactile sensor and its application, aiming to solve the deficiencies of existing tactile sensors in aspects such as micro-force detection sensitivity, multi-dimensional force direction recognition, large-area dynamic sliding perception, object feature recognition, and complex environment adaptability.

[0005] To solve the above technical problems, on the first aspect, the embodiments of the present application provide a bionic hair tactile sensor, including: a substrate and a bionic cilia array disposed on the substrate; the bionic cilia array includes a plurality of flexible bionic cilia arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia; an optical system is integrated in the substrate, and the optical system includes an imaging module and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; the imaging module is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to achieve decoupling and resolution of three-dimensional forces.

[0006] In some exemplary embodiments, the material of the flexible bionic cilia is PA6; the diameter of each flexible bionic cilium is 400 microns and the height is 8 mm.

[0007] In some exemplary embodiments, the density of the flexible bionic cilia array is 10 - 25 pieces / cm².

[0008] In some exemplary embodiments, the bionic cilia array is formed by a nylon particle hot pressing process, and the surface is activated by oxygen plasma to enhance the adhesion of the marking layer; the hot pressing process conditions are: molding at 280°C, the holding pressure time is 90 seconds, and the cooling gradient is 3°C / second.

[0009] In some exemplary embodiments, the marking layer is impregnated with a marking material through a microgroove coating device and forms a stable coating after being cured by a nano-UV LED array.

[0010] In some exemplary embodiments, the marking material is one of liquid silicone, polymer film, or metal film.

[0011] In some exemplary embodiments, the Shore hardness of the marking layer is 20A and the thickness is 200 microns.

[0012] In some exemplary embodiments, a marking layer is provided at the bottom of each flexible bionic cilium, and the thickness of the marking layer is 2 mm.

[0013] In some exemplary embodiments, the imaging module is one of an embedded camera, a flexible endoscope, a CMOS sensor, or a micro camera.

[0014] On the second aspect, the embodiments of the present application also provide an application of the bionic hair tactile sensor as described in the above embodiments in the field of intelligent robots' perception of micro-force operations and complex contact states.

[0015] The technical solution provided by the embodiments of the present application has at least the following advantages: The embodiments of the present application provide a bionic hair tactile sensor and an application. The sensor includes: a substrate and a bionic cilia array arranged on the substrate; the bionic cilia array includes a plurality of flexible bionic cilia arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia; an optical system is integrated in the substrate, and the optical system includes an imaging module and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; the imaging module is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to realize the decoupling and resolution of three-dimensional forces. The present application realizes high-resolution multi-dimensional force detection and direction detection through the change of the marking points caused by the deformation of the cilia, and expands to the functions of sliding perception and multi-modal object recognition by using the array design. Description of the Drawings

[0016] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings. These exemplary illustrations do not constitute limitations on the embodiments. Unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0017] Figure 1 It is a schematic structural diagram of a bionic hair tactile sensor provided by an embodiment of the present application.

[0018] Figure 2 It is a surface topography diagram of a bionic hair tactile sensor provided by an embodiment of the present application.

[0019] Figure 3 It is a follicle topography diagram of the back surface of the surface layer of a bionic hair tactile sensor provided by an embodiment of the present application.

[0020] Figure 4 It is a surface topography diagram of a bionic hair tactile sensor provided by an embodiment of the present application.

[0021] Figure 5 It is a schematic diagram of the elastic modulus of a bionic hair provided by an embodiment of the present application. Detailed Embodiments

[0022] As can be seen from the background technology, the existing tactile sensors have obvious deficiencies in terms of micro-force detection sensitivity, multi-dimensional force direction recognition, large-area dynamic sliding perception, object feature recognition, and complex environment adaptability.

