Bionic facial expression robot and a single-chip-based method for adapting to various refined facial expression control techniques

Abstract

This invention relates to a biomimetic facial expression robot and a method for controlling refined facial expressions based on a single-core adaptive module. The robot comprises a skeleton, a facial adaptation module, an expression driving module, a cortex, and a controller. The facial adaptation module includes multiple facial adaptation components, each with a universal interface, and these components are defined as a first adaptation component, a second adaptation component, and a third adaptation component. The expression driving module includes multiple identically configured single-core modules, each configured to have multi-degree-of-freedom actuation capabilities. The controller is configured to control the movement speed, stroke, and angle of the facial adaptation components to achieve refined biomimetic facial expression representation on the cortex. The expression driving module utilizes multiple identically configured single-core modules, reducing research and manufacturing costs. By controlling the movement speed, stroke, angle, and their changing trends of each adaptation component, the realism of the expressions is improved.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent robots, and in particular to a biomimetic facial expression robot and a method for controlling various refined facial expressions based on a single core. Background Technology

[0002] With the deep integration of artificial intelligence and robotics, biomimetic robots capable of emotional expression have become a cutting-edge research direction in the field of human-computer interaction. However, existing facial expression robot systems still face many challenges in achieving natural, vivid, and multi-part coordinated emotional interaction, such as excessively high programming complexity, long debugging cycles, slow system response, and a lack of intuitive interaction methods, which seriously restrict their popularization and application depth.

[0003] In the prior art, patent document CN115946150A discloses a head structure and robot with facial expressions, which adopts a combination of skeleton, epidermis and expression modules. By adjusting the position of the modules, it can adapt to different epidermis, which improves the assembly versatility to a certain extent. However, this solution still adopts a multi-mechanism independent drive form and does not achieve a unified configuration of a single core universal drive. It cannot make a single core compatible with multiple face shapes and appearances, resulting in a single product appearance, high R&D costs and low mass production efficiency. At the same time, it does not carry out partitioned and differentiated design of the epidermis, nor does it achieve integrated drive for wrinkles and dimples, resulting in insufficient reproduction of facial expression details and stiff movements.

[0004] Patent document CN120941428B proposes a multi-part collaborative voice AI programming facial expression robot control system, which can control facial actuators to achieve facial expressions through commands. However, this solution only focuses on control logic and external interaction, and does not solve the problem of universal compatibility between the core mechanism and facial components, making it impossible to achieve rapid switching between various detailed appearances; its control method is relatively simple, and it has not established a quantitative relationship between movement speed, stroke, angle, and speed change trend, resulting in stiff transitions in facial expressions, unclear distinction of emotions, and difficulty in reproducing realistic human facial expressions.

[0005] In summary, existing bionic facial expression robot technologies generally suffer from drawbacks such as poor versatility of the core mechanism, difficulty in adapting a single core mechanism to diverse appearances, unreasonable skin structure design, easily stiff and damaged expressions, limited functionality of the facial expression mechanism, inability to achieve precise bionic imitation of wrinkles and dimples in an integrated manner, crude expression control logic, disjointed movements, and unclear emotion differentiation. They are unable to simultaneously achieve low cost, high versatility, strong realism, and natural movements, and thus cannot meet the usage requirements of high-end bionic interaction scenarios. Summary of the Invention

[0006] The first objective of this invention is to provide a bionic facial expression robot, which aims to solve the technical problems of existing facial expression robots, such as poor versatility of the core mechanism, high cost, stiff bionic expressions, and limited facial expression structure and function, which cannot achieve precise bionic imitation.

[0007] To solve the above technical problems, a bionic facial expression robot is provided, comprising:

[0008] The main skeletal structure includes a first facial region, a second facial region, and a third facial region;

[0009] A facial adaptation module includes multiple facial adaptation components. Each facial adaptation component has a universal interface. The multiple facial adaptation components are defined as a first adaptation component, a second adaptation component, and a third adaptation component. The first adaptation component is located in the first facial region, the second adaptation component is located in the second facial region, and the third adaptation component is located in the third facial region. The first adaptation component is used for forming regular facial expressions, the second adaptation component is used for forming wrinkles, and the third adaptation component is used for forming wrinkles or dimples.

[0010] An expression driving module is installed on the skeleton body and connected to the face adaptation module. The expression driving module includes multiple single core bodies with the same configuration. The single core body is configured to have multi-degree-of-freedom driving capability and has a standardized interface that cooperates with the general interface to realize the detachable connection between the face adaptation component and the expression driving module.

[0011] The cortex is connected to the facial adaptation module and is located on the side opposite to the expression driving module;

[0012] The controller is electrically connected to the expression driving module and is configured to achieve refined bionic expression presentation of the skin by controlling the changing trends of the movement speed, movement stroke, and movement angle of the facial adaptation component.

[0013] Furthermore, the first adapter component includes a first link and a first slider. The first link is connected between the single-core body and the first slider. The first slider is fixedly installed on the dermis. The controller is configured to drive the first slider to move in a direction perpendicular to the length of the first link to tighten the dermis and form a regular expression.

[0014] Furthermore, the second adapter component includes a second link and a second slider. The second link is connected between the single-core body and the second slider. The second slider is fixedly installed on the skin. The skin has a first notch distributed at intervals on the side near the face adapter module. The controller is configured to drive the first notch to deform so that the skin forms wrinkles.

[0015] Furthermore, the third adapter component includes a third link and a third slider. The third link connects the single-mechanism body and the third slider. The third slider is fixedly installed on the skin. The skin is provided with an extended protrusion that cooperates with the third slider. The skin is provided with a second notch spaced apart on the side near the facial adapter module. The controller is configured to: within a first stroke range, drive the third adapter component to swing to generate biomimetic wrinkles in the skin; within a second stroke range, drive the third adapter component to swing to generate biomimetic dimples in the skin; and the first stroke range is smaller than the second stroke range.

