Touch sensing component, and bionic structure and sensing feedback method based thereon

Abstract

The present application discloses a touch sensing component, and a bionic structure and a sensing feedback method based thereon. The touch sensing component comprises a sensing sheet, a sensor, and a controller. The sensing sheet is disposed on a touched object, and collects a touch sensing signal when being in direct or indirect contact with a touch object, and the sensing sheet can be bent or flattened when subjected to a force; the sensor is disposed on the touched object, and collects a pose signal of the touched object and an acceleration signal of the touch object when the sensing sheet is in direct or indirect contact with the touch object; and the controller is connected to the sensing sheet and the sensor, and the controller generates touch sensing information, pose information, and acceleration information on the basis of the received touch sensing signal, pose signal and acceleration signal. The present application solves the problem that it is difficult for a conventional touch sensor to be mounted on a curved surface part, improves the appearance simulation degree of a bionic body, and facilitates subsequent analysis and comprehensive signal collection.

Description

A tactile component, a biomimetic structure and its sensory feedback method

[0001] This application is based on and claims priority to Chinese Patent Application No. 202411708853.0, filed on November 27, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of bionics, and more particularly to a tactile component, a bionic structure, and a sensory feedback method thereof. Background Technology

[0003] To enable bionic robots to sense changes in their external environment and provide feedback, touch sensors are typically installed on their outer surfaces. However, this not only hinders the realization of realistic bionic shapes, but also makes it difficult to install touch sensors on curved surfaces, which are not ideal for the robot's shape. Furthermore, touch sensors only output the content of the touch itself, which is not conducive to subsequent analysis and thus prevents user-friendly interaction with the bionic robot. Application content

[0004] The purpose of this application is to overcome the shortcomings of the prior art and provide a tactile component, a bionic structure and its sensing feedback method, which aims to solve the problem that touch sensors cannot be installed on the curved parts of bionic robots and easily hinder the bionic shape.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] In a first aspect, this application provides a tactile component, including:

[0007] A sensor is placed on the object being touched, and collects touch sensing signals when the object is in direct or indirect contact with the sensor.

[0008] A sensor is disposed on the object being touched, and during the period when the sensing sheet directly or indirectly contacts the object being touched, it collects the attitude signal and acceleration signal of the object being touched.

[0009] The controller is connected to the sensing sheet and the sensor, and generates touch sensing information, posture information and acceleration information based on the received touch sensing signal, posture signal and acceleration signal;

[0010] The sensing sheet can be bent or flattened when subjected to force.

[0011] Furthermore, the sensing sheet is provided with at least one touch sensing electrode, which generates the touch sensing signal when it directly or indirectly contacts the touch object.

[0012] Furthermore, the sensing sheet includes a substrate and a conductor, both of which are disposed on the substrate. The conductor is connected to the touch sensing electrode, and the substrate is made of a flexible and bendable material.

[0013] Furthermore, the touch sensing electrode is a capacitive electrode or a pressure electrode.

[0014] Furthermore, the sensing sheet is provided with several openings and / or openings.

[0015] Secondly, this application also provides a biomimetic structure, including a biomimetic body and the aforementioned tactile component, wherein the object to be touched is the biomimetic body, the tactile component is disposed on the biomimetic body, the biomimetic body includes a contoured skeleton and a covering layer, the covering layer is disposed on the surface of the contoured skeleton, and the sensing sheet is disposed between the contoured skeleton and the covering layer.

[0016] Furthermore, the shape of the bionic body is humanoid, animaloid, or organ-shaped.

[0017] Furthermore, the bionic body is provided with the tactile components at different parts, and also includes a main controller. The tactile components corresponding to different parts of the bionic body are all electrically connected to the same main controller.

[0018] Thirdly, this application also provides a biomimetic structure-based sensory feedback method, including:

[0019] When a touch object comes into contact with another touch object, the touch sensing information, posture information, and acceleration information collected by the tactile components of each part of the touch object are obtained.

[0020] The touch sensing information, posture information, and acceleration information collected from the tactile components of each part are comprehensively analyzed to obtain the emotional information of the touched object;

[0021] The touched object generates feedback information based on the touch object's emotional information, including sound, action behavior, and facial expressions.