[0023] Existing tactile sensors based on the deformation of optical marker points use the technique of printing high-contrast marker points on the surface of an elastomer and capturing the displacement changes of the marker points under external forces with a camera, and then estimating the magnitude and direction of the contact force. For example, the GelSight sensor developed by MIT in 2015 uses an elastomer coating combined with reflective particles to reconstruct the spatial distribution of the contact force through image processing algorithms. Its advantages are that it can achieve a spatial resolution of sub-millimeter level and can detect the texture features of the object surface. However, this technology has significant drawbacks: the sensitivity of micro-force detection is relatively low (the lower limit is about 10 mN), which cannot meet the requirement of sub-milliNewton-level force perception in scenarios such as surgical robots; the direction recognition ability is limited to two-dimensional tangential force estimation and cannot decouple three-dimensional force components; in addition, the sensor relies on visible light illumination, and its performance drops significantly in strong light or dark environments, and its environmental adaptability is poor.

[0024] For fiber-optic bionic cilia tactile sensors, such sensors combine flexible cilia structures with fiber-optic sensing technology and achieve mechanical detection through the fiber-optic transmission loss or light intensity change caused by the bending of the cilia. A typical representative is the magneto-optical composite cilia sensor proposed by Alfadhel et al. in 2016, which uses magnetic field changes to modulate the fiber-optic output signal. The advantages of this scheme are strong anti-electromagnetic interference ability and the ability to detect micro-forces of about 1 mN. However, its limitations are also obvious: the signal dimension is single, and it can only detect the normal force through the light intensity change method and cannot distinguish the tangential force direction; the expansion of the cilia array relies on complex optical multiplexing circuits, resulting in high system costs; in addition, the light intensity detection bandwidth is usually lower than 50 Hz, making it difficult to meet the real-time requirements of dynamic scenarios such as sliding detection.

[0025] In the field of visual-cilia fusion tactile system technology, this technology realizes multi-dimensional force detection by setting reflective marker points at the top of the cilia and combining a high-speed camera to track the bending trajectory of the cilia. The OmniTact system at the University of California, San Diego in 2020 is a typical representative, which uses a hemispherical cilia array in cooperation with a fish-eye lens to support three-dimensional force decoupling and large-area tactile mapping. The advantages of this scheme include three-dimensional force detection ability and array scalability, but there are still key technical bottlenecks: limited by the pixel density of the camera, the sensitivity of micro-force detection only reaches the level of 0.5 mN; the system is sensitive to environmental light conditions and requires a strictly light-shielded environment to work stably; in addition, high-precision algorithms rely on GPU acceleration processing, making it difficult to achieve low-power deployment in embedded devices, which limits the practical application in industrial scenarios.

[0026] Aiming at the technical bottlenecks of insufficient sensitivity of micro-force detection, limited dimension of direction recognition, and poor real-time performance of dynamic perception in the existing technology, an embodiment of the present application provides a bionic hair tactile sensor and its application, including: a substrate and a bionic cilia array disposed on the substrate; the bionic cilia array includes multiple flexible bionic cilia arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia; an optical system is integrated in the substrate, and the optical system includes an imaging module and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; the imaging module is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to achieve decoupling and resolution of three-dimensional forces. The present application realizes high-resolution multi-dimensional force detection and direction detection through the change of the marked points caused by the deformation of the cilia, and expands to the functions of slip perception and multi-modal object recognition by using the array design.

[0027] An embodiment of the present application provides a bionic hair tactile sensor. Through the collaborative optimization of the sub-pixel-level optical flow tracking algorithm and a high-frame-rate CMOS sensor (≥1000 fps), a resolution of 0.1 mN level is achieved, meeting the micro-force perception requirements in scenarios such as surgical robots to improve the sensitivity of micro-force detection; at the same time, the present application designs a multi-view imaging system (dual-camera layout) and a defocus depth detection algorithm. Combining the three-dimensional displacement tracking of the marked points at the top of the cilia, the accuracy error of three-dimensional force direction recognition is achieved to be <5°, breaking through the existing two-dimensional direction detection limitation and realizing accurate decoupling of three-dimensional force directions. In addition, the present application adopts a lightweight dynamic time warping (DTW) algorithm and a hardware parallel acceleration architecture to compress the slip detection response time to <10 ms, supporting real-time tactile feedback control of robots and enhancing the real-time performance of dynamic response.

