Intelligent biomimetic expression robot control system employing facial muscle neural network

Abstract

Disclosed is an intelligent biomimetic expression robot control system employing a facial muscle neural network. The control system comprises a support body (110), a power module (120), a control module (130), a muscle neural network (140), and an expression generation body (150). The control module comprises control units (131) and first sensors (132). The muscle neural network comprises mesh conductors (141) and second sensors (143). Multiple intersection points are formed between the mesh conductors, and a drive unit (142) is arranged on each intersection point. A terminal end of each mesh conductor is electrically connected to a first output port (121) of the power module. The first sensors (132) are used for sensing a longitudinal force generated by the drive units, and the second sensors (143) are used for sensing a lateral force generated by the drive units. Hence, when power differences are applied to the respective drive units, the power differences between adjacent drive units change continuously, thereby resulting in continuous deformation of deformation zones.

Description

A facial muscle neural network intelligent bionic expression robot control system Technical Field

[0001] This invention relates to the technical field of bionic facial expression robots, and more particularly to a facial muscle neural network intelligent bionic facial expression robot control system. Background Technology

[0002] With the development of science and technology, humanoid robot technology is evolving rapidly and has become a new frontier in technological competition. Domestic humanoid robots have grown from nothing to something, with numerous technological achievements, expanding application scenarios, and accelerating the localization of core components, gradually moving towards a more advanced and intelligent direction. However, due to the complex internal structure of humanoid robots, multiple power sources are needed to control multiple execution units, making it difficult to achieve precise control. Furthermore, the low consistency of coordinated control among multiple power sources results in low coordination of local control quantities in bionic robots, making it difficult to achieve natural biomimetic effects.

[0003] A micro-expression generation device and method, patent number CN 108710325 B, discloses a micro-expression driving module, N magnetic control bodies, and an expression subject, where N is an integer greater than 1. The micro-expression driving module is used to generate a magnetic field according to received control commands to control the movement trajectory of the magnetic control bodies. The magnetic control bodies are embedded in the expression subject and drive the expression subject to form different micro-expressions. This solution uses magnetic control bodies to replace complex mechanical stretching devices, which reduces the cost and control complexity of the micro-expression generation device to a certain extent. The magnetic control bodies are smaller, and more magnetic control bodies can be placed in the same volume. However, the magnetic control bodies are controlled independently of each other, resulting in poor consistency in coordinated control and poor biomimetic effect of natural expression changes.

[0004] Harbin Institute of Technology has proposed a facial expression simulation device based on a hydraulic bionic actuator (patent number CN 106393126 B). The device discloses a simulated muscle component, a flow channel control valve component, a micropump, and a logic control unit. The simulated muscle component is mounted on a simulated human face and is connected to the micropump via a flow channel. The flow channel control valve component is installed in the flow channel between the simulated muscle component and the micropump. The power supply controllers for both the flow channel control valve component and the micropump are connected to the logic control unit. This solution addresses, to some extent, the problems of complex mechanical structures, poor stability, high cost, and low simulation accuracy, as well as the bulky size, cumbersome control, and poor simulation effect of pneumatic actuators.

[0005] The inflatable dielectric elastomer hemispherical actuator, patent number CN100581039C, discloses a first pre-stretched dielectric elastomer film adhered to the inner surface of a first flexible film, and a second pre-stretched dielectric elastomer film adhered to the inner surface of a second flexible film. The inner surface of the upper end of the first pre-stretched dielectric elastomer film is bonded to the inner surface of the upper end of the second pre-stretched dielectric elastomer film, forming a hemispherical space between the first electrode and the substrate. This solution addresses, to some extent, the problems of complex mechanical structure, poor flexibility, and poor biomimetic performance in actuators.

[0006] However, while the aforementioned existing technologies have solved the problem of complex internal structures of bionic robots to some extent, the coordination of bionics cannot be well addressed, and the control quantities between various drive sources are difficult to refine. Furthermore, since the requirements for local muscle deformation vary during facial expression generation, and the requirement for continuous deformation must be ensured, it is necessary to solve the problems of continuity of bionic facial muscle deformation and improve the coordination of local control quantities in bionic robots. Summary of the Invention

[0007] The purpose of this invention is to provide a facial muscle neural network intelligent bionic expression robot control system, which aims to solve the problems of continuity of bionic facial muscle deformation and improve the coordination of local control quantities of bionic robots.

