Body-equipped robot generation method, body-equipped robot and related equipment

Abstract

The invention provides a generation method of a robot with a body, the robot with the body and related equipment. The method comprises the following steps: acquiring appearance feature customization information; determining at least one part building block and / or at least one component based on the appearance characteristic customization information; interaction feature customization information is obtained; generating interaction features based on the interaction feature customization information; under the condition that the at least one part building block and / or the at least one assembly are / is assembled to a preset robot trunk with a body, a target robot is generated; wherein each part building block corresponds to one machine part of the robot with the body, the assembly is obtained based on 3D printing, and the target robot operates the interaction mode represented by the interaction characteristics. According to the embodiment of the invention, the robot with the body, which meets the personalized customization requirements of the user and is high in appearance customization degree and biological characteristic customization degree, is provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robots, in particular to a method for generating a body-equipped robot, a body-equipped robot and related equipment. BACKGROUND

[0002] A body-equipped intelligent robot is an artificial intelligence system with a physical entity, which interacts with the physical environment through multi-modal sensors, and realizes closed-loop control of perception, decision-making and execution by integrating machine vision, natural language understanding and other technologies.

[0003] With the development of robot technology, body-equipped robots are gradually moving from the industrial field to personalized scenarios such as the family, social interaction and service. Existing body-equipped robots are mainly configured by manufacturers. That is, the body-equipped robots adopt standardized and industrialized shells, and the appearance is uniform. Moreover, the existing body-equipped robots rely on 3D printing technology in the assembly process, and the components obtained by 3D printing need to be assembled, and 3D printing consumes a long time, resulting in a long assembly cycle of the body-equipped robot.

[0004] The interactive behavior of the body-equipped robot is usually based on a few preset modes, and users cannot customize according to personal preferences, resulting in a stiff interactive experience.

[0005] Therefore, there is an urgent need for a body-equipped robot with high customization degree of appearance modeling, high customization degree of interaction, and short assembly time. SUMMARY

[0006] The embodiments of the present application provide a method for generating a body-equipped robot, a body-equipped robot and related equipment, which can solve the technical problem of low customization degree of appearance and low customization degree of interaction of the existing body-equipped robot.

[0007] In a first aspect, the embodiments of the present application provide a method for generating a body-equipped robot, characterized in that the method comprises: obtaining appearance feature customization information input by a user, the appearance feature customization information being used to represent the external modeling of the body-equipped robot; determining at least one part block and / or at least one component based on the appearance feature customization information; obtaining interaction feature customization information input by a user, the interaction feature customization information being used to represent the interaction mode of the body-equipped robot; generating an interaction feature based on the interaction feature customization information; generating a target robot in the case of assembling the at least one part block and / or the at least one component to a preset body-equipped robot torso; Each part block corresponds to a machine part of the body robot, the assembly is obtained based on 3D printing, and the target robot runs the interaction mode represented by the interaction feature.

[0008] Optionally, based on the appearance feature customization information, at least one part block is determined, including: Based on the appearance feature customization information, a first three-dimensional model is generated; The first three-dimensional model is automatically topologically optimized and lightweighted to obtain a second three-dimensional model; Based on a preset machine part area, the second three-dimensional model is cut into at least one target three-dimensional model corresponding to a machine part; Based on the at least one target three-dimensional model, at least one part block is determined.

[0009] Optionally, the appearance feature customization information includes any one of three-dimensional scanning data, image data, and appearance feature customization instructions; The first three-dimensional model is generated based on the appearance feature customization information, including: In the case that the appearance feature customization information is three-dimensional scanning data, modeling is performed based on the three-dimensional scanning data to obtain the first three-dimensional model; In the case that the appearance feature customization information is image data, three-dimensional reconstruction is performed based on the image data to obtain the first three-dimensional model; In the case that the appearance feature customization information is appearance feature customization instructions, in response to the appearance feature customization instructions, a first three-dimensional model corresponding to the appearance feature customization instructions is selected from a preset model library.

[0010] Optionally, the interaction feature customization information includes text information, voice information, and video information; The interaction feature is generated based on the interaction feature customization information, including: Text feature extraction is performed on the text information to obtain a text feature vector; The voice information is feature-extracted by a preset voice encoder to obtain an acoustic feature vector; The video information is feature-extracted by a pose estimation algorithm and an expression recognition algorithm to obtain a motion feature vector; Multi-modal feature fusion is performed on the text feature vector, the acoustic feature vector, and the motion feature vector to obtain an interaction feature embedding vector; The interaction feature embedding vector is input into a preset generative adversarial network to obtain an interaction feature.