[0022] Fourthly, this application also provides a biomimetic structure-based sensory feedback method, including:

[0023] When the touch object comes into contact with the touched object, the touch sensing information, posture information and acceleration information collected by the tactile components of various parts of the touched object are obtained, as well as the voice or text interaction information between the touch object and the touched object.

[0024] The touch sensing information, posture information, acceleration information and interaction information collected from the tactile components of each part are comprehensively analyzed to obtain the emotional information of the touched object;

[0025] The touched object generates feedback information based on the touch object's emotional information, including sound, action behavior, and facial expressions.

[0026] The advantages of this application compared to existing technologies are as follows: A tactile component includes a sensing sheet, a sensor, and a controller. The sensing sheet is disposed on the object being touched and collects touch sensing signals when directly or indirectly contacting the object. The sensing sheet can bend or flatten under force. The sensor is disposed on the object being touched and collects the attitude signal and acceleration signal of the object being touched during the period when the sensing sheet is in direct or indirect contact with the object. The controller is connected to the sensing sheet and the sensor and generates touch sensing information, attitude information, and acceleration information based on the received touch sensing signals, attitude signals, and acceleration signals. The sensing sheet of this application can bend or flatten under force, thus it can be easily installed on various curved parts of biomimetic bodies such as humanoid robots or animal-like robots, without needing to be installed on the surface of the biomimetic body. This solves the problem that traditional touch sensors are difficult to install on curved parts, improving the appearance simulation of the biomimetic body. At the same time, the tactile component can not only collect touch sensing signals but also simultaneously acquire attitude signals and acceleration signals, which is beneficial for subsequent analysis. In addition, the tactile components can be deployed as independent modules in various parts of the bionic body, which is conducive to comprehensive signal acquisition of the bionic body.

[0027] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application, it can be implemented according to the contents of the specification. In order to make the above and other objectives, features and advantages of this application more obvious and understandable, preferred embodiments are given below and described in detail. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 is a structural schematic diagram of a tactile component provided in a specific embodiment of this application;

[0030] Figure 2 is a schematic diagram of the structure of a sensor sheet in a tactile component provided in a specific embodiment of this application;

[0031] Figure 3 is a schematic diagram of the installation of a tactile component according to a specific embodiment of this application;

[0032] Figure 4 is a schematic diagram illustrating the working principle of a biomimetic structure provided in a specific embodiment of this application;

[0033] Figure 5 is a schematic diagram of the external structure of a humanoid biomimetic structure provided in a specific embodiment of this application;

[0034] Figure 6 is a schematic diagram of a human-shaped biomimetic structure provided in a specific embodiment of this application (with the covering layer removed).

[0035] Figure 7 is a schematic diagram of the sensor sheet and sensor plate in a human-shaped biomimetic structure provided in a specific embodiment of this application;

[0036] Figure 8 is a schematic diagram of the structure in which the sensor sheet and sensor plate are mounted on the biomimetic skeleton in a human-shaped biomimetic structure according to a specific embodiment of this application.

[0037] Figure 9 is a schematic diagram of a humanoid biomimetic structure in a specific embodiment of this application, in which the sensor is mounted on the biomimetic skeleton.

[0038] Figure 10 is a schematic diagram of the shape of the sensing sheet at the abdominal position in a human-shaped biomimetic structure provided in a specific embodiment of this application;

[0039] Figure 11 is a schematic diagram of the external structure of an animal-shaped biomimetic structure provided in a specific embodiment of this application;

[0040] Figure 12 is a schematic diagram of an animal-shaped biomimetic structure provided in a specific embodiment of this application (with the covering layer removed).

[0041] Figure 13 is a schematic diagram of the sensor sheet and sensor plate in an animal-shaped biomimetic structure provided in a specific embodiment of this application;

[0042] Figure 14 is a schematic diagram of a sensor sheet mounted on a biomimetic skeleton in an animal-shaped structure according to a specific embodiment of this application.

[0043] Figure 15 is a schematic diagram of the structure in which the sensor sheet and sensor plate are mounted on the biomimetic skeleton in an animal-shaped biomimetic structure provided in a specific embodiment of this application.