[0016] Furthermore, the skin layer is provided with spaced first deformation cavities, which are used to deform the skin layer. The third slider has a hooked surface, and a mating surface that mates with the hooked surface is formed on the skin layer. The skeleton body is provided with a receiving groove for accommodating part of the third slider.

[0017] Furthermore, the multiple facial adaptation components are also defined as a fourth adaptation component, which includes a fourth link and a fourth slider. The fourth link is connected between the single-mechanism main body and the fourth slider, and the fourth slider is slidably mounted on the skin. The controller is configured such that: the single-mechanism main body drives the fourth slider to swing along a first preset angle to generate biomimetic wrinkles on the skin; and the single-mechanism main body drives the fourth slider to slide along a second preset angle to generate biomimetic dimples on the skin.

[0018] Furthermore, the main skeleton divides the human face into multiple functional areas with different degrees of deformation, and the cortex is set with a gradient thickness corresponding to each functional area; the first thickness is configured to correspond to the cortex thickness of areas with large facial deformation and frequent movement, which facilitates cortex stretching and wrinkle deformation; the second thickness is configured to correspond to the cortex thickness of areas with small facial deformation and mainly provide contour support, which is used to maintain the overall stability of the facial shape; wherein, the first thickness is smaller than the second thickness, and there is a continuous and smooth transition between areas of different thicknesses.

[0019] Furthermore, the skin layer and the facial adapter component are detachable and replaceable structures, which can simulate the facial expressions and skin morphology of children, youths, middle-aged people and the elderly at different ages by replacing the skin layer with different materials, thicknesses and textures and the corresponding facial adapter components.

[0020] The slider of the facial adaptation component is configured to make surface contact with the skin. By controlling the size of the contact area between the slider and the skin, the force distribution and deformation degree of the skin can be adjusted to form different bionic expressions.

[0021] The second objective of this invention is to provide a single-core adaptive method for controlling diverse and refined facial expressions, applied to the aforementioned bionic facial robot. This method uses a unified single-core drive to control the movement speed, stroke, and angle of the single-core body corresponding to each facial region, thereby outputting bionic movements corresponding to different expressions and emotions. The method sets differentiated control parameters for different expression-forming structures, wherein the first adaptation component uses movement angle and stroke as the main control parameters; the second adaptation component uses stroke and speed as the main control parameters; and the third adaptation component uses local movement angle and gradual stroke as the main control parameters.

[0022] Furthermore, the following overall magnitude relationship exists between each facial expression / movement and its motion parameters and velocity change trends:

[0023] Speed ​​of movement: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calmness;

[0024] Exercise sequence: Laughter > Surprise > Dimples > Smile > Sadness > Wrinkles > Calmness;

[0025] From a movement perspective: laughter > surprise > dimples > smile > sadness > wrinkles > calmness;

[0026] The degree of drastic change in speed: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calmness.

[0027] Implementing the embodiments of the present invention will have the following beneficial effects:

[0028] In this embodiment, the bionic facial robot, firstly, divides the skeletal main body's face into a first facial region, a second facial region, and a third facial region, and respectively sets up a first adapter component, a second adapter component, and a third adapter component, allowing for independent control of three types of facial expressions: regular expressions, wrinkles, and dimples / wrinkles, without interference between them, resulting in richer facial expressions and a more realistic bionic effect. Secondly, the expression driving module uses multiple sets of single-core main bodies with identical configurations, and uses universal and standardized interfaces to achieve detachable connections. Different facial adapter components can be quickly replaced without changing the core, achieving compatibility of one core with multiple appearances, reducing R&D and manufacturing costs, and simplifying assembly and maintenance. Thirdly, the controller adjusts the movement speed, movement stroke, movement angle, and change trend of each adapter component to make the facial movements present a gradual transition and smooth start and stop effect, avoiding mechanical stiffness and improving the realism of the expressions and the interactive experience. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram showing the division of various parts of the bionic facial expression robot described in Embodiment 1 of the present invention;

[0031] Figure 2 This is a schematic diagram of the structure of the first adapter component described in Embodiment 1 of the present invention;

[0032] Figure 3 This is a schematic diagram of the structure of the second adapter component described in Embodiment 1 of the present invention;

[0033] Figure 4 This is a schematic diagram of the structure of the third adapter component described in Embodiment 1 of the present invention;

[0034] Figure 5 This is a schematic diagram of the structure of the fourth adapter component described in Embodiment 2 of the present invention;

[0035] Figure 6 This is a schematic diagram of the structure of the fifth adapter component described in Embodiment 3 of the present invention.

[0036] Among them: 100, bionic facial expression robot; 110, skeleton body; 111, receiving slot; 120, facial adaptation module; 121, first adaptation component; 1211, first link; 1212, first slider; 122, second adaptation component; 1221, second link; 1222, second slider; 123, third adaptation component; 1231, third link; 1232, third slider; 1233, curved surface; 124, the... Four adapter components; 1241, fourth link; 1242, fourth slider; 125, fifth adapter component; 1251, fifth link; 1252, fifth slider; 1252A, opening slot; 130, expression driving module; 131, single-core body; 140, skin layer; 141, first notch; 142, second notch; 143, extension protrusion; 1431, mating surface; 144, first deformation cavity; 145, sliding groove. Detailed Implementation

[0037] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0038] It should be noted that when a component is said to be "fixed to" another component, it can be directly attached to the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0040] Example 1:

[0041] Please refer to Figures 1-4This invention provides a bionic facial expression robot 100, including a skeleton body 110, a facial adaptation module 120, an expression driving module 130, a cortex 140, and a controller. The skeleton body 110 includes a first facial region, a second facial region, and a third facial region. The facial adaptation module 120 includes multiple facial adaptation components, each with a universal interface. These components are defined as a first adaptation component 121, a second adaptation component 122, and a third adaptation component 123. The first adaptation component 121 is located in the first facial region, the second adaptation component 122 is located in the second facial region, and the third adaptation component 123 is located in the third facial region. The first adaptation component 121 is used for forming regular facial expressions, and the second adaptation component 122 is used for forming wrinkles or dimples. An expression-driven module 130 is mounted on the skeleton body 110 and connected to the face adaptation module 120. The expression-driven module 130 includes multiple identically configured single-core bodies 131. Each single-core body 131 is configured to have multi-degree-of-freedom driving capabilities and a standardized interface that mates with a universal interface to achieve a detachable connection between the face adaptation components and the expression-driven module 130. A cortex 140 is connected to the face adaptation module 120 and positioned on the side opposite to the expression-driven module 130. A controller is electrically connected to the expression-driven module 130 and is configured to achieve refined bionic expression representation of the cortex 140 by controlling the movement speed, movement stroke, and movement angle changes of the face adaptation components. For example, the first facial region includes the eyebrow region, eye region, and mouth region; the second facial region includes the corners of the eyes, forehead region, and sides of the nose; and the third facial region includes the perioral region, cheek region, and sides of the corners of the mouth. The universal interface is a unified assembly structure set on each facial adapter component, and the standardized interface is a unified drive connection structure set on the output end of each single-core body 131. The universal interface and the standardized interface adopt the same design in terms of shape, connection orientation, and mating tolerance, and can achieve a matching connection between the single-core body 131 and the facial adapter component. For example, the universal interface and the standardized interface can be a detachable snap-fit ​​structure that can be used together. The single-core body 131 is configured to have multi-degree-of-freedom drive capability, which means that the single-core body 131 can drive the facial adapter component to perform compound movements along multiple directions and multiple angles to achieve multi-dimensional deformation of the skin 140 and complete various bionic facial expressions.

[0042] Furthermore, the facial adaptation module 120 and the expression driving module 130 are detachably connected. Users can directly disassemble and replace skin layers 140 with different structures, thicknesses, and textures and corresponding facial adaptation components according to their needs. Without replacing the main body 131 and the skeleton body 110, users can quickly switch facial appearance and expression features, thereby better simulating the facial expression features of different age groups such as children, youth, middle-aged, and elderly. Specifically, for children, a thinner, more uniformly thicker, finer, and more deformably responsive dermis 140 is used, paired with facial adaptation components that have a smaller range of motion and gentler movements to produce childlike, light, and soft facial expressions. For young adults, a dermis 140 with moderate elasticity and clear contours is used, paired with facial adaptation components that have a normal range of motion and movement to produce natural, full, and energetic facial expressions. For middle-aged and elderly individuals, a dermis 140 with gradually varying thickness in certain areas, easily forming wrinkles and lines, is used, paired with facial adaptation components that can achieve large range of motion, deep lines, and a sense of relaxation to produce refined bionic expressions with obvious wrinkles, a calm demeanor, and age-appropriate features. Through the combination of the aforementioned detachable and replaceable structure with differentiated dermis 140 and facial adaptation components, the same bionic expression robot 100 can output highly realistic and diverse facial expressions across age groups, which helps to improve the bionic realism and scene adaptability.

[0043] In this embodiment, the bionic facial robot 100, firstly, divides the face of the skeleton body 110 into a first facial region, a second facial region, and a third facial region, and respectively sets a first adapter component 121, a second adapter component 122, and a third adapter component 123, so that the three types of facial expressions—regular expressions, wrinkles, and dimples / wrinkles—can be controlled independently without interference, resulting in richer facial expressions and a more realistic bionic effect. Secondly, the facial expression driving module 130 uses multiple sets of single-core bodies 131 with the same configuration, which are connected detachably with universal and standardized interfaces. Different facial adapter components can be quickly replaced without changing the core, achieving compatibility of one core with multiple appearances, reducing R&D and manufacturing costs, and simplifying assembly and maintenance. Thirdly, the controller adjusts the movement speed, movement stroke, movement angle, and change trend of each adapter component to make the facial expressions present a gradual transition and smooth start and stop effect, avoiding mechanical stiffness and improving the realism of the expressions and the interactive experience.

[0044] Please refer to Figure 2In one possible implementation, the first adapter component 121 includes a first link 1211 and a first slider 1212. The first link 1211 connects the single-core body 131 and the first slider 1212. The first slider 1212 is fixedly mounted on the skin layer 140. The controller is configured to drive the first slider 1212 to move in a direction perpendicular to the length of the first link 1211 to tighten the skin layer 140 and form a regular expression. Exemplarily, one end of the first link 1211 is connected to the output end of the single-core body 131, and the other end is connected to the first slider 1212, so that the driving force of the single-core body 131 can be smoothly transmitted to the first slider 1212 through the first link 1211. The first slider 1212 is fixedly mounted on the inner side of the skin layer 140, maintaining a reliable connection with the skin layer 140. By controlling the movement of the single-core main body 131, the first connecting rod 1211 is driven to move, which in turn drives the first slider 1212 to swing along the length direction perpendicular to the first connecting rod 1211. During the movement, the first slider 1212 tightens or relaxes the skin layer 140, thereby causing the skin layer 140 to produce corresponding stretching, stretching or contraction deformation to form conventional bionic expressions such as smiling, opening eyes, closing eyes, opening mouth, closing mouth, raising eyebrows, and frowning, ensuring that conventional facial expressions are natural, clear and stable.