[0011] Optionally, the interaction characteristic customization information comprises basic information, character information, experience information, and skill information, and the generating of the interaction characteristic based on the interaction characteristic customization information comprises: parsing the interaction characteristic customization information to obtain the basic information, the character information, the experience information, and the skill information; performing matching based on the basic information, the character information, the experience information, and the skill information to generate the interaction characteristic.

[0012] Optionally, the interaction characteristic customization information comprises expression information, and the generating of the interaction characteristic based on the interaction characteristic customization information comprises: parsing the interaction characteristic customization information to obtain the expression information; setting the interaction mode of the face region of the embodied robot based on the expression information.

[0013] Optionally, the interaction characteristic customization information comprises voice information and action information, and the generating of the interaction characteristic based on the interaction characteristic customization information comprises: parsing the interaction characteristic customization information to obtain the voice information and the action information; setting the sound line of the embodied robot based on the voice information; setting the action amplitude and the robot posture of the embodied robot based on the action information.

[0014] In a second aspect, an embodiment of the present application provides a generating device of an embodied robot, which comprises: a first obtaining module configured to obtain user-inputted appearance characteristic customization information, the appearance characteristic customization information being used to represent the external modeling of the embodied robot; a determining module configured to determine at least one part block and / or at least one component based on the appearance characteristic customization information; a second obtaining module configured to obtain user-inputted interaction characteristic customization information, the interaction characteristic customization information being used to represent the interaction mode of the embodied robot; a first generating module configured to generate an interaction characteristic based on the interaction characteristic customization information; a second generating module configured to generate a target robot in a case where the at least one part block and / or the at least one component are assembled to a preset embodied robot torso; wherein each part block corresponds to one robot part of the embodied robot, the component is obtained based on 3D printing, and the target robot runs the interaction mode represented by the interaction characteristic.

[0015] Thirdly, embodiments of this application provide an electronic device, which includes a bus, a transceiver, an antenna, a bus interface, a processor, and a memory. The processor stores a computer program, and when the computer program is executed by the processor, it implements the steps of the embodied robot generation method described above.

[0016] Fourthly, embodiments of this application provide a sculpted robot, which is obtained using the sculpted robot generation method described above.

[0017] This application provides a method for generating a hymenical robot, a hymenical robot, and related equipment. The method includes: acquiring user-inputted shape feature customization information, which characterizes the external shape of the hymenical robot; determining at least one part block and / or at least one component based on the shape feature customization information; acquiring user-inputted interaction feature customization information, which characterizes the interaction mode of the hymenical robot; generating interaction features based on the interaction feature customization information; and generating a target robot by assembling at least one part block and / or at least one component to a preset hymenical robot torso; wherein each part block corresponds to a machine part of the hymenical robot, the component is obtained based on 3D printing, and the target robot operates according to the interaction mode characterized by the interaction features. In this application embodiment, the appearance of the hymenical robot is customized using user-inputted shape feature customization information, and the interaction mode of the hymenical robot is customized using user-inputted interaction feature customization information, thereby providing a hymenical robot that meets users' personalized customization needs, has a high degree of shape customization, and a high degree of biometric customization. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for generating an embodied robot according to an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a device for generating an embodied robot according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. 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.

[0021] See Figure 1 , Figure 1 This is a flowchart of a method for generating an embodied robot according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps: Step 101: Obtain the user-inputted shape feature customization information.

[0022] Step 102: Based on the shape feature customization information, determine at least one part of the building block and / or at least one component.

[0023] It should be noted that the method for generating a embodied robot provided in this application embodiment can be applied to electronic devices. Users input shape feature customization information and interaction feature customization information into the electronic device, thereby realizing the customization of the embodied robot.

[0024] It should be understood that the aforementioned shape customization information includes, but is not limited to: 3D scan data, image data, and shape feature customization instructions. Shape feature customization information is used to characterize the external form of the embodied robot.

[0025] The aforementioned 3D scan data can be data obtained from scanning the user's human body, or data obtained from scanning other objects using a 3D scanning device. The aforementioned image data can be 2D image data or medical image data.