[0044] Figure 16 is a schematic diagram of the shape of the sensor plate at the top of the head in a biomimetic structure in the shape of an animal provided in a specific embodiment of this application;

[0045] Figure 17 is a cross-sectional schematic diagram of a biomimetic structure in the shape of an organ provided in a specific embodiment of this application;

[0046] Figure 18 is an exploded view of a biomimetic structure in the shape of an organ provided in a specific embodiment of this application. Figure Labels

[0047] 1. Touch-sensing component; 11. Sensor sheet; 111. Touch sensing electrode; 112. Opening; 113. Opening; 12. Sensor; 13. Controller; 14. Sensing plate; 2. Bionic body; 21. Covering layer; 22. Bionic skeleton; 3. Main controller. Detailed Implementation

[0048] The technical solution of this application will be clearly and completely described below with reference to specific embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] As shown in Figures 1 to 4, this embodiment of the application provides a tactile component 1, including a sensing sheet 11, a sensor 12, and a controller 13. The sensing sheet 11 is disposed on the object being touched, and collects touch sensing signals when it directly or indirectly contacts the object. The sensing sheet 11 can be bent or flattened when subjected to force. The sensor 12 is disposed on the object being touched, and collects the posture signal and acceleration signal of the object being touched during the period when the sensing sheet 11 directly or indirectly contacts the object. The controller 13 is connected to the sensing sheet 11 and the sensor 12, and generates touch sensing information, posture information, and acceleration information based on the received touch sensing signals, posture signals, and acceleration signals.

[0050] The object being touched can be a person or an animal, and the object being touched can be a robot in humanoid or animaloid form, etc.

[0051] The sensing element 11 can bend or flatten under force, thus it can be easily installed on various curved parts of the biomimetic body 2, such as humanoid or animal-like robots, without needing to be installed on the surface of the biomimetic body 2. This solves the problem of traditional touch sensors 12 being difficult to install on curved surfaces, improving the appearance simulation of the biomimetic body 2. Simultaneously, the tactile component 1 can not only collect touch sensing signals but also simultaneously acquire attitude and acceleration signals, which is beneficial for subsequent analysis. Furthermore, the tactile component 1 can be used as an independent module and deployed in various parts of the biomimetic body 2, thereby facilitating comprehensive signal acquisition from the biomimetic body 2.

[0052] The sensor 12 and the controller 13 can be designed separately or as an integrated unit. In the integrated design, a substrate is provided, and the controller 13 and the sensor 12 are integrated on the substrate to form a sensing plate 14. The sensing plate 14 can be flexible or rigid. This design facilitates the installation of the touch component 1.

[0053] In this application, sensor 12 can refer to a single sensor or a combination of multiple types of sensors. When only one type of sensor is used, the sensor must at least have the ability to acquire the attitude signal and acceleration signal of the touched object. For example, a commercially available gyroscope sensor can provide attitude information by measuring angular velocity and is usually integrated with an accelerometer to capture the motion acceleration of the object. When multiple types of sensors are used, it can be a combination of an accelerometer and an attitude sensor, wherein the accelerometer is used to acquire the acceleration signal of the touched object, and the attitude sensor is used to acquire the attitude signal of the touched object.

[0054] In this application, the controller 13 should at least have the ability to process touch sensing signals from the sensing chip 11 and attitude and acceleration signals from the sensor 12, and analyze and process the received touch sensing signals, attitude signals, and acceleration signals to generate and output touch sensing information, attitude information, and acceleration information. The controller 13 with this capability can be a commercially available STM32, BeagleBone series, Arduino series, or Raspberry Pi series controller 13.

[0055] In this application, touch sensing information includes touch location, touch posture of the touch object (e.g., a human hand gesture when touching), and touch force. Posture information includes the current posture of the touch object and the posture change process of the touch object. Acceleration information includes the magnitude and direction of velocity change.

[0056] In one embodiment, the sensing sheet 11 is provided with at least one touch sensing electrode 111, which generates a touch sensing signal when it directly or indirectly contacts a touch object.