[0008] To address the aforementioned technical problems, a facial muscle neural network intelligent bionic expression robot control system is provided, comprising a support body, a power module, a control module, a muscle neural network, and an expression generator. The power module has multiple first output ports. The control module includes multiple control units and multiple first sensors, and the control module is electrically connected to the power module. The muscle neural network includes interconnected mesh conductors, multiple drive units, and multiple second sensors. The second sensors are connected to the mesh conductors, and multiple intersection points are formed between the mesh conductors. Each intersection point is connected to a drive unit, and the ends of the mesh conductors are electrically connected to the first output ports. The drive units generate driving force under the action of the control units. The first sensors are used to sense the longitudinal force generated by the drive units, and the second sensors are used to sense the lateral force generated by the drive units. The expression generator is connected to the support body and includes multiple deformation zones. Multiple drive units are connected to the deformation zones, and each drive unit generates a continuously changing driving force to cause continuous deformation of the deformation zones.

[0009] Furthermore, the first output port outputs a power difference through the grid conductor to each of the driving units, and the power difference between adjacent driving units changes continuously, so that each region of the deformation zone produces a continuous deformation.

[0010] Furthermore, the grid conductor includes multiple horizontal wires, multiple vertical wires, and intersection points located at the intersections of the horizontal and vertical wires. Each of the second sensors is located on the horizontal and vertical wires, and each of the driving units is located at the intersection points. The horizontal and vertical wires are respectively connected to the first output port to obtain continuous voltage signals, forming continuously changing voltage difference signals at each adjacent intersection point, so that the power difference between adjacent driving units changes continuously.

[0011] Furthermore, each of the control units forms an array located below the drive unit, and the control units are electrically connected to the power module, and the control units are fixed to the support body.

[0012] Furthermore, the power module also includes multiple second output ports, each of which is electrically connected to the control unit, and the first sensor abuts between the support body and each of the control units.

[0013] Furthermore, the muscle neural network also includes a plurality of shape memory alloys, each of which is connected around the intersection point of each of the horizontal and vertical wires, so that each of the driving units can elastically deform relative to the horizontal and vertical wires.

[0014] Furthermore, the driving unit is a first electromagnet, and the control unit is a second electromagnet.

[0015] Furthermore, when the first electromagnet and the second electromagnet generate a first attraction force, the first electromagnet moves closer to the second electromagnet on the plane; when the first electromagnet and the second electromagnet generate a second attraction force, and the second attraction force is greater than the first attraction force, the first electromagnet moves closer to the second electromagnet in space.

[0016] Furthermore, when the first electromagnet and the second electromagnet generate a first repulsive force, the first electromagnet gradually moves away from the second electromagnet in the plane; when the first electromagnet and the second electromagnet generate a second repulsive force, and the second repulsive force is greater than the first repulsive force, the first electromagnet gradually moves away from the second electromagnet in space.

[0017] Furthermore, the drive unit is a first hydraulic pump and the control unit is a second hydraulic pump; or, the drive unit is a first air pump and the control unit is a second air pump.

[0018] Implementing the embodiments of the present invention will have the following beneficial effects:

[0019] 1. The facial muscle neural network intelligent bionic expression robot control system in this embodiment has a second sensor connected to the grid conductor, and multiple intersection points are formed between the grid conductors. Each intersection point is connected to a drive unit, and the ends of the grid conductors are electrically connected to the first output port. The drive unit generates driving force under the action of the control unit. The first sensor is used to sense the longitudinal force generated by the drive unit, and the second sensor is used to sense the lateral force generated by the drive unit. Thus, when the first output port outputs the power difference through the grid conductor to act on each drive unit, the power difference between adjacent drive units changes continuously, thereby causing the deformation area to generate continuous deformation, overcoming the problem of poor continuity and coordination of bionic facial muscle deformation in the prior art.

