Physical force sense feedback system for interaction between humanoid robot and environment

Abstract

The invention relates to the technical field of force sense feedback systems, in particular to a humanoid robot and environment interaction physical force sense feedback system which comprises an object stress analysis module, an object deformation analysis module, a control module, a grasping force judgment module and a communication module. The object stress analysis module is used for analyzing and obtaining related information of object stress and transmitting the related information to the control module; the object deformation analysis module is used for analyzing and obtaining related information of object deformation and transmitting the related information to the control module; the control module obtains a grasping index according to the related information of the object stress and the related information of the object stress, and transmits the grasping index to the grasping force judgment module; the grabbing force judgment module obtains information that the force of the robot for grabbing the object is too large or proper according to the grabbing index and transmits the information to the communication module; and the communication module transmits the information that the force of the robot for capturing the object is too large or proper to the robot hand execution module.

Description

Technical Field

[0001] The present invention relates to the technical field of force feedback systems, and more particularly to a physical force feedback system for the interaction between a humanoid robot and the environment. Background Art

[0002] A humanoid robot refers to a robot that mimics the human form and behavior, usually having human structural features such as a head, torso, arms, legs, etc., and being able to interact effectively with the human environment. These robots can not only perform some simple actions but may also possess the ability of perception and autonomous decision-making. Environmental interaction refers to the physical contact or perception between the robot and its surrounding environment (including objects, other organisms, etc.). The robot can interact with the external environment through sensors and actuators, including actions such as touching, grasping, pushing, and pulling. Such interaction is a very important part when the robot is performing tasks, especially in fields such as item handling, healthcare, and services. Force feedback is when the robot interacts with the environment, and by simulating the sensations during physical contact (such as pressure, friction, vibration, etc.) to enhance the interaction experience between the robot and humans or between the robot and objects. Simply put, when the robot touches an object, it not only "knows" the existence of the object through visual or tactile sensors but can also "feel" information such as the hardness, weight, or shape of the object.

[0003] In robotics, a force feedback system can help the robot perform more precise operations and more natural interactions. For example, when grasping an item, the robot can sense the weight and elasticity of the item, thereby adjusting the force to avoid damaging the item due to excessive force application.

[0004] The application document with the publication number CN115079684A discloses a feedback method for a robot and the robot, which relates to the technical field of robotics and can improve the anthropomorphic degree of the robot. Among them, an IMU is installed in the robot, and touch sensors are arranged in multiple first preset areas on the inner side of the robot's shell. The touch sensors are used to collect the operations acting on the robot, and the IMU is used to collect the heading angle of the robot. The robot can receive a first operation acting on the first area, and the first area is any one of the multiple first preset areas. Moreover, the robot can obtain the position information of the first area, the operation type of the first operation, and a first offset amount, and the first offset amount is used to reflect the strength of the first operation acting on the robot. After that, in response to the first operation, the robot can execute a first control event corresponding to the position information of the first area, the operation type of the first operation, and the strength of the first operation acting on the robot.

[0005] The prior art mainly collects simple operation information and force offsets through touch sensors, while ignoring the judgment and adjustment of the robot's grasping behavior through more comprehensive physical force analysis. Summary of the Invention

[0006] An object of the present invention is to propose a physical force feedback system for a humanoid robot to interact with the environment in view of the above deficiencies.

[0007] The present invention adopts the following technical solutions:

[0008] A physical force feedback system for a humanoid robot to interact with the environment, the system includes an object force analysis module, an object deformation analysis module, a control module, a grasping force judgment module and a communication module; the object force analysis module is used to analyze and obtain relevant information about the object's force and transmit it to the control module; the object deformation analysis module is used to analyze and obtain relevant information about the object's deformation and transmit it to the control module; the control module obtains a grasping index based on the relevant information about the object's force and the relevant information about the object's force and transmits it to the grasping force judgment module; the grasping force judgment module obtains information indicating whether the force of the robot grasping the object is too large or appropriate based on the grasping index and transmits it to the communication module; the communication module transmits the information indicating whether the force of the robot grasping the object is too large or appropriate to the robot hand execution module.