[0045] Please refer to Figure 3 In one possible implementation, the second adapter component 122 includes a second connecting rod 1221 and a second slider 1222. The second connecting rod 1221 is connected between the single core body 131 and the second slider 1222. The second slider 1222 is fixedly installed on the skin layer 140. The skin layer 140 has a first notch 141 spaced apart on the side near the face adapter module 120. The controller is configured to drive the first notch 141 to deform, thereby forming wrinkles on the skin layer 140. Exemplarily, the second connecting rod 1221 is connected between the single core body 131 and the second slider 1222, and the second slider 1222 is fixedly installed on the inner side of the skin layer 140. The skin layer 140 has a first notch 141 spaced apart along a preset wrinkle direction on the side near the face adapter module 120. The first notch 141 reduces the local stiffness of the skin layer 140, making it easier for the skin layer 140 to form a uniform and natural wrinkle shape when subjected to force. The single-core main body 131 drives the second connecting rod 1221 and the second slider 1222 to move, causing the second slider 1222 to cause the skin layer 140 to be stretched and contracted in a directional manner, thereby causing the area where the first notch 141 is located to undergo directional deformation and closure, forming biomimetic wrinkles such as crow's feet, forehead wrinkles or nasolabial folds on the outer surface of the skin layer 140. The depth and shape of the wrinkles are adjusted by the movement stroke and speed of the second slider 1222 to improve the realism of facial expression details.

[0046] Please refer to Figure 4In one possible implementation, the third adapter component 123 includes a third link 1231 and a third slider 1232. The third link 1231 connects the single core body 131 and the third slider 1232. The third slider 1232 is fixedly mounted on the skin layer 140. The skin layer 140 is provided with an extension protrusion 143 that cooperates with the third slider 1232. The side of the skin layer 140 near the face adapter module 120 is provided with a second notch 142 spaced apart. The controller is configured to: within a first stroke range, drive the third adapter component 123 to swing to generate biomimetic wrinkles in the skin layer 140; within a second stroke range, drive the third adapter component 123 to swing to generate biomimetic dimples in the skin layer 140, and the first stroke range is smaller than the second stroke range. Exemplarily, the third link 1231 connects the single core body 131 and the third slider 1232, and the third slider 1232 is fixedly mounted on the inner side of the skin layer 140. The facial adaptation module 120 is provided with an extension protrusion 143 that cooperates with the third slider 1232. The skin layer 140 has a second notch 142 spaced apart on the side closest to the facial adaptation module 120. By controlling the single-core main body 131 to drive the third connecting rod 1231 and the third slider 1232 to swing, within a first stroke range, the third slider 1232 causes slight deformation of the skin layer 140, forming natural biomimetic wrinkles in the area of ​​the second notch 142. Within a second stroke range, the swing amplitude of the third slider 1232 increases, cooperating with the extension protrusion 143 to directionally pull the skin layer 140, forming obvious biomimetic dimples. The first stroke range is smaller than the second stroke range, allowing the same component to achieve biomimetic representation of wrinkles and dimples at different strokes. Furthermore, the skin layer 140 is provided with spaced first deformation cavities 144 to assist the skin layer 140 in producing folds and depressions more smoothly under force. The third slider 1232 has a hooked surface 1233, and the skin layer 140 forms a mating surface 1431 that adapts to the hooked surface 1233, ensuring reliable connection and stable transmission between the third slider 1232 and the skin layer 140. The skeleton body 110 has a receiving groove 111 to accommodate part of the third slider 1232, avoiding motion interference and improving structural compactness and smoothness of movement. The first stroke range is the small-amplitude swing stroke range of the third slider 1232, used to drive the skin layer 140 to produce slight and shallow deformation, so that the area of ​​the second notch 142 closes to form natural and delicate bionic wrinkles without producing obvious depressions. The second stroke range is the large-amplitude swing stroke range of the third slider 1232, used to drive the skin layer 140 to produce significant and deep traction and depression deformation, which, together with the extended protrusion 143, forms a bionic dimple with clear contours and a strong three-dimensional effect.

[0047] Please refer to Figure 4In one possible implementation, the skin layer 140 is provided with spaced-apart first deformation cavities 144 for deforming the skin layer 140. The third slider 1232 has a hooked surface 1233, and the skin layer 140 has a mating surface 1431 that mates with the hooked surface 1233. The skeleton body 110 is provided with a receiving groove 111 for accommodating part of the third slider 1232. Exemplarily, the hooked surface 1233 of the third slider 1232 is a concave arc-shaped surface facing the skin layer 140, with a smoothly transitioning arc-shaped concave shape, and the curvature of the surface matches the contour of the dimple area on the human face. The mating surface 1431 formed on the inner side of the skin layer 140 is a convex arc-shaped surface with the same curvature as the hooked surface 1233 and fits against each other. The hooked surface 1233 and the mating surface 1431 form a mating structure of concave-convex interlocking and surface contact. The functions of the curved surface 1233 and the mating surface 1431 are: to increase the contact area between the third slider 1232 and the skin layer 140, improve the connection firmness and transmission stability, and avoid local stress concentration that could cause the skin layer 140 to tear; at the same time, through the guiding and pulling effect of the curved surface, the skin layer 140 forms rounded, natural, and angleless bionic dimples and bionic wrinkles during the deformation process, thereby enhancing the realism and softness of the facial expression.