[0057] It should be noted that the touch sensing electrode 111 can contact the touch object directly or indirectly. Direct contact applies when the surface of the touch sensing electrode 111 is not covered with additional material. In the case of indirect contact, the outer surface of the touch sensing electrode 111 is covered with additional material, but this additional material must not affect the normal operation of the touch sensing electrode 111; that is, the touch sensing electrode 111 can still form a sensing contact through this layer of material.

[0058] In one embodiment, the sensing sheet 11 can be a plane or curved surface of any shape. The sensing sheet 11 includes a substrate and a conductor. The conductor and the touch sensing electrode 111 are both disposed on the substrate. The conductor is connected to the touch sensing electrode 111. The substrate is made of a flexible and bendable material.

[0059] The substrate of the sensing sheet 11 is made of a flexible and bendable material, such as polyimide (PI), polyethylene terephthalate (PET), or thermoplastic polyurethane (TPU). This ensures that the sensing sheet 11 can bend or flatten under stress, allowing it to conform to various curved surfaces during use while maintaining stable electrical performance and mechanical strength.

[0060] Conductors are formed by depositing or printing conductive materials onto a substrate. Commonly used materials include copper and silver nanowires, carbon nanotubes, or conductive polymers (such as PEDOT:PSS). These materials offer high conductivity and flexibility, making them suitable for flexible electronic devices. Conductors can be manufactured using methods such as screen printing, inkjet printing, or chemical vapor deposition. The pattern design of the conductor can be customized to meet application requirements, such as grids or stripes, to optimize touch sensitivity.

[0061] The touch sensing electrode 111 is also disposed on the substrate and directly connected to the conductor to form a complete touch sensing circuit. The touch sensing electrode 111 can be either a capacitive electrode or a pressure electrode. The two types of electrodes operate on different principles. Specifically, in practical applications, a capacitive electrode only requires designing electrodes of different shapes and sizes at the circuit ends of the sensing element 11 to transmit and receive changes in the capacitance electric field to measure touch signals. A pressure electrode, on the other hand, measures touch signals by causing deformation of the strain sensing device after pressure is applied to the electrode position, resulting in a change in its resistance.

[0062] The touch sensing electrode 111 can be made of the same material as the conductor layer or a different material to optimize contact resistance and signal transmission efficiency. The touch sensing electrode 111 is arranged in an array, such as a matrix arrangement.

[0063] The connection between the conductor and the touch sensing electrode 111 can be achieved through direct printing, conductive adhesive bonding, or co-deposition techniques to ensure the reliability and stability of the electrical connection.

[0064] In one embodiment, an FPCB is used as the substrate, copper is used as the conductor, and the touch sensing electrode 111 is a line end of different shapes and sizes.

[0065] In one embodiment, TPU and PET are used as substrates, conductive silver paste is used as conductors, and touch sensing electrodes 111 are line ends of different shapes and sizes.

[0066] In one embodiment, a fabric is used as the substrate, and flexible wires are arranged in the fabric or conductive silver paste is printed directly on the fabric. The touch sensing electrode 111 can be a capacitive electrode or a pressure electrode.

[0067] As shown in Figure 2, the sensing sheet 11 is provided with a plurality of openings 112 and / or openings 113.

[0068] The openings 112 and 113 on the sensor 11 can be designed as circular, elliptical, rectangular, or other geometric shapes. The size and distribution of the openings 112 can be adjusted according to specific requirements.

[0069] By providing openings 112 and / or openings 113 on the sensing sheet 11, positioning can be facilitated or it can be combined with other structural components to improve installation accuracy.

[0070] It should be noted that if the opening 112 and / or the opening 113 are located on the touch sensing electrode 111, and the touch sensing electrode 111 is a capacitive electrode, then the shape or area of ​​the touch sensing electrode 111 can be changed to adjust the signal.

[0071] This application embodiment also provides a biomimetic structure, including a biomimetic body 2 and the aforementioned tactile component 1. The object to be touched is the biomimetic body 2, and the tactile component 1 is disposed on the biomimetic body 2. The biomimetic body 2 includes a contoured skeleton 22 and a covering layer 21. The covering layer 21 is disposed on the surface of the contoured skeleton 22, and the sensing sheet 11 is disposed between the contoured skeleton 22 and the covering layer 21.