[0009] Optionally, the object force analysis module is used to analyze and obtain the total number of grasping points, the contact pressure of each grasping point, the contact area of each grasping point, the normal vector of the robot finger contact surface and the normal vector of the object contact surface, and transmit them to the control module; the object deformation analysis module is used to analyze and obtain the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, the minimum curvature of each reference area on the object surface after grasping, the surface area of the object, the surface area of the reference area and the volume of the object, and transmit them to the control module; the control module obtains the total number of reference areas based on the total number of grasping points, the surface area of the object, the surface area of the reference area and the volume of the object, obtains the object deformation index based on the total number of reference areas, the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping and the minimum curvature of each reference area on the object surface after grasping, obtains the contact angle of each grasping point based on the normal vector of the robot finger contact surface and the normal vector of the object contact surface, obtains the object force index based on the total number of grasping points, the contact pressure of each grasping point, the contact area of each grasping point and the contact angle of each grasping point, and obtains the grasping index based on the object force index and the object deformation index.

[0010] Optionally, the object force analysis module includes a contact pressure analysis sub-module, a contact area analysis sub-module, and a normal vector analysis sub-module; the contact pressure analysis sub-module is configured to analyze and obtain the total number of grasping points and the contact pressure of each grasping point, and transmit them to the control module; the contact area analysis sub-module is configured to analyze and obtain the contact area of each grasping point, and transmit it to the control module; the normal vector analysis sub-module is configured to analyze and obtain the normal vector of the robot finger contact surface and the normal vector of the object contact surface, and transmit them to the control module.

[0011] Optionally, the contact pressure analysis sub-module includes a contact pressure monitoring unit and a quantity statistics unit; the contact pressure monitoring unit is configured to monitor and obtain the contact pressure of each grasping point, and transmit it to the control module; the quantity statistics unit obtains the total number of grasping points based on the contact pressure of each grasping point, and transmits it to the control module.

[0012] Optionally, the contact area analysis sub-module includes a depth camera unit, an image segmentation unit, and an area calculation unit; the depth camera unit is configured to capture and obtain a contact image; the image segmentation unit uses an edge detection algorithm to segment the contact area from the contact image; the area calculation unit obtains the contact area of each grasping point based on the contact area, and transmits it to the control module.

[0013] Optionally, the normal vector analysis sub-module includes a data acquisition unit, a coordinate establishment unit, and a normal vector analysis unit; the data acquisition unit is configured to obtain information on the contact surfaces of the robot finger and the object; the coordinate establishment unit defines a local coordinate system based on the information on the contact surfaces of the robot finger and the object; the normal vector analysis unit calculates the normal vector of the robot finger contact surface and the normal vector of the object contact surface through local fitting, and transmits them to the control module.

[0014] Optionally, the object deformation analysis module includes a curvature analysis sub-module, a geometry analysis sub-module, and a data setting sub-module; the curvature analysis sub-module is configured to analyze and obtain the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, and the minimum curvature of each reference area on the object surface after grasping, and transmit them to the control module; the geometry analysis sub-module is configured to analyze and obtain the surface area and volume of the object, and transmit them to the control module; the data setting sub-module is configured to set the surface area of the reference area, and transmit it to the control module.

[0015] Optionally, the curvature analysis sub-module includes a visual sensing unit, a curvature calculation unit, and a data transmission unit; the visual sensing unit acquires the three-dimensional point cloud data of the object surface before and after grasping, and divides the object surface into multiple reference regions; the curvature calculation unit calculates the curvature of each reference region on the object surface before grasping and the curvature of each reference region on the object surface after grasping by fitting a surface, and screens out the maximum curvature of each reference region on the object surface after grasping and the minimum curvature of each reference region on the object surface after grasping; the data transmission unit transmits the curvature of each reference region on the object surface before grasping, the maximum curvature of each reference region on the object surface after grasping, and the minimum curvature of each reference region on the object surface after grasping to the control module.

[0016] Optionally, the geometric analysis sub-module includes a visual processing unit and a geometric calculation unit; the visual processing unit is used to acquire the surface contour of the object and construct a three-dimensional model of the object; the geometric calculation unit calculates the surface area and volume of the object according to the three-dimensional model of the object and transmits them to the control module.

[0017] Optionally, when the control module calculates the grasping index, it satisfies the following formula: F = PW + CV; where F is the grasping index, PW is the object force index, and CV is the object deformation index.