[0028] The following will elaborate on the embodiments of the present application in conjunction with the accompanying drawings. However, those of ordinary skill in the art can understand that in the embodiments of the present application, many technical details are proposed for readers to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can still be realized.

[0029] As Figure 1 shown, an embodiment of the present application provides a bionic hair tactile sensor, including: a substrate 1 and a bionic cilia array disposed on the substrate 1; the bionic cilia array includes multiple flexible bionic cilia 2 arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia 2; an optical system is integrated in the substrate 1, and the optical system includes an imaging module 3 and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia 2 to the imaging module 3; the imaging module 3 is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to achieve decoupling and resolution of three-dimensional forces.

[0030] The sensor provided by this application adopts a design scheme combining a bionic cilia array and an optical system. Based on the perception mechanism of biological skin hair, nylon 6 (diameter 400 microns, height 8 mm) is selected as the bionic cilia material to construct a high-density cilia array (arranged in a 5×5 pattern). The spacing between bionic cilia is 3 mm, effectively avoiding the entanglement problem. The mechanical consistency is ensured through a hot pressing process, and a marking material (black liquid silicone) is coated on the 2-mm area at the bottom of the bionic cilia as a marking layer (optical marking point). The Shore hardness of this marking layer is 20A, and the thickness is 200 microns. A highly contrastive (reflectivity difference > 85%) traceable feature is achieved through an ultraviolet curing process. The bionic cilia array is arranged in a 5×5 diagonal layout (tilt angle 15°), that is, the cilia are arranged at a 15° tilt angle, effectively reducing the physical interference between adjacent cilia and enhancing the independent deformation detection ability in a dense environment.

[0031] Figure 2 shows the surface morphology of the bionic hair tactile sensor. From Figure 2 it can be seen that: the bionic hair is inserted into the surface of the electronic skin to form the surface morphology of the bionic hair tactile sensor; Figure 3 shows the follicle morphology on the back surface of the surface layer of the bionic hair tactile sensor, that is Figure 3 shows the follicle structure at the bottom of the hair. When the surface of the hair is stimulated by the outside world, it will cause the displacement of the internal follicle structure. The bottom camera captures this change and inputs it to the deep learning network to analyze the outside stimulus; Figure 4 shows a schematic diagram of deriving tactile information by identifying the displacement of the follicle based on the principle of visual touch. The follicle structure is characterized by computer vision, making its features more obvious.

[0032] Breaking through the layout limitations of traditional CMOS sensors, the optical system of this application can adopt a micro-endoscope optical system. The sensing layer uses a medical-grade flexible endoscope (diameter 2.9 mm, field of view angle 120°) as the core imaging unit, thus replacing the traditional CMOS sensor and realizing the separation of illumination and imaging in cooperation with a dual-channel fiber bundle; the dual-channel fiber bundle (diameter 1.2 mm) realizes the separation of the illumination and imaging paths. Among them, for the illumination channel, a high-brightness LED (color temperature 5500K, illuminance dynamic range 50 - 2000 lux) is integrated; for the imaging channel, a micro prism group (reflection angle 45° ± 2°) maps the three-dimensional deformation of the marking layer at the bottom of the bionic cilia to the imaging module, and the imaging module can adopt an image sensor with a resolution of 1280×720. The support structure adopts a 3D-printed titanium alloy hollow bracket (porosity 70%), providing mechanical support and allowing the sensor to still maintain the optical alignment accuracy (deviation < 5 μm) when the bending radius ≥ 8 mm, taking into account both lightweight and mechanical strength.

[0033] It should be noted that for the optical sensor replacing the camera: In this application, an embedded camera is used to capture the deformation data of the surface marking layer. As an alternative, other types of optical sensors (e.g., CMOS sensors or micro imaging modules) can be used to achieve the same image capture function. These sensors can be replaced in terms of different sizes, performances, and costs without affecting the overall function and accuracy of the system.

