Interaction robot based on AI

Abstract

According to the AI-based interactive robot provided by the utility model, the plurality of touch interaction areas are arranged on the surface of the robot body, and the vibration sensor electrically connected with the main control chip is arranged in at least one touch interaction area, so that the interactive robot can sense vibration information, and the dimension of sensing external information by the interactive robot is increased; according to the interactive robot, the TOF sensor is arranged to collect the distance information, so that the interactive robot has the visual perception capability instead of directly collecting image data, and the risk of privacy disclosure can be reduced or reduced. Furthermore, a microphone is arranged, so that the interactive robot has the capability of collecting sound information. Thus, the interactive robot has tactile, visual and auditory perception capabilities, information acquisition channels of the interactive robot are enriched, rich external environment information can be provided for the AI model, interaction feedback is generated accordingly, and the intelligent interaction capability of the interactive robot is improved.

Description

TECHNICAL FIELD

[0001] The utility model relates to robot technical field especially, it relates to an interactive robot based on AI. BACKGROUND

[0002] With the rapid development of artificial intelligence technology, especially the deep empowerment of CHATGPT and OPENAI to interactive robot products, intelligent interactive robots usher in a new stage of development. Today's AI interactive robots not only have high intelligence, can provide natural and smooth conversation experience, but also play an increasingly important role in many key areas. In the field of education, it can help children learn various knowledge; in emotional companionship, it provides thoughtful emotional communication for users; in the entertainment field, it enriches the game experience and pushes personalized entertainment content; even in the field of life assistance and safety monitoring, it also undertakes key tasks such as daily reminders and health monitoring.

[0003] However, the current intelligent interactive robots have obvious shortcomings in information input. Most interactive robots rely on cameras to collect information, which can obtain some visual information, but the risk of privacy leakage makes consumers worried. At a time when people pay more and more attention to privacy protection, this problem seriously limits the further popularization and development of intelligent interactive robots. UTILITY MODEL CONTENT

[0004] Therefore, the utility model provides an interactive robot based on AI which can improve information security, enrich information collection channels and improve interaction performance.

[0005] The utility model provides an interactive robot based on AI, which comprises:

[0006] A body 10, the surface of the body comprises a plurality of tactile interaction areas 11;

[0007] A main control chip 20 with AI function is arranged in the body 10;

[0008] A vibration sensor 21 is arranged in at least one tactile interaction area 11, and the vibration sensor 21 is electrically connected with the main control chip 20;

[0009] A TOF sensor 23 is electrically connected with the main control chip 20;

[0010] A microphone 24 is electrically connected with the main control chip 20;

[0011] A power supply 26 is electrically connected with the main control chip 20 and is used for providing electric energy.

[0012] In an embodiment, the interactive robot further comprises an electronic skin 22 covering at least one of the haptic interaction areas 11 and electrically connected to the main control chip 20.

[0013] In an embodiment, the body 10 of the interactive robot comprises at least one of a head, a face, a back, a buttock and a limb; the haptic interaction area 11 is arranged on the surface of at least one of the head, the face, the back, the buttock and the limb.

[0014] In an embodiment, the body 10 comprises a torso and a head; the TOF sensor 23 is arranged on the front and / or back of the upper half of the torso and / or the front and / or back of the head.

[0015] In an embodiment, the head of the body 10 comprises a forehead and / or an eye socket; the TOF sensor 23 is arranged on the forehead or the eye socket.

[0016] In an embodiment, a transparent protective layer is arranged outside the arrangement position of the TOF sensor 23.

[0017] In an embodiment, the interactive robot further comprises a two-in-one speaker 25 electrically connected to the main control chip 20 and having the functions of vibration and sound emission.

[0018] In an embodiment, the head of the body 10 has a mouth and an oral cavity; the speaker 25 is arranged in the oral cavity; the mouth has an opening forming a sound propagation channel.

[0019] In an embodiment, the torso of the body 10 has an abdominal cavity or a chest cavity; the speaker 25 is arranged in the abdominal cavity or the chest cavity; the abdominal cavity or the chest cavity has an opening forming a sound propagation channel.

[0020] In an embodiment, the interactive robot further comprises a communication module connected to the main control chip; the main control chip is connected to an AI large model in the cloud through the communication module.