[0018] The beneficial effects achieved by the present invention are as follows:

[0019] 1. Through the real-time feedback of object force analysis and object deformation analysis, the grasping force can be adjusted according to the characteristics of the object (such as vulnerability, deformation ability, etc.), enabling the robot to interact safely with different types of objects;

[0020] 2. The object force analysis module can analyze the force conditions at the contact points between the object and the robot, and can monitor the applied force at each contact point in real time, including key information such as contact pressure and contact area. This information is transmitted to the control module to provide data support for the force adjustment of the robot, reducing the risk of the robot damaging the object by applying excessive force;

[0021] 3. The control module calculates the grasping index based on multiple information (mechanical and deformation data) to accurately judge whether the grasping force is appropriate, avoiding object damage or instability;

[0022] 4. The collaborative work of the object force analysis module, the object deformation analysis module, and the control module enables the robot to perform more accurate and sensitive operations in a complex environment and adapt to the grasping requirements of different objects.

[0023] To enable a further understanding of the features and technical content of the present invention, please refer to the following detailed description of the present invention and the accompanying drawings. However, the provided drawings are only for reference and illustration, and are not intended to limit the present invention. Description of the Drawings

[0024] Figure 1 Schematic diagram of the overall structure of the present invention;

[0025] Figure 2 Schematic diagram of the structure of the contact pressure analysis sub-module in the present invention;

[0026] Figure 3 Schematic diagram of the structure of the contact area analysis sub-module in the present invention;

[0027] Figure 4 Schematic diagram of the structure of the normal vector analysis sub-module in the present invention;

[0028] Figure 5 Schematic diagram of the structure of the curvature analysis sub-module in the present invention;

[0029] Figure 6 Schematic diagram of the structure of the geometric analysis sub-module in the present invention;

[0030] Figure 7 Effect diagram of the present invention;

[0031] Figure 8 Schematic diagram of the overall structure of the second embodiment of the present invention;

[0032] Figure 9 Effect diagram of the second embodiment of the present invention. Detailed Implementation Modes

[0033] The following are specific embodiments to illustrate the implementation modes of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Additionally, the drawings of the present invention are only simple schematic illustrations and are not drawn according to actual dimensions, hereby declared in advance. The following implementation modes will further detail the related technical content of the present invention, but the disclosed content is not intended to limit the protection scope of the present invention.

[0034] Embodiment 1: This embodiment provides a physical force feedback system for the interaction between a humanoid robot and the environment, as shown in combination with Figures 1 to 7 shown.

[0035] Physical force feedback system for the interaction between a humanoid robot and the environment. The system includes an object force analysis module, an object deformation analysis module, a control module, a grasping force judgment module, and a communication module. The object force analysis module is used to analyze and obtain relevant information about the force on the object and transmit it to the control module. The object deformation analysis module is used to analyze and obtain relevant information about the deformation of the object and transmit it to the control module. The control module obtains a grasping index based on the relevant information about the force on the object and the relevant information about the force on the object and transmits it to the grasping force judgment module. The grasping force judgment module obtains information indicating whether the force of the robot grasping the object is too large or appropriate based on the grasping index and transmits it to the communication module. The communication module transmits the information indicating whether the force of the robot grasping the object is too large or appropriate to the robot hand execution module.

[0036] Specifically, when the grasping force judgment module makes a judgment, the following judgment principle is referred to: when the grasping index is greater than or equal to the selected threshold of the grasping index, it indicates that the force of the robot grasping the object is too large; when the grasping index is less than the selected threshold of the grasping index, it indicates that the force of the robot grasping the object is appropriate. The selected threshold of the grasping index is set by those skilled in the art. The reason for not covering the situation of "the force of the robot grasping the object is too small" in the above two cases is that when the robot grasps the object and lifts it to an appropriate height (set by those skilled in the art), the grasping index is calculated only when the object is stably maintained at the appropriate height. Assuming the object drops, the above lifting action should be repeated after adjustment until the object is maintained at the appropriate height and then the calculation is performed.