[0026] In this step, the user-inputted shape feature customization information is obtained, and based on this shape feature customization information, at least one part block and / or at least one component is determined. The aforementioned part block corresponds to a machine part of the embodied robot, and the aforementioned component is obtained through 3D printing. It should be noted that 3D printing instructions are generated based on the user-input shape feature customization information, and then the component is obtained through 3D printing.

[0027] Step 103: Obtain customized information on user-input interaction features.

[0028] Step 104: Generate interactive features based on the customized information of the interactive features.

[0029] It should be understood that the aforementioned customized interaction features include, but are not limited to, text information, voice information, and video information. This customized interaction feature information is used to characterize the interaction methods of the embodied robot.

[0030] The aforementioned text information may be a piece of text input by the user; the aforementioned voice information may be a piece of audio input by the user; and the aforementioned video information may be a piece of video input by the user.

[0031] In this step, user-inputted interaction feature customization information is obtained, and interaction features are generated based on this information. For details on how to generate interaction features, please refer to subsequent embodiments.

[0032] Step 105: By assembling at least one part block and / or at least one component into a preset embodied robot torso, a target robot is generated.

[0033] In this step, one optional implementation is to assemble at least one part of the building blocks into a preset body robot torso to obtain the target robot.

[0034] Another alternative implementation involves assembling at least one component into a pre-defined body robot to obtain the target robot.

[0035] Another alternative implementation involves assembling at least one part block and at least one component into a pre-defined body robot torso to obtain the target robot.

[0036] After obtaining the interaction characteristics, the interaction method of the embodied robot is set to obtain the target robot. In this embodiment, the appearance of the embodied robot is customized by user-inputted shape feature customization information, and the interaction method of the embodied robot is customized by user-inputted interaction feature customization information. This provides an embodied robot that meets users' personalized customization needs, has a high degree of appearance customization, and a high degree of biometric customization.

[0037] Furthermore, in this embodiment of the application, a hymen robot can be obtained by assembling building blocks and / or components onto the torso, thereby eliminating the need for a large amount of time spent on 3D printing operations and reducing the assembly time of the hymen robot.

[0038] In addition, designers can be deeply involved in the production process of embodied robots, and can customize the appearance and interaction methods of embodied robots in depth, thereby increasing the designer's sense of participation.

[0039] Optionally, based on the shape feature customization information, at least one part of the building block is determined, including: Based on the customized information of the aforementioned shape features, a first three-dimensional model is generated; The first 3D model is automatically topologically optimized and lightweighted to obtain the second 3D model. Based on the preset machine part regions, the second three-dimensional model is cut into at least one target three-dimensional model that corresponds one-to-one with the machine parts. Based on the at least one target 3D model, determine at least one part of the building blocks.

[0040] In this embodiment, after receiving the user's input of shape feature customization information, a first three-dimensional model is generated based on the shape feature customization information.

[0041] Furthermore, the first 3D model is subjected to automatic topology optimization and lightweighting.

[0042] Optionally, a voxel-based topology optimization method or a boundary representation-based topology optimization method can be used to automatically optimize the topology of the first 3D model.

[0043] Alternatively, a mesh simplification algorithm or feature simplification method can be used to lightweight the first 3D model.

[0044] The first 3D model, after automatic topology optimization and lightweighting, is determined as the second 3D model. Based on preset machine part regions, the second 3D model is segmented to obtain target 3D models that correspond one-to-one with the machine parts. Then, the part blocks corresponding to these target 3D models are determined.

[0045] In this embodiment, a 3D model is created based on the user-inputted shape feature customization information, and the 3D model is divided into target 3D models that correspond one-to-one with the machine parts, thereby realizing the customized design of the appearance of the embodied robot.

[0046] Optionally, the shape feature customization information includes any one of three-dimensional scan data, image data, and shape feature customization instructions; The step of generating a first 3D model based on the customized information of the shape features includes: When the shape feature customization information is three-dimensional scan data, a model is created based on the three-dimensional scan data to obtain a first three-dimensional model; When the shape feature customization information is image data, a three-dimensional reconstruction is performed based on the image data to obtain a first three-dimensional model; When the shape feature customization information is a shape feature customization instruction, in response to the shape feature customization instruction, a first three-dimensional model corresponding to the shape feature customization instruction is selected from a preset model library.

[0047] As mentioned above, the shape feature customization information includes any one of the following: 3D scan data, image data, and shape feature customization instructions.