[0021] The interactive robot based on AI provided by the utility model has the advantages that the interactive robot can perceive vibration information and increase the dimension of the interactive robot for perceiving external information, in addition, the interactive robot has visual perception ability by setting the TOF sensor 23 to collect distance information, instead of directly collecting image data, so that the risk of privacy leakage can be reduced and information security can be improved, further, the interactive robot has the ability of collecting sound information by setting the microphone 24, so that the interactive robot has tactile, visual and auditory perception ability, the information acquisition channel of the interactive robot is enriched, rich external environment information can be provided for the AI model, and interaction feedback can be generated based on the information, so that the intelligent interaction ability of the interactive robot can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 The structure schematic view of the interactive robot based on AI provided by the utility model embodiment is shown in the figure.

[0023] Figure 2 The structure schematic view of the front of the interactive robot based on AI in the embodiment is shown in the figure.

[0024] Figure 3 The structure schematic view of the back of the interactive robot is shown in the figure. Figure 2 The structure schematic view of the back of the interactive robot is shown in the figure.

[0025] Figure 4 The structure schematic view of the front of the interactive robot of embodiment 1 is shown in the figure.

[0026] Figure 5 The structure schematic view of the back of the interactive robot is shown in the figure. Figure 4 The structure schematic view of the back of the interactive robot is shown in the figure.

[0027] Figure 6 The structure schematic view of the back of the interactive robot of embodiment 3 is shown in the figure.

[0028] Figure 7 The structure schematic view of the back of the interactive robot is shown in the figure. Figure 6 The structure schematic view of the back of the interactive robot is shown in the figure.

[0029] In the figure: 10, body; 11, tactile interaction area; 20, main control chip; 21, vibration sensor; 22, electronic skin; 23, TOF sensor; 24, microphone; 25, loudspeaker; 26, power supply. DETAILED DESCRIPTION

[0030] In the following, the utility model is further described in combination with the drawings and the specific embodiment, and it should be noted that the following described embodiments or technical features can be combined to form new embodiments without conflict.

[0031] It should be noted that all directional indications (such as up, down, left, right, front, back, inner, outer, top, bottom, etc.) in the embodiments of the present application are intended to facilitate the understanding of relative positions between various components, and are not intended to indicate the specific orientation of the components in a particular attitude (as shown in the drawings). If the specific attitude changes, the directional indications will also change accordingly.

[0032] It should also be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or there can be a middle element. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or there can be a middle element.

[0033] As shown in Figures 1 to 3 An embodiment of the present application provides an interactive robot based on AI, which comprises a body 10, a main control chip 20, a vibration sensor 21, a TOF sensor 23, a microphone 24 and a power supply 26.

[0034] Among them, the body 10 of the interactive robot can be designed as a human or animal shape. The humanoid robot has a body structure similar to that of a human being, including a head, a torso, limbs, etc. This similarity makes the interactive robot more naturally imitate human actions and postures on the one hand. On the other hand, it helps the interactive robot to communicate and interact with humans more intuitively, such as conveying information and intentions through gestures, facial expressions and body language.

[0035] The animal-shaped robot usually has a cute or unique appearance. This appearance feature can attract people's attention and stimulate curiosity on the one hand, and help the robot establish an emotional connection with humans on the other hand, increasing the interest and interactivity of the interaction. As shown in Figure 2 and Figure 3 As shown in an embodiment, the body 10 of the interactive robot is designed as a puppy shape, and in other embodiments, the body of the interactive robot can also be designed as a vehicle such as a car, an airplane, etc.

[0036] As shown in Figure 1As shown, the vibration sensor 21, the TOF sensor 23, the microphone 24 and the power supply 26 are all electrically connected with the master control chip 20. The master control chip 20 has an AI function. The AI function can be implemented in two ways. The first way is that the master control chip 20 can directly integrate an AI model inside, so that the interactive robot can process and analyze data locally in real time, realize fast response and intelligent decision-making. The second way is that the master control chip 20 can further be provided with a communication module, which can communicate with the AI model deployed in the cloud, and use the powerful computing power and rich data resources of the cloud to provide more accurate and personalized intelligent interaction experience for users. By using the AI model, the interactive robot can realize deep interaction with the user, understand the user's intention, and provide intelligent services and support.