[0048] Please refer to Figure 1In one possible implementation, the skeleton body 110 divides the human face into multiple functional areas with different degrees of deformation. The dermis 140 is configured with a gradient thickness corresponding to each functional area. A first thickness is configured for areas with large facial deformation and frequent movement, facilitating stretching and wrinkling deformation of the dermis 140. A second thickness is configured for areas with small facial deformation and primarily providing contour support, maintaining overall facial shape stability. The first thickness is less than the second thickness, and there is a continuous and smooth transition between areas of different thicknesses. For example, the skeleton body 110 divides the human face into multiple functional areas with different degrees of deformation. The dermis 140 is integrally molded with a gradient thickness according to the movement frequency, deformation amplitude, and support requirements of each area. The thickness of different areas changes continuously without steps or breaks, ensuring the overall consistency and smooth appearance of the dermis 140. The first thickness corresponds to high-deformation areas of the face, characterized by large facial deformation, frequent movement, and the need for frequent stretching and wrinkling. In these areas, the dermal layer 140 is thinner, softer, and has lower stretching resistance. Driven by the facial adaptation components, it more easily produces subtle wrinkles, depressions, and stretching deformations, facilitating the realistic portrayal of dynamic expressions such as smiles, laughter, wrinkles, and dimples. This avoids stiff expressions and inaccurate deformation caused by an overly rigid dermal layer 140. By ensuring the first thickness is smaller than the second thickness and a smooth, continuous transition between different thickness areas, the dermal layer 140 becomes more sensitive in easily deformable areas and more stable in supporting areas, balancing facial realism, structural stability, and lifespan. This enhances the overall interactive experience and durability of the bionic facial robot 100.

[0049] Please refer to Figure 1In one possible implementation, the skeleton body 110 divides the face into high-deformation areas, medium-deformation areas, and low-deformation areas according to deformation intensity. The high-deformation areas include the perioral area, the corners of the eyes, and the area beside the nose, used for smiling, laughing, squinting, and wrinkle formation. The medium-deformation areas include the eyebrow area and the cheek area, used for raising eyebrows, creating dimples, and slight deformation. The low-deformation areas include the central forehead area, the bridge of the nose area, and the jawline area, used to support the facial contour. Exemplarily, the skeleton body 110 divides the face into high-deformation areas, medium-deformation areas, and low-deformation areas according to deformation intensity. The high-deformation areas include the perioral area, the corners of the eyes, and the area beside the nose, used for smiling, laughing, squinting, and wrinkle formation. The medium-deformation areas include the eyebrow area and the cheek area, used for raising eyebrows, creating dimples, and slight deformation. The low-deformation areas include the central forehead area, the bridge of the nose area, and the jawline area, used to support the facial contour. The high-deformation areas include the perioral region, the corners of the eyes, and the area beside the nose. These areas are the core expression areas where human facial muscles move most frequently, deform the most, and form the most dense wrinkles. They are mainly used to complete high-frequency, large-amplitude movements such as smiling, laughing, squinting, opening and closing the mouth, and the formation of biomimetic wrinkles. By coordinating with the 140-gradient thickness of the dermis, this area becomes sensitive to deformation and produces natural wrinkles, ensuring the vividness and fidelity of the core expressions. The medium-deformation areas include the eyebrow area and the cheek area. These areas have a moderate frequency of movement and deformation amplitude. They are mainly used to complete medium-level facial expressions such as raising eyebrows, slightly lifting the cheeks, and the formation of biomimetic dimples. They balance expressiveness and structural stability, making facial expressions richer and more layered. The low-deformation area includes the central forehead area, the bridge of the nose area, and the jawline area. This area does not participate in facial changes and mainly plays the role of supporting the overall facial contour, shaping the form, and stabilizing the structure. It is used to maintain the three-dimensional regularity of the face, prevent facial collapse or deformation, and ensure that the robot's face maintains a stable and beautiful appearance during long-term driving and use.

[0050] The second objective of this invention is to provide a single-core adaptive method for controlling diverse and refined facial expressions, applied to the aforementioned bionic facial expression robot 100. The method drives a unified single-core main body 131, controlling the movement speed, stroke, and angle of the single-core main body 131 corresponding to each facial region to output bionic movements corresponding to different expressions and emotions. The method sets differentiated control parameters for different expression-forming structures. Specifically, the first adaptation component 121 uses movement angle and stroke as the main control parameters; the second adaptation component 122 uses stroke and speed as the main control parameters; and the third adaptation component 123 uses local movement angle and gradual stroke as the main control parameters. For example, the first adaptation component 121 is used to form basic expressions such as smiling, opening eyes, and closing the mouth. The movements are mainly directional traction and overall extension, requiring high precision in movement amplitude and position. By using movement angle and stroke as core control parameters, the stretching amplitude and position of the cortex 140 can be precisely controlled, ensuring that the regular facial expression movements are standard, stable, and without deviation, achieving clear and controllable basic expressions. The second adapter component 122 is used to create fine facial expressions such as wrinkles. The depth, density, and naturalness of the wrinkles directly depend on the deformation speed and stretching stroke. By using the movement stroke and speed as core control parameters, the speed and depth of wrinkle formation can be precisely adjusted, allowing the wrinkles to present a gradual effect from shallow to deep and from sparse to dense, avoiding stiff and abrupt wrinkles and enhancing the biomimetic realism. The third adapter component 123 is used to alternately create wrinkles and dimples on the same structure, requiring the switching between the two forms to be completed in different stroke ranges. By using the local motion angle and gradual stroke as core control parameters, delicate wrinkles can be achieved with a small stroke, while deep dimples can be achieved with a large stroke, making the movement of the same component smooth and the form distinct, balancing structural simplicity and expressive richness.