[0020] 2. In the facial muscle neural network intelligent bionic expression robot control system of this embodiment, since the grid conductor includes multiple horizontal wires, multiple vertical wires and intersection points located at the intersection of the horizontal and vertical wires, and the second sensor is distributed on the horizontal and vertical wires, and each driving unit is located at the intersection point, the horizontal and vertical wires are respectively connected to the first output port to obtain continuous voltage signals, so that a continuously changing voltage difference signal is formed at each adjacent intersection point and acts on the driving unit, thereby causing the power difference value of adjacent driving units to change continuously;

[0021] 3. The facial muscle neural network intelligent bionic expression robot control system in this embodiment includes multiple memory alloys, and each memory alloy is connected around the intersection of each horizontal and vertical wire, so that each driving unit can generate elastic deformation relative to the horizontal and vertical wires, thereby enabling the driving unit to drive the deformation area to generate deformation and natural recovery deformation.

[0022] 4. In the facial muscle neural network intelligent bionic expression robot control system of this embodiment, since the driving unit is the first electromagnet and the control unit is the second electromagnet, when the first electromagnet and the second electromagnet generate a first attraction force, the first electromagnet moves closer to the second electromagnet on the plane, thereby causing the deformation area to produce a planar deformed expression; when the first electromagnet and the second electromagnet generate a second attraction force, and the second attraction force is greater than the first attraction force, the first electromagnet moves closer to the second electromagnet in space, thereby causing the deformation area to produce a three-dimensional concave deformed expression; when the first electromagnet and the second electromagnet generate a first repulsive force, the first electromagnet moves away from the second electromagnet on the plane, thereby causing the deformation area to produce a planar deformed expression; when the first electromagnet and the second electromagnet generate a second repulsive force, and the second repulsive force is greater than the first repulsive force, the first electromagnet moves away from the second electromagnet in space, thereby causing the deformation area to produce a three-dimensional bulging deformed expression. Attached Figure Description

[0023] 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.

[0024] Figure 1 is a front view of the muscle neural network described in an embodiment of the present invention;

[0025] Figure 2 is a circuit diagram of the facial muscle neural network intelligent bionic expression robot control system according to an embodiment of the present invention;

[0026] Figure 3 is a side view of the facial muscle neural network intelligent bionic expression robot control system according to an embodiment of the present invention;

[0027] Figure 4 is a magnified view of a portion of point A in Figure 3;

[0028] Figure 5 is a schematic diagram of the facial muscle neural network intelligent bionic expression robot control system according to an embodiment of the present invention;

[0029] Figure 6 is a schematic diagram of the structure of the facial expression generator according to an embodiment of the present invention.

[0030] Wherein: 100, robot control system; 110, support body; 120, power module; 121, first output port; 122, second output port; 130, control module; 131, control unit; 132, first sensor; 140, muscle neural network; 141, mesh conductor; 1411, horizontal wire; 1412, vertical wire; 142, drive unit; 143, second sensor; 144, shape memory alloy; 150, facial expression generator; 151, deformation zone. Detailed Implementation

[0031] 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.

[0032] 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.

[0033] 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 application belongs. The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It must be noted that as used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

[0034] Please refer to Figures 1-6. This embodiment of the invention provides an intelligent bionic facial expression robot control system 100 with a facial muscle neural network 140, including a support body 110, a power module 120, a control module 130, a muscle neural network 140, and an expression generator 150. The power module 120 has multiple first output ports 121; the control module 130 includes multiple control units 131 and multiple first sensors 132, and the control module 130 is electrically connected to the power module 120; the muscle neural network 140 includes interconnected mesh conductors 141, multiple drive units 142, and multiple second sensors 143. The second sensors 143 are connected to the mesh conductors 141. Multiple intersection points are formed between the grid conductors 141, and each intersection point is connected to a drive unit 142. The ends of the grid conductors 141 are electrically connected to the first output port 121. The drive unit 142 generates a driving force under the action of the control unit 131. The first sensor 132 is used to sense the longitudinal force generated by the drive unit 142, and the second sensor 143 is used to sense the lateral force generated by the drive unit 142. The expression generator 150 is connected to the support body 110. The expression generator 150 includes multiple deformation zones 151. Multiple drive units 142 are connected to the deformation zones 151. Each drive unit 142 generates a continuously changing driving force so that the deformation zone 151 generates a continuous deformation. In specific applications, the muscle neural network 140 may include at least two network conductors. Each network conductor can be fitted with a refined mesh to refine the expression changes. Since a second sensor 143 is connected to the mesh conductor 141, multiple intersection points are formed between the mesh conductors 141. A driving unit 142 is connected to each intersection point, and the ends of the network conductors are electrically connected to a first output port 121. It can be understood that the mesh conductor 141 is similar to a mesh made of interwoven wires, with four intersection points on each small mesh. Each intersection point radiates to an end, connecting to the first output port 121. The driving unit 142 is connected at the intersection point and receives the signal from the first output port 121. Therefore, under the action of the control unit 131, the driving unit 142 generates a changing driving force. The first sensor 143 connects to the second sensor 143 connected to the second sensor 143 connected to the third sensor 143 connected ... Sensor 132 is connected to control unit 131, which generates a reaction force. The first sensor 132 is used to sense the longitudinal force generated by drive unit 142, so that power module 120 can provide feedback to adjust the magnitude of the longitudinal force of drive unit 142. The second sensor 143 is connected to grid conductor 141 and is used to sense the lateral force generated by drive unit 142, so that power module 120 can provide feedback to adjust the magnitude of the longitudinal force of drive unit 142. In particular, when the first output port 121 outputs the power difference through grid conductor 141 to each drive unit 142, the power difference between adjacent drive units 142 changes continuously, so that deformation area 151 generates continuous deformation, thereby solving the problem of poor continuity and coordination of bionic facial muscle deformation.