[0037] On this basis, the embodiment sets multi-dimensional sensors for the interactive robot, so that the interactive robot has multi-dimensional perception ability, and then uses the AI function to improve the interaction performance of the interactive robot.

[0038] Specifically, as shown in the figure, Figures 1 to 3 The surface of the body 10 of the interactive robot is provided with a plurality of tactile interaction areas 11. The tactile interaction area 11 refers to a specific area on the body of the interactive robot for tactile interaction. By setting a specific sensor in the tactile interaction area 11, the robot can perceive the user's interaction information from the tactile dimension. In specific applications, the location of the interaction area 11 on the body 10 can be set according to the user's interaction habits. For example, the interactive robot is in the form of a puppy, and the user is usually used to touching or patting the head and / or back of the puppy to convey information to the puppy, so the tactile interaction area 11 of the interactive robot in this form can be set on the head and back. For another example, the interactive robot is a humanoid robot, and the user is used to touching or patting the head, back and limbs of the robot to convey information to the robot, so the tactile interaction area 11 of the interactive robot in this form can be set on the head, back and limbs.

[0039] In an embodiment, at least one tactile interaction area 11 is provided with a vibration sensor 21. The vibration sensor 21 is electrically connected with the master control chip 20, and can convert mechanical vibration signals into electrical signals, so that the interactive robot can perceive vibration information, adding a new dimension to the interactive robot's perception of the external environment. In this way, when the user pats the tactile interaction area 11, the vibration sensor 21 can perceive the vibration information and the patting position.

[0040] The interactive robot is also provided with a TOF sensor 23 electrically connected with the main control chip 20. The TOF sensor 23 is used to collect distance information. The traditional interactive robot is usually provided with a camera to collect visual information, and this technical means brings the risk of privacy leakage. In the embodiment, the TOF sensor 23 is used to collect visual information, which measures the flight time of light pulses between the sensor and the target object, accurately calculates the distance, and then constructs visual information based on the distance information. For example, the TOF sensor 23 continuously collects face depth data at a high frequency of 5-60 times per second, and using the existing AI model, the subtle position changes of key facial feature points such as eye corners, mouth corners, and nose tips can be tracked in real time. Since it does not directly collect image data, it can reduce or reduce the risk of privacy leakage and improve information security.

[0041] The interactive robot is also provided with a microphone 24 electrically connected with the main control chip 20, which can collect sound information, and using the existing AI model can identify the user's voice instructions and / or emotions.

[0042] The interactive robot is also provided with a power supply 26 electrically connected with the main control chip 20, which provides power for the interactive robot.

[0043] The AI-based interactive robot of the utility model, through setting multiple tactile interaction areas 11 on the surface of the body, at least one tactile interaction area 11 is provided with a vibration sensor 21 electrically connected with the main control chip 20, so that the interactive robot can perceive vibration information, which increases the dimension of the interactive robot to perceive external information. In addition, the interactive robot collects distance information by setting a TOF sensor 23 to make the interactive robot have visual perception ability, instead of directly collecting image data, which can reduce or reduce the risk of privacy leakage. Further, by setting a microphone 24, the interactive robot has the ability to collect sound information. In this way, the interactive robot has tactile, visual and auditory perception ability, which enriches the information acquisition method of the interactive robot, can provide rich external environment information for the AI model, and generate interactive feedback by fusing multi-dimensional information, and improve its intelligent interaction ability.

[0044] In an embodiment, at least one tactile interaction area 11 can be provided with an electronic skin 22. The electronic skin 22 is electrically connected with the main control chip 20 and is used to perceive touch information. In this way, when a user touches the tactile interaction area 11, the electronic skin 22 can perceive the touch information and its touch position. Combined with the electronic skin 22 and the vibration sensor 21, the interactive robot can have tactile perception ability.

[0045] In an embodiment, the body 10 of the interactive robot comprises at least one of a head, a face, a back, a buttock and a limb; the tactile interaction area 11 is arranged on the surface of at least one of the head, the face, the back, the buttock and the limb. In this way, the tactile interaction area 11 can be arranged at the position corresponding to the human or animal body part on the head, the face, the back, the buttock and the limb of the robot, so that the interaction between the robot and the human is more natural, which helps to reduce the interaction barrier between the human and the robot and helps the user to quickly understand the interaction mode of the robot.