[0037] Optionally, the object force analysis module is used to analyze and obtain the total number of grasping points, the contact pressure of each grasping point, the contact area of each grasping point, the normal vector of the robot finger contact surface, and the normal vector of the object contact surface, and transmit them to the control module. The object deformation analysis module is used to analyze and obtain the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, the minimum curvature of each reference area on the object surface after grasping, the surface area of the object, the surface area of the reference area, and the volume of the object, and transmit them to the control module. The control module obtains the total number of reference areas based on the total number of grasping points, the surface area of the object, the surface area of the reference area, and the volume of the object, obtains the object deformation index based on the total number of reference areas, the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, and the minimum curvature of each reference area on the object surface after grasping, obtains the contact angle of each grasping point based on the normal vector of the robot finger contact surface and the normal vector of the object contact surface, obtains the object force index based on the total number of grasping points, the contact pressure of each grasping point, the contact area of each grasping point, and the contact angle of each grasping point, and obtains the grasping index based on the object force index and the object deformation index.

[0038] Optionally, the object force analysis module includes a contact pressure analysis sub-module, a contact area analysis sub-module, and a normal vector analysis sub-module; the contact pressure analysis sub-module is used to analyze and obtain the total number of grasping points and the contact pressure of each grasping point, and transmit them to the control module; the contact area analysis sub-module is used to analyze and obtain the contact area of each grasping point, and transmit it to the control module; the normal vector analysis sub-module is used to analyze and obtain the normal vector of the robot finger contact surface and the normal vector of the object contact surface, and transmit them to the control module.

[0039] Specifically, the normal vector of the robot finger contact surface and the normal vector of the object contact surface can be detected in the following way. Measure the contact force acting on the finger through a torque sensor installed on the robot finger: F r =(F x , F y , F z ), where F x , F y , F z are the force components of the robot finger in the robot base coordinate system (the robot base coordinate system is set by those skilled in the art, and the conversion method can apply the homogeneous transformation matrix). The direction of the contact force is consistent with the normal direction of the contact surface. The corresponding normal vector of the robot finger contact surface: The reaction force of the object on the finger can be measured through a torque sensor installed on the object contact surface or a force sensor on the contact surface of the robot finger: F o =(F' x , F′ y , F' z ), where F' x , F′ y , F' z are the force components of the object contact point in the robot base coordinate system. The direction of the normal vector is the same as the direction of the contact force but opposite. The corresponding normal vector of the object contact surface:

[0040] The negative sign indicates that the direction of the normal vector points outside the object surface.

[0041] Optionally, the contact pressure analysis sub-module includes a contact pressure monitoring unit and a quantity statistics unit; the contact pressure monitoring unit is used to monitor and obtain the contact pressure of each grasping point, and transmit it to the control module; the quantity statistics unit obtains the total number of grasping points based on the contact pressure of each grasping point, and transmits it to the control module.

[0042] Optionally, the contact area analysis sub-module includes a depth camera unit, an image segmentation unit, and an area calculation unit; the depth camera unit is configured to capture and obtain a contact image; the image segmentation unit uses an edge detection algorithm to segment the contact area from the contact image; the area calculation unit obtains the contact area of each grasping point based on the contact area and transmits it to the control module.

[0043] Optionally, the normal vector analysis sub-module includes a data acquisition unit, a coordinate establishment unit, and a normal vector analysis unit; the data acquisition unit is configured to obtain information on the contact surface between the robot finger and the object; the coordinate establishment unit defines a local coordinate system based on the information on the contact surface between the robot finger and the object; the normal vector analysis unit calculates the normal vector of the contact surface of the robot finger and the normal vector of the contact surface of the object through local fitting and transmits them to the control module.

[0044] Optionally, the object deformation analysis module includes a curvature analysis sub-module, a geometric analysis sub-module, and a data setting sub-module; the curvature analysis sub-module is configured to analyze and obtain the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, and the minimum curvature of each reference area on the object surface after grasping, and transmit them to the control module; the geometric analysis sub-module is configured to analyze and obtain the surface area and volume of the object and transmit them to the control module; the data setting sub-module is configured to set the surface area of the reference area and transmit it to the control module.

[0045] Optionally, the curvature analysis sub-module includes a visual sensing unit, a curvature calculation unit, and a data transmission unit; the visual sensing unit acquires the three-dimensional point cloud data of the object surface before and after grasping and divides the object surface into multiple reference areas; the curvature calculation unit calculates the curvature of each reference area on the object surface before grasping and the curvature of each reference area on the object surface after grasping by fitting a surface, and filters out the maximum curvature of each reference area on the object surface after grasping and the minimum curvature of each reference area on the object surface after grasping based on the curvature of each reference area on the object surface after grasping; the data transmission unit transmits the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, and the minimum curvature of each reference area on the object surface after grasping to the control module.