[0035] In one possible implementation, the grid conductor 141 includes multiple horizontal conductors 1411, multiple vertical conductors 1412, and junction points located at the intersections of the horizontal conductors 1411 and the vertical conductors 1412. Each second sensor 143 is located on the horizontal conductors 1411 and the vertical conductors 1412, and each drive unit 142 is located at the junction point. The horizontal conductors 1411 and the vertical conductors 1412 are respectively connected to the first output port 121 to obtain continuous voltage signals, forming continuously changing voltage difference signals at each adjacent junction point, so that the power difference between adjacent drive units 142 changes continuously. In specific applications, since the grid conductor 141 includes multiple horizontal conductors 1411, multiple vertical conductors 1412, and intersection points located at the intersections of the horizontal conductors 1411 and the vertical conductors 1412, and the second sensors 143 are distributed on the horizontal conductors 1411 and the vertical conductors 1412, and each drive unit 142 is located at the intersection point, the horizontal conductors 1411 and the vertical conductors 1412 are respectively connected to the first output port 121 to obtain continuous voltage signals. Thus, a continuously changing voltage difference signal is formed at each adjacent intersection point, acting on the drive unit 142, thereby causing the power difference value of adjacent drive units 142 to change continuously. In this way, it is not necessary to connect wires separately to the drive unit 142, avoiding the problems of chaotic internal wiring and signal interference. It should be noted that, for example, the input terminals of adjacent drive units 142 on the horizontal guide wire 1411 are the same, but the output terminals are connected to different vertical guide wires 1412. In this case, if the signal controlling the adjacent vertical guide wire 1412 changes continuously, the adjacent drive units 142 on the horizontal guide wire 1411 will generate continuous power output values. Similarly, if the output terminals of adjacent drive units 142 on the vertical guide wire 1412 are the same, but the input terminals are connected to different horizontal guide wires 1411, the signal controlling the adjacent horizontal guide wire 1411 changes continuously, the adjacent drive units 142 on the vertical guide wire 1412 will generate continuous power output values, thereby enabling the adjacent drive units 142 to act on the deformation zone 151 to produce continuous and coordinated deformation.

[0036] In one possible implementation, each control unit 131 forms an array located below the drive unit 142, and the control unit 131 is electrically connected to the power module 120. The control unit 131 is fixed to the support body 110. In practical applications, since multiple control units 131 are arrayed below the drive unit 142 and electrically connected to the power module 120, when the drive unit 142 moves one unit, the corresponding control unit 131 continuously exerts a force on the drive unit 142, enabling the drive unit 142 to move multiple units. This allows the drive unit 142 to cause greater deformation of the deformation area 151, resulting in exaggerated facial expressions from the expression generator 150. Furthermore, it is worth noting that each control unit 131 can be controlled independently. For example, the drive unit 142 can generate a first driving force under the action of the control unit 131 in the first unit, and a second driving force different from the first driving force under the action of the control unit 131 in the second unit. This results in different deformation force effects on the deformation area 151, allowing the expression generator 150 to better refine the generation effect of simulated facial expressions.