[0048] In an optional embodiment, if the shape feature customization information is three-dimensional scan data, then three-dimensional modeling can be performed based on the three-dimensional scan data to obtain a first three-dimensional model.

[0049] In another alternative embodiment, if the shape feature customization information is image data, then the two-dimensional image data can be reconstructed in three dimensions to obtain a first three-dimensional model.

[0050] In another optional embodiment, a model library is pre-set, which stores multiple three-dimensional models. If the shape feature customization information is a shape feature customization instruction, then the model pointed to by the shape feature customization instruction in the model library is determined as the first three-dimensional model.

[0051] Optionally, the customized interaction features include text information, voice information, and video information; The process of generating interaction features based on the customized information of the interaction features includes: Text features are extracted from the text information to obtain a text feature vector; The speech information is processed by a preset speech encoder to extract features and obtain an acoustic feature vector. The video information is used to extract features using pose estimation and facial expression recognition algorithms to obtain action feature vectors. Multimodal feature fusion is performed on the text feature vector, the acoustic feature vector, and the action feature vector to obtain an interaction feature embedding vector; The interaction features are embedded into a vector and input into a preset generative adversarial network to obtain the interaction features.

[0052] The interactive feature customization information includes text information, voice information, and video information.

[0053] In this embodiment, for text information, text features are extracted to obtain a text feature vector. For speech information, features are extracted from the speech information using a preset speech encoder to obtain an acoustic feature vector. For video information, pose feature vectors are extracted from the video information using a pose estimation algorithm, and facial expression feature vectors are extracted from the video information using an expression recognition algorithm. The pose feature vectors and facial expression feature vectors are then used to determine the action feature vectors.

[0054] Furthermore, multimodal feature fusion is performed on the text feature vector, acoustic feature vector, and action feature vector to obtain the interaction feature embedding vector. The interaction feature embedding vector is then input into a pre-defined generative adversarial network to obtain the interaction features.

[0055] In this embodiment, a text feature vector is generated based on the text information input by the user; an acoustic feature vector is generated based on the voice information input by the user; and an action feature vector is generated based on the video information input by the user. The text feature vector, acoustic feature vector, and action feature vector are fused using multimodal features to obtain an interaction feature embedding vector. The interaction feature embedding vector is then input into a preset generative adversarial network to obtain interaction features, thereby realizing the customized design of the interaction method for the embodied robot.

[0056] Optionally, the customized interaction feature information includes basic information, personality trait information, experience information, and skill information. Generating interaction features based on the customized interaction feature information includes: The interaction feature customization information is analyzed to obtain basic information, personality trait information, experience information, and skill information; Based on the basic information, personality traits, experience, and skills, interaction features are generated through matching.

[0057] The aforementioned customized interactive features include basic information, personality traits, experience, and skills.

[0058] In this embodiment, after obtaining the interaction feature customization information, the interaction feature customization information is parsed to obtain basic information, personality feature information, experience information, and skill information.

[0059] The system generates interactive features by matching basic information, personality traits, experience, and skills. Optionally, the basic information, personality traits, experience, and skills can be input into a preset database that stores the mapping relationship between feature information and interactive features. Interactive features can then be generated by customizing information based on these interactive features.

[0060] Optionally, the interaction feature customization information includes facial expression information, and the step of generating interaction features based on the interaction feature customization information includes: The interaction feature customization information is analyzed to obtain the facial expression information; Based on the facial expression information, the interaction method for the face area of ​​the avatar robot is set.

[0061] The aforementioned interactive feature customization information includes emoticon information.

[0062] In this embodiment, after obtaining the interaction feature customization information, the interaction feature customization information is parsed to obtain expression information, and based on the expression information, the interaction mode of the robot's face area is set.

[0063] Optionally, the interaction feature customization information includes voice information and motion information, and the step of generating interaction features based on the interaction feature customization information includes: The interaction feature customization information is analyzed to obtain the sound information and the action information; Based on the sound information, the voice of the embodied robot is set; Based on the motion information, the range of motion and posture of the embodied robot are set.

[0064] The aforementioned interactive feature customization information includes voice information and motion information.

[0065] In this embodiment, after obtaining the interaction feature customization information, the interaction feature customization information is parsed to obtain sound information and motion information. Based on the sound information, the voice of the embodied robot is set; specifically, the voice of the embodied robot is set to be consistent with the voice represented by the sound information. Based on the motion information, the motion range and robot posture of the embodied robot are set.