[0051] The dermal layer 140 and the facial adapter components are detachable and replaceable. By replacing the dermal layer 140 with different materials, thicknesses, and textures, and the corresponding facial adapter components, facial expressions and skin morphology at different ages—children, youth, middle-aged, and elderly—can be simulated. For example, for children, a thinner, more elastic, finer-textured, and wrinkle-free dermal layer 140, combined with a facial adapter component with short stroke and low force, results in a sensitive and gentle facial expression, presenting a youthful, light, and wrinkle-free look. For youth, a moderately thicker, more elastic, and firm dermal layer 140, combined with a facial adapter component with standard motion parameters, achieves a natural, full, and energetic basic expression with minimal wrinkles and a youthful appearance. For middle-aged individuals, a dermal layer 140 with gradually varying thickness and prone to wrinkling in certain areas, combined with a facial adapter component driven by medium stroke and medium force, can create natural fine lines, mild nasolabial folds, and crow's feet, resulting in a composed expression that reflects their age. Targeting the elderly, the product utilizes a dermal layer 140 with high gradient thickness, high relaxation, and dense preset texture, combined with facial adaptation components driven by large stroke and deep deformation, to create deep wrinkles, obvious sagging contours, and static lines, highly replicating the facial expressions of the elderly.

[0052] The slider of the facial adapter component is configured for surface contact with the skin layer 140. By controlling the size of the contact area between the slider and the skin layer 140, the force distribution and deformation degree of the skin layer 140 can be adjusted to form bionic expressions of different shapes. For example, the slider of the facial adapter component uses surface contact with the skin layer 140, rather than point or line contact, so that the driving force applied by the slider can be evenly transmitted to the skin layer 140 in a surface manner, avoiding local stress concentration that could cause tearing, deformation, or jamming of the skin layer 140. By controlling the size of the contact area between the slider and the skin layer 140, the force distribution and deformation amplitude of the skin layer 140 can be precisely adjusted, thereby forming bionic expressions of different levels of detail and shapes. The larger contact area allows the cortex 140 to experience a wider range of force and more uniform stress, resulting in softer, fuller, and more natural deformation. This is suitable for driving the cortex 140 to produce large-area, large-radius, and rounded facial expressions, such as smiling, laughing, cheek bulging, and biomimetic dimples, ensuring natural facial contours without sharp angles and more closely resembling real muscle movements. The smaller contact area allows the cortex 140 to experience more concentrated force and more sensitive local deformation, forming obvious wrinkles and depressions with less driving force. This is suitable for driving the cortex 140 to produce fine, deep, and localized wrinkle deformation, such as crow's feet, forehead wrinkles, and nasolabial folds, improving wrinkle clarity and detail realism.

[0053] In one possible implementation, the following overall magnitude relationship exists between each facial expression and motion parameter, and the trend of speed change:

[0054] Speed ​​of movement: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calmness;

[0055] Exercise sequence: Laughter > Surprise > Dimples > Smile > Sadness > Wrinkles > Calmness;

[0056] From a movement perspective: laughter > surprise > dimples > smile > sadness > wrinkles > calmness;

[0057] The intensity of speed changes is as follows: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calm. The range of motion directly determines the depth and amplitude of deformation in cortical area 140. Laughter and surprise require significant stretching of cortical area 140, resulting in the largest range of motion; dimples require a large range of motion to drive the slider's swing, creating a noticeable indentation in cortical area 140, thus the range of motion is greater than that of a smile; a smile involves a medium range of motion; sadness involves a smaller range of motion; wrinkles require only a very small range of motion to close the gap in cortical area 140, resulting in the smallest range of motion; the calm state has no motion, and the range of motion is zero. The angle of motion corresponds to the amplitude of the slider's swing, directly determining the type of facial expression change. Laughter and surprise involve the largest deflection amplitude and the largest swing angle; dimples require a relatively large swing angle to create a three-dimensional indentation, thus the angle is greater than that of a smile; a smile involves a normal swing angle; sadness has a smaller angle; wrinkles require only a tiny angle to trigger wrinkle formation, resulting in the smallest angle; the calm state has no angle change. The speed of motion matches the rhythmic requirements of the facial expression. Laughter and surprise are explosive movements, with the slider swinging the fastest; smiles and dimples are gentle expressions, with similar and moderate speeds; sadness is a smooth movement with a slower speed; wrinkles are fine, static lines, requiring only low-speed micro-movements, resulting in the lowest speed; calmness has no movement speed. The trend of speed change corresponds to the smoothness of the expression's start and stop. Laughter and surprise start rapidly with a strong rhythm and the most dramatic changes; smiles and dimples start and stop smoothly with a gentle transition and consistent degree of change; sadness gradually softens and has a lower degree of change; wrinkles move subtly and steadily with a weak degree of change; calmness has no speed change and the trend is the smoothest.