[0037] In one possible implementation, the power module 120 further includes a plurality of second output ports 122, each of which is electrically connected to a control unit 131. A first sensor 132 abuts against the support 110 and each control unit 131. In specific applications, the second output ports 122 are electrically connected to the control units 131, and the first sensor 132 abuts against the support 110 and each control unit 131. The second output ports 122 adapt and adjust their command signals according to the command signals from the power module 120 to the first output ports 121. The first sensor 132 collects the longitudinal feedback signals generated by the control units 131 into the power module 120, enabling the power module 120 to dynamically balance and adjust the command signals of the second output ports 122.

[0038] In one possible implementation, the muscle neural network 140 further includes a plurality of shape memory alloys 144, each shape memory alloy 144 being connected around the intersection points of each horizontal conductor 1411 and each vertical conductor 1412, so that each driving unit 142 can elastically deform relative to the horizontal conductor 1411 and the vertical conductor 1412. Specifically, each shape memory alloy 144 is connected around the intersection points of each horizontal conductor 1411 and each vertical conductor 1412. It can be understood that there are four vertices in the grid conductor 141, each vertex being an intersection point, and each intersection point having four edges, with shape memory alloy 144 connected to each edge. Thus, the driving unit 142 located at the intersection point can move in any direction, allowing each driving unit 142 to elastically deform relative to the horizontal conductor 1411 and the vertical conductor 1412 on the shape memory alloy 144, thereby enabling the driving unit 142 to drive the deformation region 151 to generate deformation and natural recovery deformation.

[0039] In one possible implementation, the driving unit 142 is a first electromagnet, and the control unit 131 is a second electromagnet. When the first electromagnet and the second electromagnet generate a first attraction force, the first electromagnet moves closer to the second electromagnet on a plane. When the first electromagnet and the second electromagnet generate a second attraction force, and the second attraction force is greater than the first attraction force, the first electromagnet moves closer to the second electromagnet in space. When the first electromagnet and the second electromagnet generate a first repulsion force, the first electromagnet moves further away from the second electromagnet on a plane. When the first electromagnet and the second electromagnet generate a second repulsion force, and the second repulsion force is greater than the first repulsion force, the first electromagnet moves further away from the second electromagnet in space. In specific applications, when the first electromagnet and the second electromagnet generate a first attraction force, the first electromagnet is closer to the second electromagnet on the plane. It is worth noting that at this time, the first attraction force can serve as a load for the deformation of the expression generator 150 and the lateral deformation of the shape memory alloy 144, thereby causing the deformation area 151 to produce a planar deformed expression. When the first electromagnet and the second electromagnet generate a second attraction force, and the second attraction force is greater than the first attraction force, the first electromagnet is closer to the second electromagnet in space. It is worth noting that at this time, the first attraction force can serve as a load for the deformation of the expression generator 150 and the lateral and longitudinal deformation of the shape memory alloy 144, thereby causing the deformation area 151 to produce a three-dimensional concave deformed expression. Thus, the expression generator 150 can simulate the three-dimensional concave effect of the deformation area 151.

[0040] When the first electromagnet and the second electromagnet generate a first repulsive force, the first electromagnet gradually moves away from the second electromagnet in the plane. It is worth noting that at this time, the first repulsive force can serve as a load for the deformation of the expression generator 150 and the lateral deformation of the shape memory alloy 144, thereby causing the deformation area 151 to produce a planar deformed expression. When the first electromagnet and the second electromagnet generate a second repulsive force, and the second repulsive force is greater than the first repulsive force, the first electromagnet gradually moves away from the second electromagnet in space. It is worth noting that at this time, the second repulsive force can serve as a load for the deformation of the expression generator 150 and the lateral and longitudinal deformation of the shape memory alloy 144, thereby causing the deformation area 151 to produce a three-dimensional bulging deformed expression. Thus, the expression generator 150 can simulate the three-dimensional bulging effect of the deformation area 151.

