A robot scene simulation method applied to the digital robot industry chain

Abstract

The present invention discloses a robot scene simulation method applied to the digital robot industry chain, which relates to the technical field of scene simulation. The present invention obtains three-dimensional morphological data of a fixture in an actual operating environment, and combines geometric structure information to perform data fusion correction, effectively compensating for morphological deviations caused by manufacturing tolerances, wear and mechanical errors; in the robot grasping modeling process, the part posture changes and grasping errors in the clamping process are calculated through simulation analysis, and the rotation transformation matrix and stiffness matrix are used to analyze the offset of the parts in the clamping process, dynamically adjust the grasping points, and effectively reduce the assembly deviation caused by the clamping error; in the assembly path optimization stage, the path of the robot performing the assembly task is adjusted by the correction parameters in the fixture database, and the robot insertion force and assembly trajectory are optimized by comprehensively considering the grasping error and fixture morphological deviation, so that the robot can adaptively adjust the assembly strategy and improve the assembly accuracy.

Description

Technical Field

[0001] The present invention relates to the field of scene simulation technology, and in particular to a robot scene simulation method applied to a digital robot industry chain. Background Art

[0002] In the existing simulation methods of the digital robotics industry chain, robotic assembly tasks rely on CAD models for path planning, and the working state of the fixture is modeled by combining point cloud data, finite state machine control logic, etc.

[0003] This type of method is difficult to accurately simulate when the robot actually performs grasping, clamping, and assembly operations due to the mechanical errors of the fixture itself, the instability of the clamping force, and the slight deformation of the parts. As a result, the posture of the parts grasped by the robot in actual operation has slight deviations from the simulation results.

[0004] To address the above issues, some traditional simulation solutions preset fixed clamping forces and standard clamping positions during the fixture simulation process to approximately simulate the fixture forces in the real environment. In addition, some solutions combine visual inspection systems to perform secondary position correction after the robot grasps. However, rigid modeling cannot dynamically reflect the subtle changes of the fixture under different working conditions, and vision-based error compensation can often only be adjusted after grasping, and the robot's motion path cannot be optimized in advance, which affects the overall assembly efficiency. Therefore, a robot scene simulation solution is urgently needed to solve such problems. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a robot scene simulation method applied to the digital robot industry chain to solve the problem that existing robot simulation methods are difficult to accurately simulate the clamping force of the fixture, mechanical errors and part deformation, resulting in robot grasping posture deviation and reduced assembly accuracy.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] The embodiment of the present invention provides a robot scene simulation method applied to a digital robot industry chain, which includes:

[0009] Step S1: Based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment, a digital twin model of the robot and the fixture is constructed;

[0010] Step S2: Based on the digital twin model of the fixture in step S1, the robot grasping process is modeled, and the interaction between the robot actuator and the fixture is simulated in a simulation environment. The part posture changes and grasping errors during the gripping process are calculated, and the grasping points of the robot actuator are adjusted based on the simulation analysis.

[0011] Step S3, based on the grasping error calculated in step S2, obtain correction parameters from the fixture database and adjust the path of the robot performing the assembly task;

[0012] In step S4, assembly simulation is performed according to the path optimized in step S3, and the position deviation of parts in the actual assembly process is collected by a visual sensor and compared with the simulation data.

[0013] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, in which: in step S1, the point cloud data or visual measurement data of the fixture in the actual operating environment is obtained, the geometric structure information is corrected, and the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters are stored in the fixture database.

[0014] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, the steps of constructing a digital twin model of the robot and the fixture based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment are as follows:

[0015] Use laser scanning or visual measurement equipment to obtain the three-dimensional point cloud data of the fixture in the actual operating environment, and define the point cloud set as P:

[0016] P={p i |p i =(x i ,y i ,z i )},i=1,2,…,N,

[0017] Among them, P represents the point cloud data set of the fixture, p i represents the i-th sampling point, x i ,y i ,z i Represents point p i The coordinate value in the three-dimensional coordinate system, N represents the total number of sampling points,

[0018] Based on the point cloud data P, the least squares method or surface fitting method is used to reconstruct the geometric surface of the fixture. The reconstruction process is expressed as:

[0019]

[0020] Among them, S(x,y) represents the fixture surface equation after fitting, M represents the highest order of fitting, and a mn is the fitting coefficient to be determined, m, n represent the power exponent of the fitting equation, and control the order of the surface shape, x m ,y n Respectively represent the calculated values ​​of coordinates x and y at different powers,