[0034] The deformation detection system is based on the monocular vision three-dimensional reconstruction technology. Through the improved EPnP algorithm combined with nonlinear optimization, the two-dimensional images captured by the endoscope are converted into three-dimensional displacement data of the cilia. Feature tracking uses ORB descriptors (500 feature points are extracted per frame), with a matching accuracy of ±1.5 pixels, achieving an axial displacement resolution of ±8 microns (corresponding to a 0.2 mN force detection). The dynamic anti-interference module integrates a six-axis IMU sensor (BMI160, sampling rate 1000 Hz), and compensates for the image blur caused by mechanical vibration in real time through gyroscope and acceleration data. The light intensity adaptive system uses a PID control algorithm to adjust the LED brightness (50 - 2000 lux), and still maintains the illuminance fluctuation less than 5% in a strong interference environment such as an operating room shadowless lamp. Temperature drift suppression is compensated by a closed-loop of a PT1000 high-precision sensor, ensuring that the temperature drift rate is lower than 0.01% / °C in the range of -20°C to 60°C.

[0035] In some embodiments, the material of the flexible bionic cilia 2 is PA6; the diameter of each flexible bionic cilia 2 is 400 microns, and the height is 8 mm. PA6 is polyamide 6, that is, nylon 6.

[0036] In some embodiments, the density of the flexible bionic cilia array is 10 - 25 pieces / cm².

[0037] In some embodiments, the bionic cilia array is formed by a nylon particle hot pressing process, and the surface is activated by oxygen plasma to enhance the adhesion of the marking layer; the hot pressing process conditions are: molding at 280°C, holding pressure time of 90 seconds, and cooling gradient of 3°C / second.

[0038] In some embodiments, the marking layer is impregnated with a marking material by a microgroove coating device and forms a stable coating after being cured by a nano-UV LED array.

[0039] Specifically, the preparation process flow of the sensor provided in this application is as follows: The flexible bionic cilia are formed by hot pressing nylon particles (molded at 280°C, holding pressure for 90 seconds, cooling gradient of 3°C / second), and the surface is activated by oxygen plasma (processed at a power of 80 watts for 90 seconds) to enhance the adhesion of the silicone marker layer (marker points). The marker layer is made by impregnating black liquid silicone at a speed of 0.5 mm / second through a microgroove coating device and cured for 30 seconds by a 395-nm ultraviolet LED array (intensity of 80 mW / cm²) to form a stable coating. In the optical system integration stage, laser-assisted centering technology is used, and the illumination and imaging are separated by cooperating with a dual-channel fiber bundle to ensure that the coaxiality error between the endoscope and the fiber bundle is ≤5 microns, and the precision butt joint of the V-groove makes the fiber insertion loss less than 0.5 dB. The final assembly achieves IP68-level waterproof performance through epoxy resin encapsulation and can withstand continuous immersion in 1-meter-deep water for 1 hour.

[0040] It should be noted that different types of marker layers can be used. In this application, black silicone markers are used, and ultraviolet lithography technology is used to make patterns. Alternative solutions can use different types of marker materials (such as transparent films, polymers, or metal films) and different pattern-making technologies (such as laser etching, hot pressing, etc.). These changes can adapt to different application requirements, such as use in extreme environments or more miniaturized devices.

[0041] In some embodiments, the marker material is one of liquid silicone, polymer film, or metal film.

[0042] In some embodiments, the Shore hardness of the marker layer is 20A, and the thickness is 200 microns.

[0043] In some embodiments, a marker layer is provided at the bottom of each flexible bionic cilium, and the thickness of the marker layer is 2 mm.

[0044] In some embodiments, the imaging module is one of an embedded camera, a flexible endoscope, a CMOS sensor, or a micro camera.

[0045] In addition, the embodiments of this application also provide an application of the bionic hair tactile sensor as described in the above embodiments in the field of intelligent robots' perception of micro-force operations and complex contact states.