[0072] The shape of the bionic body 2 can be humanoid, animaloid, or organoid. Animal-shaped bionic bodies 2 can resemble dogs, cats, foxes, etc., while organ-shaped bionic bodies 2 can mimic features such as mouths and noses. It is important to note that organ-shaped bionic bodies 2 can exist independently or be combined with humanoid or animal-shaped bionic bodies 2 as part of the humanoid or animal-shaped bionic body 2.

[0073] The covering layer 21, serving as the external structure of the biomimetic skeleton 22, can be one or more layers of material, made of elastic or rigid materials, such as silicone, fur, or leather. The color, texture, and thickness of the covering layer 21 can be adjusted as needed to achieve the best biomimetic effect. The covering layer 21 can be fixed to the biomimetic skeleton 22 using adhesive bonding, mechanical fixation, heat fusion, or zippers.

[0074] The sensor 11 can be mounted on the inner side of the cover layer 21 or on the surface of the contour frame 22. Specifically, the sensor 11 can be directly mounted on the inner side of the cover layer 21 or the surface of the contour frame 22, or the mounting location can be designed with some structures to assist in mounting or positioning. For example, the inner side of the cover layer 21 or the surface of the contour frame 22 is designed with a protruding structure that corresponds to the opening 113 or aperture 112 on the sensor 11. In addition, the sensor 11 can be fixed to the inner side of the cover layer 21 or the surface of the contour frame 22 by means of adhesive bonding, sewing, or pressing.

[0075] The sensor 11 is positioned between the contoured skeleton 22 and the cover layer 21, preventing the sensor 11 from being directly exposed on the outer surface of the cover layer 21, thereby improving the appearance simulation of the bionic body 2. The sensor plate 14, which integrates the sensor 12 and the controller 13, can be fixed to the contoured skeleton 22 by screws or clips, and is located inside the cover layer 21. This design also prevents the sensor plate 14 from being directly exposed on the outer surface of the cover layer 21, thereby further improving the appearance simulation of the bionic body 2.

[0076] In one embodiment, the bionic body 2 is provided with tactile components 1 at different parts, and the bionic structure also includes a main controller 3. The tactile components 1 corresponding to different parts of the bionic body 2 are all electrically connected to the same main controller 3.

[0077] As shown in Figure 4, two or more haptic components 1 are connected to the same main controller 3. The main controller 3 can comprehensively collect and analyze the touch sensing information, posture information, and acceleration information collected from each haptic component 1. The comprehensive analysis by the main controller 3 can be based on existing machine learning models. During the development phase, the machine learning model is burned into the flash memory of the main controller 3 in the form of code. The main controller 3 can be an ARM Cortex series, AVR series MCU, or a single-chip microcomputer (SoC), etc.

[0078] As shown in Figures 5 to 10, in one embodiment, the bionic body 2 is in humanoid form (hereinafter referred to as a humanoid robot). The humanoid robot has tactile components installed on its upper arm, forearm, thigh, calf, abdomen, back, buttocks, chest, and face. The shape and size of the sensor plate 11 of the tactile component will be different for different installation locations.

[0079] Figure 5 illustrates the appearance of the humanoid robot. Since the tactile components are installed between the cover layer and the contoured skeleton 22, they are not visible from the robot's exterior, thus enhancing its realistic appearance. Figure 6 shows the humanoid robot with the cover layer removed, retaining only the contoured skeleton 22. In this state, no tactile components are present. Figure 7 shows the arrangement of multiple tactile components. The contoured skeleton 22 is not shown in this figure. As can be seen from Figure 7, the shape and size of the sensor plates 11 vary depending on the location of the contoured skeleton 22. Figure 8 shows the state where the tactile components are installed on the contoured skeleton 22. As can be seen from Figure 8, sensor plates 11 that conform to the curved surfaces can still be installed on certain curved surfaces of the contoured skeleton 22. Figure 9 shows the state where only the sensor plates 11 are installed on the contoured skeleton 22, without the sensor plate 14. Figure 10 shows the shape of the sensor plate 11 located on the abdomen of the humanoid robot.

[0080] As shown in Figures 11 to 16, in one embodiment, the bionic body 2 is in the shape of a panda (hereinafter referred to as the panda robot). The panda robot has tactile components installed on its upper arms, forearms, thighs, calves, abdomen, head, and face. The shape and size of the sensor plate 11 of the tactile component will be different for different installation locations.