[0046] In an embodiment, the electronic skin 22 and the vibration sensor 21 can be arranged on the same or different tactile interaction areas 11 to adapt to the use habit of the user. For example, the user is used to patting or touching the back of the interactive robot, and the tactile interaction area 11 can be arranged on the back, and the electronic skin 22 and the vibration sensor 21 are arranged on the tactile interaction area 11. However, the user is used to touching the forehead of the interactive robot, and a tactile interaction area 11 can be arranged on the forehead of the interactive robot, and the electronic skin 22 is arranged on the tactile interaction area 11.

[0047] In an embodiment, the body comprises a torso and a head, and the TOF sensor 23 is arranged on the front and / or back of the upper half of the torso and / or the front and / or back of the head. By arranging the TOF sensor 23 on the upper half of the torso and / or the head, the TOF sensor 23 has a better field of view, which can reduce the interference of the ground or other low objects and clearly capture the spatial information of the head and the surrounding environment without obstruction.

[0048] Specifically, only one TOF sensor 23 can be arranged on the front of the upper half of the torso or the front of the head to collect the distance information of the front of the interactive robot and perceive the environment in front of the interactive robot. In an example, two TOF sensors 23 can be arranged on the front and the back of the interactive robot to collect the distance information of the front and the back of the interactive robot and perceive the environment in front of and behind the interactive robot. For example, the TOF sensor 23 is arranged on the front of the upper half of the torso and the back of the head. For another example, the TOF sensor 23 is arranged on the front and the back of the upper half of the torso. For another example, the TOF sensor 23 is arranged on the front and the back of the head.

[0049] In a specific embodiment, the upper half of the torso can be the chest position of the interactive robot, and the TOF sensor 23 is arranged on the chest position of the interactive robot.

[0050] In a specific embodiment, the head of the body comprises a forehead and / or an eye socket, and the TOF sensor 23 is arranged on the forehead or the eye socket.

[0051] In a specific application, the installation position of the TOF sensor 23 can be flexibly set according to the actual height of the interactive robot. If the interactive robot is relatively high, the TOF sensor 23 can be arranged at the chest position of the interactive robot. If the interactive robot is relatively low, in order to have a better collection field of view, the TOF sensor 23 can be arranged at the forehead or the eye socket of the interactive robot.

[0052] In an embodiment, a transparent protective layer for protecting the TOF sensor 23 is arranged outside the arrangement position of the TOF sensor 23. In this way, the protective effect of the TOF sensor can be achieved.

[0053] Specifically, the transparent protective layer can be a PMMA (acrylic) transparent protective layer. In order to achieve the best performance, the higher the light transmittance of the transparent protective layer is, the better. However, the light transmittance is usually required to be above 80%. In order to further improve the performance, an AR film can be coated on the PMMA (acrylic) transparent protective layer to increase the transmittance of PMMA and reduce the loss of light intensity. The transmittance of the PMMA material without coating film is 92%, and the transmittance of the PMMA material with double-sided coating film can be increased to 96%.

[0054] In an embodiment, the interactive robot is further provided with a loudspeaker 25 electrically connected with the main control chip 20. The interactive robot determines the interaction strategy by collecting multi-dimensional information and using the analysis result of AI. If the interaction strategy includes language interaction, the loudspeaker 25 can be used for feedback.

[0055] The loudspeaker 25 is a two-in-one loudspeaker with vibration and sound functions, which can improve the interaction performance of the interactive robot. When the interactive robot needs to interact with the user through voice, the audio signal is transmitted to the sound unit of the loudspeaker 25, and the sound is clearly and loudly played out through the carefully designed audio amplification and processing circuit. At the same time, when it is necessary to convey specific emotions or states, such as excitement, excitement, etc., the main control chip 20 sends corresponding vibration signals to the vibration unit of the loudspeaker 25. The vibration unit generates vibrations of different frequencies, amplitudes and lengths according to these signals, which are transmitted to the user through the shell of the interactive robot, so that the user can intuitively feel the emotional intensity expressed by the interactive robot. For example, when the interactive robot detects that the user has achieved a certain achievement and shows extreme excitement, it will not only express congratulations with cheerful and high-pitched voice, but also let the user feel its excitement through strong and rhythmic vibration, further enhancing the infectivity and authenticity of the interaction.