[0046] The bionic hair tactile sensor provided in this application has a bionic cilium array with a rigid substrate (E≥1 GPa), a flexible top (E≤100 kPa), and a fluorescent / highly reflective marker layer, and integrates an optical system of a flexible endoscope and a polarization filter. The diameter is ≤3 mm, the field of view coverage rate is ≥90%, and the three-dimensional force direction recognition accuracy error is <5°, breaking through the limitations of existing two-dimensional direction detection.

[0047] Compared with the prior art, the bionic hair tactile sensor provided by this application has the following advantages: (1) Higher tactile perception accuracy and complexity.

[0048] The prior art usually relies on traditional resistive or capacitive sensors to sense tactile signals. These sensors are insufficient in capturing complex tactile actions (such as pinching and sliding), especially with significant limitations in the perception of force distribution and direction. In contrast, this application combines the visual tactile perception principle with deep learning algorithms, which can accurately capture the deformation of the surface marking layer and analyze the three-dimensional force distribution, enabling more detailed identification of complex tactile interactions. For example, actions such as sliding and pinching can be recognized more precisely. This vision-based tactile perception technology can greatly improve the system's recognition accuracy for complex actions.

[0049] (2) Superior three-dimensional force decoupling ability compared to the prior art.

[0050] Traditional force sensors (such as resistive and piezoelectric sensors) usually cannot effectively decouple forces in different directions and of different types, resulting in limited tactile perception information being provided, especially for the precise measurement of multi-dimensional forces. This application combines the image data captured by a camera with deep learning algorithms to effectively achieve the decoupling and resolution of three-dimensional forces, greatly enhancing the accuracy and fineness of force perception. Compared with the prior art, this application can not only identify simple pressures but also accurately analyze complex force distributions and interaction methods such as sliding and pinching, solving the limitations of traditional force sensors.

[0051] (3) Higher environmental adaptability and robustness.

[0052] The prior art is often limited by the illumination or noise of the sensor itself. Especially in an environment with large changes in illumination conditions, the recognition accuracy may be greatly reduced. This application adopts dynamic brightness adjustment and image enhancement technologies, which can effectively resist the interference of environmental illumination changes and improve the stability and robustness of the system in complex environments. This is much better than resistive or capacitive sensors in the prior art, which are usually sensitive to illumination changes and cannot be adjusted or optimized by software means.

[0053] (4) Multi-task learning ability and data utilization efficiency.

[0054] Force perception and tactile recognition in the prior art often require separately designed independent systems or models. However, this application adopts deep learning architectures (such as ResNeXt and R(2+1)D convolutional networks), which can achieve multi-task learning of tactile action recognition and three-dimensional force perception, effectively improving the data utilization efficiency and the adaptability of the model. Through a unified deep learning framework, this application can complete multiple tasks simultaneously in a single system, avoiding the waste of computing resources and performance bottlenecks caused by task separation in traditional technologies.

[0055] (5) Improvement in cost-effectiveness.

[0056] Traditional flexible electronic skin sensors require multiple separate sensor components (such as pressure sensors, temperature sensors, etc.). These components not only increase the complexity of the system but also bring relatively high costs. In this application, by combining visual perception and deep learning, the tasks that originally required multiple sensors are concentrated in an integrated system, reducing the hardware cost while simplifying the system design and improving the integration and reliability of the system.

[0057] (6) Better user interaction experience.

[0058] Although existing electronic skin technologies can provide certain tactile perception capabilities, they often cannot achieve rich emotional interactions or high-quality human-machine interactions. In this application, by realizing fine tactile recognition and three-dimensional force decoupling, it can not only recognize simple tactile actions but also distinguish different emotional interactions (such as caressing, pinching, etc.), making the emotional expression of robots or interactive devices more natural and flexible, greatly enhancing the user's interaction experience.