[0046] Optionally, the geometric analysis sub-module includes a visual processing unit and a geometric calculation unit; the visual processing unit is configured to obtain the surface contour of the object and construct a three-dimensional model of the object; the geometric calculation unit calculates the surface area and volume of the object based on the three-dimensional model of the object and transmits them to the control module.

[0047] Optionally, when the control module calculates the grasping index, the following formula is satisfied: F = PW + CV; where F is the grasping index, PW is the object force index, and CV is the object deformation index.

[0048] Optionally, when the control module calculates the grasping index, the following formula is satisfied:

[0049] where I is the total number of grasping points, p i is the contact pressure of the i-th grasping point, a i is the contact area of the i-th grasping point, θ i is the contact angle of the i-th grasping point;

[0050] n r is the normal vector of the contact surface of the robot finger, n o is the normal vector of the contact surface of the object;

[0051] N is the total number of reference areas, ql n is the curvature of the n-th reference area on the object surface before grasping, is the maximum curvature of the n-th reference area on the object surface after grasping, is the minimum curvature of the n-th reference area on the object surface after grasping;

[0052] s all is the surface area of the object, s ck is the surface area of the reference area, v all is the volume of the object.

[0053] Specifically, when the control module calculates the grasping index, refer to the following program code:

[0054]

[0055]

[0056] Specifically, the following points need to be noted when calculating the grasping index: This embodiment studies three-dimensional objects with non-abnormal structures.

[0057] The total number of grasping points is illustrated by the following example: Assume that the total number of fixed grasping points on the robot hand is 20. After the robot grasps, the total number of grasping points activated as sensed by the sensor (activation means that the sensor measures that the contact pressure is greater than 0) is 10, then the corresponding total number of grasping points is 10; on the robot's hand, especially for robots designed to be similar to human hands, the grasping points are located in the following parts: fingertips, finger roots, the opposite side of the thumb, the center of the palm, the palm-finger connection part, finger pads, the sides of the fingers, and the wrist.

[0058] The unit of the contact pressure of each gripping point is Newton.

[0059] The unit of the contact area of each gripping point is square centimeter.

[0060] The unit of the contact angle of each gripping point is radian, which refers to the included angle between the object surface and the contact point of the robot finger.

[0061] The units of the surface area of the object and the surface area of the reference region are both square meters; the contact angle of each gripping point is obtained through the normal vector of the corresponding surface, because the contact point on each object surface can be regarded as a part of the object surface, and the normal vector of each contact surface describes the overall orientation of the contact surface.

[0062] The unit of the volume of the object is cubic meter.

[0063] The surface areas of each reference region are the same, and the positions of the corresponding reference regions need to be set by those skilled in the art. Those skilled in the art generally select a plane as the reference region, so there is only one curvature before gripping. When the corresponding reference region deforms after gripping, there are multiple curvatures after gripping. At this time, the corresponding maximum curvature and minimum curvature are obtained through analysis and screening.

[0064] The above units are just examples. Those skilled in the art can set different units according to actual needs when implementing this solution.

[0065] This embodiment solves the problem that the traditional feedback system has a relatively high risk of damaging the object. The object force analysis module can monitor the applied force of each contact point in real time by analyzing the force conditions at the contact points between the object and the robot, including key information such as contact pressure and contact area. This information is transmitted to the control module to provide data support for the force adjustment of the robot, reducing the risk of the robot damaging the object by applying excessive force.

[0066] Embodiment 2: This embodiment includes all the contents of Embodiment 1 and provides a physical force perception feedback system for the interaction between a humanoid robot and the environment, as shown in combination with Figure 8 and Figure 9 as shown.

[0067] A physical force feedback system for a humanoid robot to interact with the environment, which also includes a position analysis module and an information storage module; the position analysis module is used to analyze and obtain the applied force before adjusting the highest grasping point, and transmit it to the control module; the information storage module is used to store the selection threshold of the grasping index and transmit it to the control module; the control module calculates the applied force after adjusting the highest grasping point based on the applied force before adjusting the highest grasping point, the selection threshold of the grasping index, the grasping index and the total number of grasping points, and transmits it to the communication module; the communication module transmits the applied force after adjusting the highest grasping point to the robot hand execution module; the robot makes adjustments according to the applied force after adjusting the highest grasping point.