[0081] Figure 11 illustrates the appearance of the panda robot. Since the tactile components are installed between the cover layer and the contoured skeleton 22, they are not visible from the robot's exterior, thus enhancing its realistic appearance. Figure 12 shows the panda robot with the cover layer removed, leaving only the contoured skeleton 22. In this state, no sensing components are present. Figure 13 shows the arrangement of multiple tactile components. The contoured skeleton 22 is not shown in this figure. As can be seen from Figure 13, the shape and size of the sensing plates 11 vary depending on the location of the contoured skeleton 22. Figure 14 shows the contoured skeleton 22 with only sensing plates 11 installed, without the sensing plate 14. Figure 15 shows the tactile components installed on the contoured skeleton 22. As can be seen from Figure 15, sensing plates 11 that conform to the curved surfaces can still be installed on certain curved surfaces of the contoured skeleton 22. Figure 16 shows the shape of the sensing plate 11 on the top of the panda robot's head.

[0082] As shown in Figures 17 and 18, in one embodiment, the bionic body 2 is in the shape of an organ, which has a channel, specifically simulating a human mouth or nostril. A human finger or other part can enter this channel. Since sensor plates 11 are distributed along the length of the channel, when a human finger or other part enters and moves within the channel, data such as the specific position of the finger or other part entering the channel, the dwell time, and the force applied can be collected, and data such as the movement speed can be indirectly calculated.

[0083] In one embodiment, the main controller 3 integrates a wireless communication module, which can be Wi-Fi, Bluetooth, Zigbee, LoRa, or other modules suitable for short- or medium-range wireless transmission. The main controller 3 wirelessly connects to the VR device or an electronic device with VR functionality (such as a smartphone, tablet, VR headset, etc.) through the wireless communication module. The initial position and posture of the virtual target object in the VR animation are set to be consistent with the posture of the bionic body 2. The information collected by the main controller 3 (including touch sensing information, posture information, acceleration information, and voice information, etc.) is sent to the VR device or an electronic device with VR functionality. The virtual target object in the VR animation will be synchronously presented in the same form as the bionic body 2. For example, when the arm of the humanoid robot is raised, the virtual target object in the VR animation also presents the same posture, ensuring the user's immersion in the VR environment.

[0084] It is worth noting that the information interaction between the VR device and the bionic host 2 can be viewed as a two-way information flow process. The VR device is not merely a passive data receiving terminal; it can also generate a large amount of useful information. For example, the VR device can capture the user's interactive actions in the virtual world (such as gestures and head movements), user perspective data, and the device's environmental perception information (such as spatial positioning and boundary awareness). This information can be merged with touch sensing information, posture information, acceleration information, and voice interaction information from the bionic host 2. The merged information will then be sent back to the VR device, synchronizing the virtual target object's movements with the user's physical interactions.

[0085] It should be noted that if multiple bionic bodies 2 need to be synchronized to the same VR device, data aggregation and distribution can be performed through network protocols (such as MQTT or WebSocket) on the main controller 3, thereby enabling multiple virtual characters to perform collaborative actions in the same VR environment.

[0086] In addition to VR applications, it can also be extended to AR scenarios, that is, mapping the posture information of the bionic body 2 onto AR devices, allowing users to see virtual objects superimposed on the real scene and interact with the bionic body 2 through AR glasses or mobile phones.

[0087] This application embodiment also provides a sensing feedback method for a biomimetic structure, which is applicable to the above-mentioned biomimetic structure. The method specifically includes the following steps: S10A-S30A.

[0088] S10A: When the touch object comes into contact with the object being touched, the touch sensing information, posture information and acceleration information collected by the tactile components of each part of the object being touched are obtained.

[0089] In this embodiment, the object being touched is a human, and the object being touched is a humanoid robot.

[0090] The tactile components of each part of the humanoid robot are all connected to the same main controller. The main controller can receive the touch sensing information, posture information and acceleration information collected by the tactile components of each part.

[0091] S20A: The touch sensing information, posture information and acceleration information collected by the touch sensing components of each part are comprehensively analyzed to obtain the emotional information of the touched object.