[0066] Optionally, the interaction feature customization information includes text information, and the step of generating interaction features based on the interaction feature customization information includes: The text information is semantically segmented to obtain multiple text segments; each text segment represents a different semantic meaning. Text feature extraction is performed on the multiple text segments to obtain multiple text feature vectors; The multiple text feature vectors are aggregated to obtain an interaction feature embedding vector; The interaction features are embedded into a vector and input into a preset generative adversarial network to obtain the interaction features.

[0067] As mentioned above, the customized interaction features include text information. It should be noted that these interaction features include, but are not limited to, interaction methods such as actions, tone of voice, and facial expressions.

[0068] In this embodiment, semantic analysis algorithms can be used to perform semantic analysis on the text information input by the user, and text belonging to different semantic categories can be segmented into different paragraphs to obtain multiple text segments.

[0069] Text features are extracted from multiple text segments to obtain multiple text feature vectors, and these multiple text feature vectors are aggregated to obtain an interactive feature embedding vector.

[0070] It should be noted that this embodiment pre-sets a generative adversarial network, and trains the generative adversarial network using supervised learning. The training data is feature vectors, the output is interaction features, and the labels are positive labels representing the feature vectors and interaction features, and negative labels representing the feature vectors and interaction features that do not correspond.

[0071] In this embodiment, after obtaining the interaction feature embedding vector, the interaction feature embedding vector is input into a preset generative adversarial network to obtain the interaction features.

[0072] In this embodiment, based on the text information input by the user, an interaction feature embedding vector is generated, and then the interaction feature embedding vector is input into a preset generative adversarial network to obtain interaction features, thereby realizing the customized design of the interaction method of the embodied robot.

[0073] It should be understood that in other embodiments, users can customize the interactive features of the avatar robot and interact with it via touch or remote control.

[0074] Optionally, after aggregating the multiple text feature vectors to obtain the interaction feature embedding vector, the method further includes: The target 3D model is input into a pre-defined 3D convolutional neural network to obtain the shape feature vector; The shape feature vector is input into a preset generative adversarial network to obtain candidate features; The interactive features are obtained by adjusting the candidate features using the interactive feature embedding vector.

[0075] It should be noted that a three-dimensional convolutional neural network is pre-configured, which is used to convert three-dimensional vectors into one-dimensional feature vectors.

[0076] In this embodiment, after obtaining the target 3D model, the target 3D model is input into a preset 3D convolutional neural network to obtain shape feature vectors, and the above shape feature vectors are input into a preset generative adversarial network to obtain candidate features. Furthermore, the candidate features are adjusted by inter-feature embedding vectors to obtain inter-feature features.

[0077] In this embodiment, interactive features are obtained based on the shape feature vector representing the appearance and the interactive feature vector obtained based on the text information input by the user. These interactive features are adapted to the appearance of the robot and meet the user's customization needs.

[0078] Optionally, the method further includes: The target 3D model is input into a preset 3D convolutional neural network to obtain the shape feature vector; The shape feature vector is input into a preset generative adversarial network to obtain interaction features; By assembling the component blocks into the pre-set embodied robot torso, the target robot is generated; The part blocks are obtained by 3D printing the target 3D model, and the target robot operates the interaction mode represented by the interaction features.

[0079] In this embodiment, after obtaining the target 3D model, the target 3D model is input into a preset 3D convolutional neural network to obtain the shape feature vector, and the above shape feature vector is input into a preset generative adversarial network to obtain the interaction feature.

[0080] The target 3D model is 3D printed to obtain component blocks, which are then assembled into the torso of the embodied robot. After obtaining the interaction features, the interaction mode of the embodied robot is set to obtain the target robot.

[0081] In this embodiment, the shape feature vector is obtained based on the target 3D model, and then the shape feature vector is input into a preset generative adversarial network to obtain interaction features. Based on the appearance of the embodied robot, the interaction mode of the embodied robot is set.