[0058] The innovation of this solution lies in its use of a single-core main body 131 with identical configuration and a detachable connection structure featuring universal and standardized interfaces. This allows one core to be compatible with multiple facial adapter components and the skin layer 140, enabling rapid switching of appearance and expression without replacing the skeleton or drive. This significantly reduces R&D, production, and maintenance costs, while improving assembly efficiency and scalability. The skin layer 140 and facial adapter components employ a detachable and replaceable structure, matching different skin types and shapes for children, youth, middle-aged, and elderly individuals. This allows the same robot to output highly realistic expressions across different age groups, significantly enhancing scene adaptability. Through the same linkage slider structure of the third adapter component 123, wrinkles are output with a small stroke and small angle, while dimples are output with a large stroke and large angle. This achieves integrated driving and one-click switching between two fine expressions: wrinkles and dimples. This breaks through the single-function limitations of traditional mechanisms, resulting in richer expression layers and a more streamlined structure. The face is divided into high, medium, and low deformation zones. The dermal layer 140 employs a gradient thickness design: the high deformation zone is thinner, allowing for easy stretching and wrinkling, while the low deformation zone is thicker, providing support and stability. Combined with notches and deformation cavities, this results in natural, non-stiff expressions and stable, non-collapsed contours, balancing realism and durability. A surface contact mechanism is used, adjusting the stress distribution on the dermal layer 140 by controlling the contact area: large contact areas create soft, full deformations like dimples and smiles; small contact areas create fine, deep lines like wrinkles, preventing stress concentration and tearing of the dermal layer 140, thus improving movement stability and facial realism. Differentiated master control parameters are set for different adapter components, and movement speed, stroke, angle, and speed change trends are quantified and graded, resulting in smooth start and stop of expressions, natural transitions, and clear emotional distinctions. This completely eliminates the problems of mechanical stiffness and abrupt movements, achieving highly human-like emotional interaction.

[0059] Example 2:

[0060] Please refer to Figure 5 In this embodiment, the fourth adapter component 124 is an alternative to the third adapter component 123 in the first embodiment. Compared with the first embodiment, the difference lies in the structure and the way the dimples are generated.

[0061] Please refer to Figure 5In one possible implementation, the plurality of facial adaptation components are further defined as a fourth adaptation component 124. The fourth adaptation component 124 includes a fourth link 1241 and a fourth slider 1242. The fourth link 1241 is connected between the single-core body 131 and the fourth slider 1242. The fourth slider 1242 is slidably mounted on the skin layer 140. The controller is configured to: drive the single-core body 131 to swing the fourth slider 1242 along a first preset angle to generate biomimetic wrinkles on the skin layer 140; and drive the single-core body 131 to slide the fourth slider 1242 along a second preset angle to generate biomimetic dimples on the skin layer 140. Exemplarily, the fourth link 1241 is connected between the single-core body 131 and the fourth slider 1242, and the fourth slider 1242 is slidably mounted on the inner side of the skin layer 140. The fourth slider 1242 is driven by the single-core main body 131 to swing along a first preset angle, causing local wrinkle deformation in the skin layer 140 to form biomimetic wrinkles. The fourth slider 1242 is driven by the single-core main body 131 to slide along a second preset angle, directionally stretching and concave shaping the skin layer 140 to form biomimetic dimples. By switching between the two motion modes of swinging and sliding, the fourth adapter component 124 can independently output both wrinkles and dimples, further enriching the details of expressions and the versatility of the structure. A sliding groove 145 is provided on the skin layer 140. The fourth slider 1242 is a cylinder, and the shape of the sliding groove 145 is the same as the shape of the fourth slider 1242. The fourth slider 1242 is slidably installed in the sliding groove 145. In this embodiment, the first preset angle is defined as 0 degrees, and the second preset angle is 90 degrees. When the fourth slider 1242 moves toward the first preset angle, there is no relative displacement between the fourth slider 1242 and the sliding groove 145. When the fourth slider 1242 moves toward the second preset angle, the fourth slider 1242 slides in the sliding groove 145. That is to say, the sliding groove 145 extends in the direction of the second preset angle.

[0062] Apart from the differences mentioned above, the structure of the bionic facial expression robot 100 and its components provided in this embodiment can be optimized with reference to Embodiment 1, and will not be described in detail here.

[0063] Example 3:

[0064] This embodiment provides a method for forming dimples, the specific method is as follows:

[0065] Please refer to Figure 6The fifth adapter component 125 includes a fifth link 1251 and a fifth slider 1252. The fifth link 1251 connects the single mechanism body 131 and the fifth slider 1252. The fifth slider 1252 is disposed in non-contact with the leather layer 140. The fifth slider 1252 has an opening groove 1252A. The opening groove 1252A has an opening on the side facing the leather layer 140. In the direction from the leather layer 140 to the single mechanism body 131, the diameter of the opening groove 1252A gradually decreases. During operation, the single-core main body 131 drives the fifth connecting rod 1251 to perform high-speed reciprocating swing motion, and simultaneously drives the fifth slider 1252 to swing back and forth rapidly along a preset trajectory; the high-speed swinging fifth slider 1252 drives the airflow inside the opening slot 1252A to flow synchronously at high speed. According to Bernoulli's principle of fluid dynamics, fluid velocity and fluid static pressure are negatively correlated. The convergent and gradually narrowing structure of the opening slot 1252A causes the airflow velocity inside the slot to continuously increase along the direction of the diameter contraction, thereby making the air pressure inside the opening slot 1252A significantly lower than the atmospheric pressure outside the outer skin layer 140 of the opening slot 1252A, forming a stable and controllable negative pressure difference inside and outside the opening slot 1252A. The negative pressure difference continuously generates an inward pneumatic traction force on the side of the skin layer 140 facing the fifth slider 1252, causing the skin layer 140 in the corresponding cheek area to undergo elastic indentation deformation under the action of negative pressure suction. Combined with the constraining and shaping effect of the opening contour of the opening groove 1252A, a biomimetic dimple expression with a rounded contour and a natural and stable shape is ultimately formed. At the same time, since the fifth slider 1252 and the skin layer 140 are set to be non-contact throughout the process, problems such as local stress concentration, wear and tear of the skin layer 140, and surface indentation damage caused by the rigid compression of the skin layer 140 by the slider are completely avoided. By adjusting the swing speed of the fifth link 1251 and the convergence angle of the opening groove 1252A, the magnitude of the negative pressure difference can be precisely adjusted, thereby controlling the dimple indentation depth and deformation amplitude, and realizing refined and smooth dynamic control of the dimple expression.