[0059] To experimentally verify the bionic hair tactile sensor provided in this application, COMSOL Multiphysics software is used for simulation verification and mechanical response simulation. The conditions of the simulation experiment are: a single cilium (diameter 500 microns, height 5 mm) is deflected by 1° under a tangential force. Experimental results: the error from the actual measurement is less than 5%; the output matching degree of the sensor array is 92%.

[0060] ANSYS finite element analysis is used for mechanical response simulation. The model adopted is to simplify the bionic cilium into a flexible beam (elastic modulus 50 kPa). The experimental results obtained are: the maximum stress is 12.3 MPa (at the root of the cilium), and the measured fracture strength ≥ 15 MPa; the theoretical sensitivity is 6.63 μT / mN, and the measured error is 7.7%.

[0061] To characterize the elastic modulus of the device, a tensile test is carried out with a tensile machine, and then its elastic modulus is calculated with the help of corresponding physical formulas; the elastic modulus of the bionic hair is as Figure 5 shown.

[0062] With the above technical solution, the embodiment of the present application provides a bionic hair tactile sensor and application. The sensor includes: a substrate and a bionic cilia array disposed on the substrate; the bionic cilia array includes a plurality of flexible bionic cilia arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia; an optical system is integrated in the substrate, and the optical system includes an imaging module and a micro prism, and the micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; the imaging module is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to realize the decoupling and resolution of three-dimensional forces. The present application realizes high-resolution multi-dimensional force detection and direction detection through the change of the marking points caused by the deformation of the cilia, and uses the array design to expand to the functions of sliding perception and multi-modal object recognition.

[0063] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present application, and in actual applications, various changes can be made in form and details without departing from the spirit and scope of the present application. Any person skilled in the art can make various changes and modifications within the spirit and scope of the present application. Therefore, the protection scope of the present application should be subject to the scope defined by the claims.

Claims

1. A bionic hair tactile sensor, characterized in that, Comprising: a substrate and a bionic cilia array disposed on the substrate; The bionic cilia array includes a plurality of flexible bionic cilia arranged in an array, and a marking layer is provided at the bottom of the flexible bionic cilia; An optical system is integrated in the substrate, and the optical system includes an imaging module and a micro prism. The micro prism reflects the deformation of the marking layer at the bottom of the flexible bionic cilia to the imaging module; the imaging module is used to capture the deformation data of the marking layer and send the deformation data to the deep learning algorithm module to achieve decoupling and resolution of three-dimensional forces.

2. The bionic hair tactile sensor according to claim 1, wherein, The material of the flexible bionic cilia is PA6; the diameter of each flexible bionic cilium is 400 microns and the height is 8 mm.

3. The bionic hair tactile sensor according to claim 1, characterized in that The density of the flexible bionic cilia array is 10 - 25 pieces / cm².

4. The bionic hair tactile sensor according to claim 1, wherein The bionic cilia array is formed by a nylon particle hot pressing process, and the surface is activated by oxygen plasma to enhance the adhesion of the marking layer; the hot pressing process conditions are: molding at 280°C, holding pressure time of 90 seconds, and cooling gradient of 3°C / second.

5. The bionic hair tactile sensor according to claim 1, characterized in that The marking layer is impregnated with a marking material through a microgroove coating device and forms a stable coating after curing by a nano-UV LED array.

6. The bionic hair tactile sensor according to claim 5, characterized in that, The marking material is one of liquid silicone, polymer film or metal film.

7. The bionic hair tactile sensor according to claim 1, wherein, The Shore hardness of the marking layer is 20A and the thickness is 200 microns.

8. The bionic hair tactile sensor according to claim 1, characterized in that A marking layer is provided at the bottom of each flexible bionic cilium, and the thickness of the marking layer is 2 mm.

9. The bionic hair tactile sensor according to claim 1, characterized in that The imaging module is one of an embedded camera, a flexible endoscope, a CMOS sensor or a micro camera.

10. Application of a bionic hair tactile sensor according to any one of claims 1 to 9 in the field of perception of micro-force operation and complex contact states by an intelligent robot.

Images