[0068] Optionally, the position analysis module includes a contact point recognition sub-module, a position recognition sub-module and a data feedback sub-module; the contact point recognition sub-module recognizes the area where the robot finger contacts the object and the distribution of the contact points through image processing and force distribution analysis, and outputs the number and position coordinates of the contact points; the position recognition sub-module calculates the highest grasping point according to different position coordinates; the data feedback sub-module selects the applied force before adjusting the highest grasping point corresponding to the highest grasping point and transmits it to the control module.

[0069] Optionally, when the control module calculates the applied force after adjusting the highest grasping point, the following formula is satisfied:

[0070] where AM is the applied force after adjusting the highest grasping point, yh is the applied force before adjusting the highest grasping point, and f ref is the selection threshold of the grasping index.

[0071] Optionally, when the control module calculates the applied force after adjusting the highest grasping point, the following program code is referenced:

[0072]

[0073] Specifically, when the grasping index is greater than or equal to the selection threshold of the grasping index, it means that the force of the robot grasping the object is too large. At this time, the force applied to the contact point at the highest position can be adjusted. When the object remains unchanged at the appropriate height after adjustment, the corresponding grasping index can be continuously calculated and then it can be judged whether the requirements are met. Assuming that the requirements are not met, those skilled in the art can perform other adjustment operations (such as adjusting the force applied to the contact point at the highest position according to the same method as above). When the object drops after adjustment, those skilled in the art can perform other adjustment operations; the unit of the applied force after adjusting the highest grasping point and the applied force before adjusting the highest grasping point is Newton; the applied force before adjusting the highest grasping point refers to the force preset by the robot corresponding to the grasping point, which is not necessarily the same as the measured contact pressure.

[0074] The above units are just examples. Those skilled in the art can set different units according to actual needs when implementing this solution.

[0075] This embodiment solves the problem of poor flexibility of traditional feedback systems. It can handle diverse grasping requirements and flexibly adjust the applied force according to actual situations.

[0076] The content disclosed above is only the preferred and feasible embodiment of the present invention, and does not limit the protection scope of the present invention. Therefore, all equivalent technical changes made by using the content of the specification and drawings of the present invention are included in the protection scope of the present invention. In addition, the elements therein can be updated with the development of technology.

Claims

1. A physical force feedback system for humanoid robot to interact with the environment, characterized in that: The system includes an object force analysis module, an object deformation analysis module, a control module, a grasping force judgment module and a communication module; The object force analysis module is used to analyze and obtain relevant information about the object force, and transmit it to the control module; The object deformation analysis module is used to analyze and obtain relevant information about the object deformation, and transmit it to the control module; The control module obtains a grasping index based on the relevant information of the force applied to the object and the relevant information of the force applied to the object, and transmits the information to the grasping force determination module; The grasping force judgment module obtains information about whether the force of the robot grasping the object is too strong or appropriate according to the grasping index, and transmits the information to the communication module; The communication module transmits information that the force of the robot grasping the object is too strong or appropriate to the robot hand execution module.

2. The physical force feedback system for humanoid robot interaction with environment as claimed in claim 1, characterized in that: The object force analysis module is used to analyze and obtain the total number of grasping points, the contact pressure of each grasping point, the contact area of ​​each grasping point, the normal vector of the robot finger contact surface and the normal vector of the object contact surface, and transmit them to the control module; The object deformation analysis module is used to analyze and obtain the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, the minimum curvature of each reference area on the object surface after grasping, the surface area of ​​the object, the surface area of ​​the reference area and the volume of the object, and transmit them to the control module; The control module obtains the total number of reference areas based on the total number of grasping points, the surface area of ​​the object, the surface area of ​​the reference area and the volume of the object; obtains the object deformation index based on the total number of reference areas, the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping and the minimum curvature of each reference area on the object surface after grasping; obtains the contact angle of each grasping point based on the normal vector of the robot finger contact surface and the normal vector of the object contact surface; obtains the object force index based on the total number of grasping points, the contact pressure of each grasping point, the contact area of ​​each grasping point and the contact angle of each grasping point; and obtains the grasping index based on the object force index and the object deformation index.