[0066] Apart from the differences mentioned above, the structure of the bionic facial expression robot 100 and its components provided in this embodiment can be optimized with reference to Embodiment 1, and will not be described in detail here.

[0067] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A biomimetic facial expression robot, characterized in that, include: The main skeletal structure includes a first facial region, a second facial region, and a third facial region; A facial adaptation module includes multiple facial adaptation components. Each facial adaptation component has a universal interface. The multiple facial adaptation components are defined as a first adaptation component, a second adaptation component, and a third adaptation component. The first adaptation component is located in the first facial region, the second adaptation component is located in the second facial region, and the third adaptation component is located in the third facial region. The first adaptation component is used for forming regular facial expressions, the second adaptation component is used for forming wrinkles, and the third adaptation component is used for forming wrinkles or dimples. An expression driving module is installed on the skeleton body and connected to the face adaptation module. The expression driving module includes multiple single core bodies with the same configuration. The single core body is configured to have multi-degree-of-freedom driving capability and has a standardized interface that cooperates with the general interface to realize the detachable connection between the face adaptation component and the expression driving module. The cortex is connected to the facial adaptation module and is located on the side opposite to the expression driving module; The controller is electrically connected to the expression driving module and is configured to achieve refined bionic expression presentation of the skin by controlling the changing trends of the movement speed, movement stroke, and movement angle of the facial adaptation component.

2. The bionic facial expression robot according to claim 1, characterized in that, The first adapter component includes a first link and a first slider. The first link is connected between the single-core body and the first slider. The first slider is fixedly installed on the dermis. The controller is configured to drive the first slider to move in a direction perpendicular to the length of the first link to tighten the dermis and form a regular expression.

3. The bionic facial expression robot according to claim 1, characterized in that, The second adapter component includes a second link and a second slider. The second link is connected between the single-core body and the second slider. The second slider is fixedly installed on the skin. The skin has a first notch distributed at intervals on the side near the face adapter module. The controller is configured to drive the first notch to deform so that the skin forms wrinkles.

4. The bionic facial expression robot according to claim 1, characterized in that, The third adapter component includes a third link and a third slider. The third link connects the single-core body and the third slider. The third slider is fixedly installed on the skin. The skin is provided with an extended protrusion that cooperates with the third slider. The skin is provided with a second notch spaced apart on the side near the facial adapter module. The controller is configured to: within a first stroke range, drive the third adapter component to swing to generate biomimetic wrinkles in the skin; within a second stroke range, drive the third adapter component to swing to generate biomimetic dimples in the skin; and the first stroke range is smaller than the second stroke range.

5. The bionic facial expression robot according to claim 4, characterized in that, The skin layer is provided with spaced first deformation cavities, which are used to deform the skin layer. The third slider has a hooked surface, and a mating surface that mates with the hooked surface is formed on the skin layer. The skeleton body is provided with a receiving groove for accommodating part of the third slider.

6. The bionic facial expression robot according to claim 1, characterized in that, The multiple facial adaptation components are further defined as a fourth adaptation component, which includes a fourth link and a fourth slider. The fourth link is connected between the single-mechanism main body and the fourth slider. The fourth slider is slidably mounted on the skin. The controller is configured such that: the single-mechanism main body drives the fourth slider to swing along a first preset angle to generate biomimetic wrinkles on the skin; and the single-mechanism main body drives the fourth slider to slide along a second preset angle to generate biomimetic dimples on the skin.

7. The bionic facial expression robot according to claim 1, characterized in that, The main skeleton divides the human face into multiple functional areas with different degrees of deformation. The cortex is set with a gradient thickness corresponding to each functional area. The first thickness is configured to correspond to the cortex thickness of areas with large facial deformation and frequent movement, which facilitates cortex stretching and wrinkling deformation. The second thickness is configured to correspond to the cortex thickness of areas with small facial deformation and mainly provide contour support, which is used to maintain the overall stability of the facial shape. The first thickness is smaller than the second thickness, and the areas with different thicknesses transition continuously and smoothly.

8. The bionic facial expression robot according to claim 1, characterized in that, The skin layer and the facial adapter components are detachable and replaceable structures. By replacing the skin layer with different materials, thicknesses, and textures and the corresponding facial adapter components, the facial expressions and skin morphology of children, young adults, middle-aged adults, and the elderly can be simulated respectively. The slider of the facial adaptation component is configured to make surface contact with the skin. By controlling the size of the contact area between the slider and the skin, the force distribution and deformation degree of the skin can be adjusted to form different bionic expressions.

9. A method for controlling diverse and refined facial expressions based on a single core, applied to the bionic facial expression robot described in any one of claims 1-8, characterized in that, The method uses a unified single-core drive to control the movement speed, movement stroke, and movement angle of the single-core body corresponding to each facial region, so as to output bionic movements corresponding to different expressions and emotions. The method sets differentiated control parameters for different expression formation structures, wherein the first adapter component uses movement angle and movement stroke as the main control parameters; the second adapter component uses movement stroke and movement speed as the main control parameters; and the third adapter component uses local movement angle and gradual movement stroke as the main control parameters.

10. The method for controlling diverse and refined facial expressions based on a single core module according to claim 9, characterized in that, The following overall magnitude relationship exists between each facial expression / movement and its motion parameters and velocity change trends: Speed ​​of movement: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calmness; Exercise sequence: Laughter > Surprise > Dimples > Smile > Sadness > Wrinkles > Calmness; From a movement perspective: laughter > surprise > dimples > smile > sadness > wrinkles > calmness; The degree of drastic change in speed: Laughter > Surprise > Smile = Dimples > Sadness > Wrinkles > Calmness.

Images