[0056] In an embodiment, the head of the body 10 has a mouth and an oral cavity, and the loudspeaker 25 is arranged in the oral cavity; the mouth has an opening forming a sound propagation channel. In this way, the mouth can be used as an opening forming a sound propagation channel, which can guide the sound to form a relatively closed and optimized sound channel in the oral cavity after the sound is emitted from the loudspeaker.

[0057] The torso of the body 10 has an abdominal cavity or a chest cavity, the speaker 25 is arranged in the abdominal cavity or the chest cavity, and the abdominal cavity or the chest cavity has an opening forming a sound propagation channel. In this way, the abdominal cavity or the chest cavity can serve as a natural resonance cavity of the speaker as a relatively large cavity structure.

[0058] The interactive robot based on AI of the utility model utilizes TOF sensor, microphone, vibration sensor, electronic skin, two-in-one speaker with vibration and sound emission functions and other multi-dimensional sensors to realize multi-dimensional data fusion sensing detection, so as to improve the interactive experience and function implementation.

[0059] Taking the form of the interactive robot as a puppy as an example, the utility model provides three embodiments of different installation positions of the above-mentioned devices.

[0060] Embodiment 1

[0061] As shown in Figure 4 and Figure 5 , the TOF sensor 23 is arranged at a prominent position on the front of the interactive robot, so as to ensure that the TOF sensor 23 can clearly and unobstructedly collect spatial information of the head and the surrounding environment. The TOF sensor 23 is installed in the eye socket of the eye part of the puppy, and a PMMA (acrylic) transparent protective layer is arranged at the outermost side of the eye. In order to achieve the best performance, the higher the light transmittance of the transparent protective layer is, the better. However, the light transmittance is usually required to be above 80%. In order to further improve the performance, an AR film can be coated on the PMMA (acrylic) transparent protective layer to increase the transmittance of PMMA and reduce the loss of light intensity. The transmittance of the PMMA material without coating is 92%, and the transmittance can be increased to 96% by coating a double-sided film. The TOF sensor 23 can be installed in only one eye socket or two eye sockets respectively. In this way, the field of view can be expanded, and more spatial information can be obtained. Preferably, the emission end and the receiving end of the TOF sensor 23 are both directed to the front.

[0062] The microphone 24 is arranged in the ear canal of the ear of the puppy. Preferably, two microphones 24 are arranged, that is, one microphone 24 is arranged in each ear, so as to realize stereo sound collection. At the same time, the time difference, phase difference and intensity difference of sound arriving at the ear from different directions can be used to spatially locate the sound source. Of course, the microphone can also be arranged on the head or other positions of the body, and the number thereof can be one or more.

[0063] The vibration sensor 21 is arranged at the outermost skin position of the body, such as the head, the back or the buttocks. The closer the vibration sensor is to the outer skin layer of each part, the higher the sensitivity is. In this way, when the puppy is hit by a hand, especially when the hit position is close to or hits the installation position of the vibration sensor 21, the sensitive detection can be realized.

[0064] The electronic skin 22 uniformly covers the surface area of the interactive robot that is easily contacted by a person, ensuring that the touch, pressing and other actions of the human body can be accurately perceived. For example, the electronic skin 22 is arranged on the head, or the back, or the buttocks, or the legs, etc. When a hand or other object contacts, such as touches a certain position of these parts, the puppy can perceive different input information according to the direction, force, speed and other action changes of the contact or touch.

[0065] The loudspeaker 25 is preferably arranged in the mouth of the puppy, and the mouth is preferably slightly open, with a gap between the teeth, so that the sound emitted by the loudspeaker in the mouth can be transmitted through the small opening and the gap. Preferably, the loudspeaker 25 is a two-in-one device integrating vibration and sound, which can ensure that the functions of vibration and sound emission are realized at the same time in a small loudspeaker volume. For example, when the puppy is shouting due to excitement, the dog's mouth also vibrates violently, which will have a stronger real immersive feeling.