[0092] Since the main controller can perform comprehensive analysis based on existing machine learning models, specifically, it collects touch sensing information, posture information, and acceleration information from each tactile component. This data is normalized to adjust data of different dimensions to a uniform scale, and filters are used to eliminate noise and improve data quality. A deep neural network (DNN) or convolutional neural network (CNN) is selected to divide the normalized data into training, validation, and test sets. The model is trained using a feedforward neural network, enabling it to learn the relationship between tactile data and expected output. Backpropagation algorithms and optimizers (such as Adam or SGD) are used to update model parameters. The loss function is monitored to ensure gradual model convergence. Hyperparameter tuning involves adjusting the network structure (such as the number of layers and neurons) and training parameters (such as learning rate and batch size) to optimize model performance. The trained model is then burned into the main controller. After burning, the main controller can analyze the data transmitted from each tactile component in real time to analyze human emotional information, such as whether the transmitted emotions are friendly, aggressive, mischievous, happy, angry, or sad.

[0093] S30A: The touched object generates feedback information based on the emotional information of the touched object. The feedback information includes sound, action behavior and facial expression.

[0094] In this embodiment, appropriate sounds, such as speech, music, or other sound effects, are played through the humanoid robot's built-in or remotely connected audio output device. The content of the sound is related to emotional information, such as cheerful speech or comforting music. The humanoid robot's control actuators enable body movements, such as waving, nodding, and hugging, which correspond to emotional states. For example, if a person simultaneously grasps both of the humanoid robot's upper arms and laughs heartily, the robot will determine that the person is likely experiencing happiness and wants to communicate and share with the robot. When the robot recognizes that a person is happy, it can make movements such as opening its arms. The humanoid robot possesses facial animation capabilities, changing facial expressions through servo motors or flexible materials to reflect emotional information. For example, when it recognizes that a touched object is showing worry, its face may display a concerned expression.

[0095] It should be noted that the humanoid robot's movements and facial expressions can be achieved through mechanical actions. The mechanical structure design of the humanoid robot can be referenced in Chinese patents with publication numbers CN117124343A and CN117138362A.

[0096] This application embodiment also provides a sensing feedback method for a biomimetic structure, which is applicable to the above-mentioned biomimetic structure. The method specifically includes the following steps: S10B-S30B.

[0097] S10B: When the touch object contacts the touched object, the touch sensing information, posture information and acceleration information collected by the tactile components of various parts of the touched object are obtained, as well as the voice or text interaction information between the touch object and the touched object are obtained.

[0098] In this embodiment, the object being touched is a human, and the object being touched is a humanoid robot.

[0099] The tactile components of each part of the humanoid robot are all connected to the same main controller 3. The main controller 3 can receive the touch sensing information, posture information and acceleration information collected by the tactile components of each part, as well as the voice or text interaction information between the human and the humanoid robot.

[0100] It should be noted that if a person interacts with a humanoid robot using language, the robot's speech-to-text module can be used to convert the spoken content into text for further processing, thus reducing the complexity of the process. Of course, the speech-to-text module can be integrated into the main controller or exist independently.

[0101] S20B: The touch sensing information, posture information, acceleration information and interaction information collected from the touch-sensing components of each part are comprehensively analyzed to obtain the emotional information of the touched object.

[0102] Since the main controller can perform comprehensive analysis based on existing machine learning models, specifically, it collects touch sensing information, posture information, and acceleration information from each tactile component. This data is normalized to adjust data of different dimensions to a uniform scale, and filters are used to eliminate noise and improve data quality. A deep neural network (DNN) or convolutional neural network (CNN) is selected to divide the normalized data into training, validation, and test sets. The model is trained using a feedforward neural network, enabling it to learn the relationship between tactile data and expected output. Backpropagation algorithms and optimizers (such as Adam or SGD) are used to update model parameters. The loss function is monitored to ensure gradual model convergence. Hyperparameter tuning involves adjusting the network structure (such as the number of layers and neurons) and training parameters (such as learning rate and batch size) to optimize model performance. The trained model is then burned into the main controller. After burning, the main controller can analyze the data transmitted from each tactile component in real time to analyze human emotional information, such as whether the transmitted emotions are friendly, aggressive, mischievous, happy, angry, or sad. The algorithm for sentiment analysis based on voice or text interaction information can be found in Chinese patent publication number CN118916466A.