[0082] This application also provides an apparatus for generating a hymenoid robot; for details, please refer to [link to specific examples]. Figure 2 , Figure 2 This is a schematic diagram of the structure of a body robot generation device provided in an embodiment of this application, as shown below. Figure 2 As shown, the embodied robot generation device 200 includes: The first acquisition module 201 is used to acquire the shape feature customization information input by the user, the shape feature customization information being used to characterize the external shape of the embodied robot; The determining module 202 is used to determine at least one part block and / or at least one component based on the shape feature customization information; The second acquisition module 203 is used to acquire user-inputted interaction feature customization information, which is used to characterize the interaction mode of the embodied robot. The first generation module 204 is used to generate interactive features based on the customized information of the interactive features; The second generation module 205 is used to generate a target robot by assembling at least one part block and / or at least one component into a preset body robot torso. Each component block corresponds to a part of the embodied robot. The components are obtained through 3D printing, and the target robot operates according to the interaction methods represented by the interaction features. Optionally, the determining module 202 is specifically used for: Based on the customized information of the aforementioned shape features, a first three-dimensional model is generated; The first 3D model is automatically topologically optimized and lightweighted to obtain the second 3D model. Based on the preset machine part regions, the second three-dimensional model is cut into at least one target three-dimensional model that corresponds one-to-one with the machine parts. Based on the at least one target 3D model, determine at least one part of the building blocks.

[0083] Optionally, the shape feature customization information includes any one of three-dimensional scan data, image data, and shape feature customization instructions; The first generation module 202 is further specifically used for: When the shape feature customization information is three-dimensional scan data, a model is created based on the three-dimensional scan data to obtain a first three-dimensional model; When the shape feature customization information is image data, a three-dimensional reconstruction is performed based on the image data to obtain a first three-dimensional model; When the shape feature customization information is a shape feature customization instruction, in response to the shape feature customization instruction, a first three-dimensional model corresponding to the shape feature customization instruction is selected from a preset model library.

[0084] Optionally, the customized interaction features include text information, voice information, and video information; The second generation module 204 is further specifically used for: Text features are extracted from the text information to obtain a text feature vector; The speech information is processed by a preset speech encoder to extract features and obtain an acoustic feature vector. The video information is used to extract features using pose estimation and facial expression recognition algorithms to obtain action feature vectors. Multimodal feature fusion is performed on the text feature vector, the acoustic feature vector, and the action feature vector to obtain an interaction feature embedding vector; The interaction features are embedded into a vector and input into a preset generative adversarial network to obtain the interaction features.

[0085] Optionally, the customized interaction feature information includes basic information, personality trait information, experience information, and skill information; The second generation module 204 is further specifically used for: The interaction feature customization information is analyzed to obtain basic information, personality trait information, experience information, and skill information; Based on the basic information, personality traits, experience, and skills, interaction features are generated through matching.

[0086] Optionally, the interaction feature customization information includes facial expression information; The second generation module 204 is further specifically used for: The interaction feature customization information is analyzed to obtain the facial expression information; Based on the facial expression information, the interaction method for the face area of ​​the avatar robot is set.

[0087] Optionally, the interaction feature customization information includes voice information and motion information; The second generation module 204 is further specifically used for: The interaction feature customization information is analyzed to obtain the sound information and the action information; Based on the sound information, the voice of the embodied robot is set; Based on the motion information, the range of motion and posture of the embodied robot are set.

[0088] For details, see Figure 3 This application also provides an electronic device.

[0089] The transceiver 302 is used to acquire shape feature customization information input by the user, and the shape feature customization information is used to characterize the external shape of the embodied robot. The processor 305 is used to determine at least one part block and / or at least one component based on the shape feature customization information; The transceiver 302 is used to acquire user-inputted interaction feature customization information, which is used to characterize the interaction mode of the embodied robot. The processor 305 is used to generate interactive features based on the customized information of the interactive features; A target robot is generated by assembling at least one part block and / or at least one component into a pre-defined embodied robot torso. Each component block corresponds to a machine part of the embodied robot. The components are obtained based on 3D printing, and the target robot operates the interaction mode represented by the interaction features.

[0090] exist Figure 3In this context, a bus architecture (represented by bus 301) is used. Bus 301 can include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 305 and memory represented by memory 306. Bus 301 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 303 provides an interface between bus 301 and transceiver 302. Transceiver 302 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 305 is transmitted over a wireless medium via antenna 330. Furthermore, antenna 303 also receives data and transmits data to processor 305.

[0091] Processor 305 manages bus 301 and general processing, and also provides various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. Memory 306 can be used to store data used by processor 305 during operation.

[0092] Optionally, the processor 305 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD).