[0066] The TOF sensor 23, the microphone 24, the vibration sensor 21, the electronic skin 22 and the loudspeaker 25 are all connected to the main control chip 20, and by using the existing AI model, the functions of face recognition and identity determination, expression depth analysis, multi-object recognition and motion response, intelligent obstacle avoidance and path planning, etc. can be realized, and multi-dimensional data fusion perception and interaction can be realized based thereon.

[0067] The intelligent interactive robot of the utility model innovatively deeply fuses the TOF sensor 23 and the microphone 24, the vibration sensor 21, the electronic skin 22 and the two-in-one loudspeaker 25 with vibration and sound emission functions and other various sensors. Taking an actual scene as an example, when the TOF sensor 23 recognizes that the owner is nearby and judges that the owner is in a smiling state through expression analysis, the microphone 24 synchronously captures the owner's laughter, and the electronic skin 22 sensitively perceives the owner's touching action and the behavior of patting the head. These multi-dimensional data from different sensors are deeply fused and analyzed in the main control chip inside the interactive robot, and according to the preset complex interaction logic, or by means of neural network model for joint training and analysis, the interactive robot can accurately plan the next interaction content with the owner, realizing the natural and smooth multi-dimensional information detection perception and interaction experience between people.

[0068] The two-in-one speaker 25 with vibration and sound functions plays an important role in the interaction process of the interactive robot. When the interactive robot needs to interact with the user through voice, the audio signal is transmitted to the sound unit of the speaker 25, and through the carefully designed audio amplification and processing circuit, the sound is clearly and loudly played out. At the same time, when it is necessary to convey specific emotions or states, such as excitement, excitement, etc., the control system will send corresponding vibration signals to the vibration unit of the speaker 25. The vibration unit generates vibrations of different frequencies, amplitudes and lengths according to these signals, and transmits them to the user through the shell of the interactive robot, so that the user can intuitively feel the emotional intensity expressed by the interactive robot. For example, when the interactive robot detects that the owner has achieved a certain achievement and shows extreme excitement, it will not only express congratulations with a happy and high-pitched voice, but also let the owner feel its excitement through strong and rhythmic vibrations, further enhancing the infectivity and authenticity of the interaction.

[0069] Through the above-mentioned careful design of hardware integration and software algorithm implementation, the interactive robot of the utility model can fully exert the synergistic advantages of various sensors, realize safe, efficient, rich and highly emotional interactive functions, and bring users an unprecedented interactive robot experience.

[0070] Embodiment 2

[0071] As shown in Figure 2 and Figure 3 , the TOF sensor 23 is arranged at a prominent position on the front of the interactive robot, ensuring that the sensor 23 can clearly and unobstructedly collect spatial information of the head and the surrounding environment. The TOF sensor 23 is installed on the forehead of the dog, and a PMMA (acrylic) transparent protective layer is arranged on the outermost side of the forehead. In order to achieve optimal performance, the higher the light transmittance of the transparent protective layer, the better, but it is usually required to be more than 80%. In order to further improve the performance, an AR film can be coated on the PMMA (acrylic) transparent protective layer to increase the transmittance of PMMA and reduce the loss of light intensity. The transmittance of the PMMA material without coating is 92%, and the transmittance can be increased to 96% by coating a double-sided film. The TOF sensor can be installed only on the forehead, and preferably, the emission end and the receiving end of the TOF sensor are both directed forward (towards the direction of the dog's eyes looking straight ahead). Figure 2 The direction of the dog's eyes looking straight ahead.

[0072] The TOF sensor 23 can also be installed on the back of the head. The TOF sensors 23 are arranged on the front and back of the interactive robot, respectively, which can realize spatial detection in front and behind, expand the field of view, and obtain more spatial information.

[0073] The installation mode of the microphone 24, the vibration sensor 21, the electronic skin 22 and the speaker 25 can be the same as that of embodiment 1, which will not be described here again.