[0041] In one possible implementation, the drive unit 142 is a first hydraulic pump and the control unit 131 is a second hydraulic pump; or, the drive unit 142 is a first air pump and the control unit 131 is a second air pump. In specific applications, when the drive unit 142 is the first hydraulic pump and the control unit 131 is the second hydraulic pump, the first output port 121 controls the power of the first hydraulic pump, and the first hydraulic pump generates a driving force that acts on the deformation zone 151 to produce lateral deformation. The second output port 122 controls the power of the second hydraulic pump, and the second hydraulic pump generates a driving force that acts on the deformation zone 151 to produce longitudinal deformation, thereby enabling the expression generator 150 to generate expressions through hydraulic pressure as the driving force. Similarly, when the drive unit 142 is the first air pump and the control unit 131 is the second air pump, the first output port 121 controls the power of the first air pump, and the first air pump generates a driving force that acts on the deformation zone 151 to produce lateral deformation. The second output port 122 controls the power of the second air pump, and the second air pump generates a driving force that acts on the deformation zone 151 to produce longitudinal deformation, thereby enabling the expression generator 150 to generate expressions through air pressure as the driving force.

[0042] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, 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 scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A facial muscle neural network intelligent bionic expression robot control system, characterized in that, The application relates to a muscle nerve network, which comprises a support body, a power module, a control module, a muscle nerve network, an expression generating body and a plurality of first output ports. The power module is provided with a plurality of first output ports. The control module comprises a plurality of control units and a plurality of first sensors, and is electrically connected with the power module. The muscle nerve network comprises mesh conductors, a plurality of driving units and a plurality of second sensors, the mesh conductors are connected with the second sensors, a plurality of intersection points are formed between the mesh conductors, each intersection point is connected with the driving unit, the ends of the mesh conductors are electrically connected with the first output ports, the driving units generate driving force under the action of the control units, the first sensors are used for sensing the longitudinal force generated by the driving units, and the second sensors are used for sensing the transverse force generated by the driving units. The expression generating body is connected with the support body, the expression generating body comprises a plurality of deformation zones, the deformation zones are connected with a plurality of driving units, each driving unit generates continuously changing driving force to make the deformation zones generate continuous deformation amount. The first output ports output power difference through the mesh conductors to act on each driving unit, and the power difference of adjacent driving units continuously changes to make each region of the deformation zones generate continuous deformation amount. The mesh conductors comprise a plurality of transverse wires, a plurality of longitudinal wires and intersection points located at the intersection of the transverse wires and the longitudinal wires, each second sensor is located on the transverse wire and the longitudinal wire, and each driving unit is located at the intersection point, the transverse wire and the longitudinal wire are connected with the first output ports to obtain continuous voltage signals, and continuous voltage difference signals are formed at each adjacent intersection point to make the power difference of adjacent driving units continuously change. Each control unit forms an array below the driving units, and the control units are electrically connected with the power module and fixed to the support body.

2. The facial muscle neural network intelligent biomimetic expression robot control system of claim 1, wherein, The power module further comprises a plurality of second output ports, the second output ports are respectively electrically connected with the control units, and the first sensors are abutted between the support body and each control unit.

3. The facial muscle neural network intelligent biomimetic expression robot control system of claim 2, wherein, The muscle nerve network further comprises a plurality of memory alloys, each memory alloy is connected around the intersection point of each transverse wire and each longitudinal wire to make each driving unit elastically deform relative to the transverse wire and the longitudinal wire.

4. The facial muscle neural network intelligent biomimetic expression robot control system of claim 1, wherein, The driving units are first electromagnets, and the control units are second electromagnets.

5. The facial muscle neural network intelligent biomimetic expression robot control system of claim 4, wherein, When the first electromagnet and the second electromagnet generate a first attractive force, the first electromagnet approaches the second electromagnet in a plane; when the first electromagnet and the second electromagnet generate a second attractive force, and the second attractive force is greater than the first attractive force, the first electromagnet approaches the second electromagnet in space.

6. The facial muscle neural network intelligent biomimetic expression robot control system of claim 5, wherein, ​ 7. The facial muscle neural network intelligent biomimetic expression robot control system of claim 5, wherein, When the first electromagnet and the second electromagnet generate a first repulsive force, the first electromagnet gradually moves away from the second electromagnet in a plane; when the first electromagnet and the second electromagnet generate a second repulsive force, and the second repulsive force is greater than the first repulsive force, the first electromagnet gradually moves away from the second electromagnet in space.

8. The facial muscle neural network intelligent biomimetic expression robot control system of claim 1, wherein, The driving unit is a first hydraulic pump, and the control unit is a second hydraulic pump; or, the driving unit is a first air pump, and the control unit is a second air pump.

Images