[0103] S30B: The touched object generates feedback information based on the emotional information of the touched object. The feedback information includes sound, action behavior and facial expression.

[0104] In this embodiment, appropriate sounds, such as speech, music, or other sound effects, are played through the humanoid robot's built-in or remotely connected audio output device. The content of the sound is related to emotional information, such as cheerful speech or comforting music. The humanoid robot's control actuators enable body movements, such as waving, nodding, and hugging, which correspond to emotional states. For example, if a person simultaneously grasps both of the humanoid robot's upper arms and laughs heartily, the robot will determine that the person is likely experiencing happiness and wants to communicate and share with the robot. When the robot recognizes that a person is happy, it can make movements such as opening its arms. The humanoid robot possesses facial animation capabilities, changing facial expressions through servo motors or flexible materials to reflect emotional information. For example, when it recognizes that a touched object is showing worry, its face may display a concerned expression.

[0105] It should be noted that the humanoid robot's movements and facial expressions can be achieved through mechanical actions. The mechanical structure design of the humanoid robot can be referenced in Chinese patents with publication numbers CN117124343A and CN117138362A.

[0106] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A tactile component, characterized in that, include: A sensor is placed on the object being touched, and collects touch sensing signals when the object is in direct or indirect contact with the sensor. A sensor is disposed on the object being touched, and during the period when the sensing sheet directly or indirectly contacts the object being touched, it collects the attitude signal and acceleration signal of the object being touched. The controller is connected to the sensing sheet and the sensor, and generates touch sensing information, posture information and acceleration information based on the received touch sensing signal, posture signal and acceleration signal; The sensing sheet can be bent or flattened when subjected to force.

2. The tactile component according to claim 1, characterized in that, The sensing sheet is provided with at least one touch sensing electrode, which generates the touch sensing signal when it comes into direct or indirect contact with the touch object.

3. A tactile component according to claim 2, characterized in that, The sensing sheet includes a substrate and a conductor. The conductor and the touch sensing electrode are both disposed on the substrate. The conductor is connected to the touch sensing electrode. The substrate is made of a flexible and bendable material.

4. A tactile component according to claim 3, characterized in that, The touch sensing electrode is a capacitive electrode or a pressure electrode.

5. A tactile component according to claim 1, characterized in that, The sensor sheet is provided with several openings and / or openings.

6. A biomimetic structure, characterized in that, The device includes a bionic body and at least one tactile component as described in any one of claims 1-5, wherein the object being touched is the bionic body, the tactile component is disposed on the bionic body, the bionic body includes a contoured skeleton and a covering layer, the covering layer is disposed on the surface of the contoured skeleton, and the sensing sheet is disposed between the contoured skeleton and the covering layer.

7. A biomimetic structure according to claim 6, characterized in that, The bionic body is shaped like a human, an animal, or an organ.

8. A biomimetic structure according to claim 6 or 7, characterized in that, The bionic body is provided with the tactile components at different parts, and also includes a main controller. The tactile components corresponding to different parts of the bionic body are all electrically connected to the same main controller.

9. A biomimetic structural perception feedback method, characterized in that, include: When a touch object comes into contact with another touch object, the touch sensing information, posture information, and acceleration information collected by the tactile components of each part of the touch object are obtained. The touch sensing information, posture information, and acceleration information collected from the tactile components of each part are comprehensively analyzed to obtain the emotional information of the touched object; The touched object generates feedback information based on the touch object's emotional information, including sound, action behavior, and facial expressions.

10. A biomimetic structural perception feedback method, characterized in that, include: When the touch object comes into contact with the touched object, the touch sensing information, posture information and acceleration information collected by the tactile components of various parts of the touched object are obtained, as well as the voice or text interaction information between the touch object and the touched object. The touch sensing information, posture information, acceleration information and interaction information collected from the tactile components of each part are comprehensively analyzed to obtain the emotional information of the touched object; The touched object generates feedback information based on the touch object's emotional information, including sound, action behavior, and facial expressions.

Images