[0074] Embodiment 3

[0075] As shown in Figure 6 and Figure 7 The TOF sensor 23 is arranged at a prominent position on the front of the interactive robot, ensuring that the sensor 23 can clearly and unobstructedly collect spatial information of the human head and the surrounding environment. The TOF sensor 23 is mounted on the upper part of the chest of the puppy, and a PMMA (acrylic) transparent protective layer is arranged on the outermost side of the upper part of the chest. In order to achieve optimal performance, the higher the light transmittance of the transparent protective layer, the better, but it is generally required to be above 80%. In order to further improve the performance, an AR film can be coated on the PMMA (acrylic) transparent protective layer to increase the transmittance of PMMA, reduce the loss of light intensity, and the transmittance of the PMMA material without coating is 92%, and the transmittance can be increased to 96% by coating a double-sided film. The TOF sensor 23 can be mounted only on the upper part of the chest, and preferably, the emission end and the receiving end of the TOF sensor 23 are both directed forward. Figure 6 The puppy's eyes are directed forward.

[0076] The loudspeaker 25 is preferably arranged in the chest cavity or abdominal cavity of the puppy, and a small opening (such as the navel) is preferably arranged at the corresponding position of the chest cavity or abdominal cavity, so that the sound emitted by the loudspeaker in the chest cavity or abdominal cavity can be transmitted through the small opening. Preferably, the loudspeaker is a two-in-one device integrating vibration and sound, which can ensure that the vibration and sound functions are realized at the same time in a relatively small loudspeaker volume. For example, when the puppy roars due to excitement, it also vibrates violently, which can provide a stronger sense of real immersion.

[0077] The microphone 24, the vibration sensor 21, and the electronic skin 22 can be mounted in the same way as in Embodiment 1, which will not be described here.

[0078] The above embodiments are only preferred embodiments of the present application, and cannot be used to limit the scope of protection of the present application. Any non-essential changes and substitutions made by those skilled in the art based on the present application are within the scope of the present application.

Claims

1. An AI-based interactive robot, characterized by, The application relates to an interactive robot, comprising the following parts: a body (10), the surface of which comprises a plurality of tactile interaction areas (11); a main control chip (20) with an AI function arranged in the body (10); a vibration sensor (21) arranged in at least one of the tactile interaction areas (11), which is electrically connected with the main control chip (20); a TOF sensor (23) electrically connected with the main control chip (20); a microphone (24) electrically connected with the main control chip (20); a power supply (26) electrically connected with the main control chip (20) and used for providing electric energy. 2.The AI-based interactive robot according to claim 1, characterized in that, The interactive robot further comprises an electronic skin (22) arranged on at least one of the tactile interaction areas (11) and electrically connected with the main control chip (20). 3.The AI-based interactive robot according to claim 1 or 2, characterized in that, The body (10) of the interactive robot comprises at least one of a head, a face, a back, a buttock and a limb; the tactile interaction area (11) is arranged on the surface of at least one of the head, the face, the back, the buttock and the limb. 4.The AI-based interactive robot according to claim 1, characterized in that, The body (10) comprises a trunk and a head; the TOF sensor (23) is arranged on the front and / or back surface of the upper half of the trunk and / or the front and / or back surface of the head. 5.The AI-based interactive robot according to claim 4, characterized in that, The head of the body (10) comprises a forehead and / or an eye socket; the TOF sensor (23) is arranged on the forehead or the eye socket. 6.The AI-based interactive robot according to claim 1, 4, or 5, wherein, A transparent protective layer is arranged outside the arrangement position of the TOF sensor (23). 7.The AI-based interactive robot according to claim 1, 4, or 5, wherein, The interactive robot further comprises a two-in-one loudspeaker (25) electrically connected with the main control chip (20) and having a vibration and sound emission function. 8.The AI-based interactive robot according to claim 7, characterized in that, The head of the body (10) has a mouth and an oral cavity; the loudspeaker (25) is arranged in the oral cavity; the mouth has an opening forming a sound propagation channel. 9.The AI-based interactive robot according to claim 7, characterized in that, The trunk of the body (10) has an abdominal cavity or a chest cavity; the loudspeaker (25) is arranged in the abdominal cavity or the chest cavity; the abdominal cavity or the chest cavity has an opening forming a sound propagation channel. 10.The AI-based interactive robot according to claim 1, wherein, The interactive robot further comprises a communication module connected with the main control chip; the main control chip is connected with an AI large model in the cloud through the communication module.

Images