[0093] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described embodiment of the embodied robot generation method and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0094] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0095] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0096] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for generating an embodied robot, characterized in that, The method includes: Obtain the shape feature customization information input by the user, which is used to characterize the external shape of the embodied robot; Based on the shape feature customization information, at least one part of the building block and / or at least one component is determined; Obtain user-inputted interaction feature customization information, which is used to characterize the interaction mode of the embodied robot; Based on the customized information of the interaction features, interaction features are generated; A target robot is generated by assembling at least one part block and / or at least one component into a pre-defined embodied robot torso. Each component block corresponds to a machine part of the embodied robot. The components are obtained based on 3D printing, and the target robot operates the interaction mode represented by the interaction features.

2. The method according to claim 1, characterized in that, Based on the aforementioned shape feature customization information, at least one component block is determined, including: Based on the customized information of the aforementioned shape features, a first three-dimensional model is generated; The first 3D model is automatically topologically optimized and lightweighted to obtain the second 3D model. Based on the preset machine part regions, the second three-dimensional model is cut into at least one target three-dimensional model that corresponds one-to-one with the machine parts. Based on the at least one target 3D model, determine at least one part of the building blocks.

3. The method according to claim 2, characterized in that, The shape feature customization information includes any one of three-dimensional scan data, image data, and shape feature customization instructions; The step of generating a first 3D model based on the customized information of the shape features includes: When the shape feature customization information is three-dimensional scan data, a model is created based on the three-dimensional scan data to obtain a first three-dimensional model; When the shape feature customization information is image data, a three-dimensional reconstruction is performed based on the image data to obtain a first three-dimensional model; When the shape feature customization information is a shape feature customization instruction, in response to the shape feature customization instruction, a first three-dimensional model corresponding to the shape feature customization instruction is selected from a preset model library.

4. The method according to claim 1, characterized in that, The customized interactive features include text information, voice information, and video information; The process of generating interaction features based on the customized information of the interaction features includes: Text features are extracted from the text information to obtain a text feature vector; The speech information is processed by a preset speech encoder to extract features and obtain an acoustic feature vector. The video information is used to extract features using pose estimation and facial expression recognition algorithms to obtain action feature vectors. Multimodal feature fusion is performed on the text feature vector, the acoustic feature vector, and the action feature vector to obtain an interaction feature embedding vector; The interaction features are embedded into a vector and input into a preset generative adversarial network to obtain the interaction features.

5. The method according to claim 1, characterized in that, The customized interaction feature information includes basic information, personality trait information, experience information, and skill information. The generation of interaction features based on the customized interaction feature information includes: The interaction feature customization information is analyzed to obtain basic information, personality trait information, experience information, and skill information; Based on the basic information, personality traits, experience, and skills, interaction features are generated through matching.

6. The method according to claim 1, characterized in that, The customized interaction feature information includes facial expression information, and the generation of interaction features based on the customized interaction feature information includes: The interaction feature customization information is analyzed to obtain the facial expression information; Based on the facial expression information, the interaction method for the face area of ​​the avatar robot is set.

7. The method according to claim 1, characterized in that, The customized interaction feature information includes voice information and motion information. The step of generating interaction features based on the customized interaction feature information includes: The interaction feature customization information is analyzed to obtain the sound information and the action information; Based on the sound information, the voice of the embodied robot is set; Based on the motion information, the range of motion and posture of the embodied robot are set.

8. A device for generating a embodied robot, characterized in that, The device includes: The first acquisition module is used to acquire the shape feature customization information input by the user, the shape feature customization information being used to characterize the external shape of the embodied robot; A determination module is used to determine at least one part of the building block and / or at least one component based on the shape feature customization information; The second acquisition module is used to acquire user-inputted interaction feature customization information, which is used to characterize the interaction mode of the embodied robot. The first generation module is used to generate interaction features based on the customized information of the interaction features; The second generation module is used to generate a target robot by assembling at least one part block and / or at least one component into a preset body robot torso. Each component block corresponds to a machine part of the embodied robot. The components are obtained based on 3D printing, and the target robot operates the interaction mode represented by the interaction features.

9. An electronic device, characterized in that, The electronic device includes a bus, a transceiver, an antenna, a bus interface, a processor, and a memory. The processor stores a computer program, which, when executed by the processor, implements the steps of the method for generating the embodied robot as described in any one of claims 1 to 7.

10. A embodied robot, characterized in that, The embodied robot is obtained using the embodied robot generation method as described in any one of claims 1 to 7.

Images