3. The physical force feedback system for humanoid robot interaction with environment as claimed in claim 2, characterized in that: The object force analysis module includes a contact pressure analysis submodule, a contact area analysis submodule and a normal vector analysis submodule; The contact pressure analysis submodule is used to analyze and obtain the total number of gripping points and the contact pressure of each gripping point, and transmit the results to the control module; The contact area analysis submodule is used to analyze and obtain the contact area of ​​each grasping point, and transmit the result to the control module; The normal vector analysis submodule is used to analyze and obtain the normal vector of the robot finger contact surface and the normal vector of the object contact surface, and transmit them to the control module.

4. The physical force feedback system for humanoid robot interaction with environment as claimed in claim 3, characterized in that: The contact pressure analysis submodule includes a contact pressure monitoring unit and a quantity statistics unit; The contact pressure monitoring unit is used to monitor and obtain the contact pressure of each grasping point and transmit it to the control module; The number counting unit obtains the total number of grasping points according to the contact pressure of each grasping point, and transmits the total number to the control module.

5. The physical force feedback system for interaction between a humanoid robot and an environment as claimed in claim 3, characterized in that: The contact area analysis submodule includes a depth camera unit, an image segmentation unit and an area calculation unit; The depth camera unit is used to capture and obtain a contact image; The image segmentation unit segments the contact area from the contact image using an edge detection algorithm; The area calculation unit obtains the contact area of ​​each grasping point according to the contact region and transmits the result to the control module.

6. The physical force feedback system for interaction between a humanoid robot and an environment as claimed in claim 3, characterized in that: The normal vector analysis submodule includes a data acquisition unit, a coordinate establishment unit and a normal vector analysis unit; The data acquisition unit is used to obtain information about the contact surface between the robot finger and the object; The coordinate establishing unit defines a local coordinate system according to information of a contact surface between the robot finger and the object; The normal vector analysis unit calculates the normal vector of the robot finger contact surface and the normal vector of the object contact surface through local fitting, and transmits them to the control module.

7. The physical force feedback system for interaction between a humanoid robot and an environment as claimed in claim 2, characterized in that: The object deformation analysis module includes a curvature analysis submodule, a geometry analysis submodule and a data setting submodule; The curvature analysis submodule is used to analyze and obtain the curvature of each reference area on the surface of the object before grasping, the maximum curvature of each reference area on the surface of the object after grasping, and the minimum curvature of each reference area on the surface of the object after grasping, and transmit them to the control module; The geometric analysis submodule is used to analyze and obtain the surface area and volume of the object, and transmit them to the control module; The data setting submodule is used to set the surface area of ​​the reference area and transmit it to the control module.

8. The physical force feedback system for humanoid robot interaction with environment as claimed in claim 7, characterized in that: The curvature analysis submodule includes a visual sensing unit, a curvature calculation unit and a data transmission unit; The visual sensing unit acquires three-dimensional point cloud data of the surface of the object before and after grasping, and divides the surface of the object into a plurality of reference areas; The curvature calculation unit calculates the curvature of each reference area on the surface of the object before grasping and the curvature of each reference area on the surface of the object after grasping by a method of fitting the surface, and obtains the maximum curvature of each reference area on the surface of the object after grasping and the minimum curvature of each reference area on the surface of the object after grasping according to the curvature of each reference area on the surface of the object after grasping; The data transmission unit transmits the curvature of each reference area on the object surface before grasping, the maximum curvature of each reference area on the object surface after grasping, and the minimum curvature of each reference area on the object surface after grasping to the control module.

9. The physical force feedback system for humanoid robot interaction with environment as claimed in claim 7, characterized in that: The geometric analysis submodule includes a visual processing unit and a geometric calculation unit; The visual processing unit is used to obtain the surface contour of the object and construct a three-dimensional model of the object; The geometric calculation unit calculates the surface area and volume of the object according to the three-dimensional model of the object, and transmits the calculated results to the control module.

10. The physical force feedback system for interaction between a humanoid robot and an environment as claimed in claim 2, characterized in that: When the control module calculates the grasping index, the following formula is satisfied: F = PW + CV; Among them, F is the grasping index, PW is the object force index, and CV is the object deformation index.

Images