[0021] By comparing the point cloud data P with the ideal fixture geometry model S0(x,y), the morphological deviation is calculated using the following formula:

[0022]

[0023] Where Δd i Indicates the morphological deviation value of the i-th point, x i,0 ,y i,0 ,z i,0 is the coordinate value of the corresponding point of the ideal geometric model of the fixture,

[0024] Construct a morphological deviation compensation model and correct the geometric model of the fixture. The model formula is:

[0025]

[0026] in, represents the geometry model of the fixture after compensation, S(x,y) is the uncorrected fitting surface equation, and w i is the weight coefficient, Δd i Indicates the morphological deviation correction value.

[0027] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, the step of correcting the geometric structure information and storing the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters in the fixture database is as follows:

[0028] Based on the morphological deviation compensation model, the geometric structure information of the fixture is updated, and the error distribution function is introduced. The function formula is:

[0029]

[0030] Among them, E(x,y) represents the mean square deviation of the morphological error, is the average morphological deviation of all sampling points,

[0031] By correcting the morphological error, the clamping force compensation parameters are calculated. The calculation formula is:

[0032]

[0033] Among them, F crepresents the clamping force after correction, F0 is the clamping force under ideal conditions, k is the influencing factor of the morphological deviation,

[0034] The corrected fixture geometry model Morphological deviation parameter Δd i And clamping force compensation parameter F c Store in fixture database DB:

[0035]

[0036] Among them, DB represents the fixture database, Store the corrected geometric model data, Δd stores the morphological deviation data, F c Stores the gripping force compensation parameters.

[0037] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, in which: in step S2, the corrected grasping point data, clamping force compensation parameters and simulation correction data are stored in the fixture database.

[0038] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, the steps of modeling the robot grasping process, simulating the interaction between the robot's actuator and the fixture in a simulation environment, calculating the part posture change and grasping error during the clamping process, and adjusting the grasping point of the robot actuator based on the simulation analysis are as follows:

[0039] Based on the digital twin model of the fixture, define the grasping equation of the robot actuator:

[0040]

[0041] Among them, G(x,y,z) represents the grasping position of the robot actuator, R(θ) is the rotation transformation matrix, and θ is the rotation angle. is the geometric model of the compensated fixture, T is the translation vector, which represents the displacement of the robot actuator relative to the fixture,

[0042] Calculate the change in part posture during the clamping process using the following formula:

[0043] Φ=J -1 F c ,

[0044] Where Φ is the angular offset vector of the part, J is the clamping stiffness matrix, which represents the elastic deformation relationship between the fixture and the part, -1 represents the inverse operation of the matrix, and F c is the corrected clamping force,

[0045] Calculate the gripping error e based on simulation analysis and adjust the gripping point:

[0046] e=||G-G0||,

[0047] G'=G-αe,

[0048] Among them, e is the grasping error, which represents the Euclidean distance between the actual grasping point G and the target grasping point G0, G0 is the target grasping point, which represents the grasping position of the robot in the ideal state, G' is the corrected grasping point, and α is the adjustment coefficient.

[0049] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, in which: during the adjustment of the path, the robot insertion force and assembly trajectory are optimized based on the corrected grasping point data, clamping force compensation parameters and fixture shape deviation.

[0050] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, the step of obtaining correction parameters from the fixture database and adjusting the path of the robot performing the assembly task is as follows:

[0051] Based on the grasping error e, the assembly path is corrected. The correction formula is:

[0052] P'=P-βe,

[0053] Among them, P is the original assembly path, which represents the initial trajectory of the robot performing the assembly task, P' is the corrected assembly path, β is the path correction coefficient,

[0054] Combined with the fixture shape deviation, the insertion force F a Optimize, the optimization formula is:

[0055]

[0056] Among them, F' a is the optimized insertion force, F a is the initial insertion force, which indicates the initial force value when the robot performs the assembly task, γ is the influence factor of the morphological deviation, and w i is the weight coefficient, which indicates the influence of different morphological deviations on the insertion force, Δd i is the morphological deviation correction value.

[0057] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, in which: in step S4, based on the comparison results, the morphological deviation parameters and clamping force compensation parameters in the fixture database are dynamically corrected, and the fixture digital twin model is updated.

[0058] As a preferred solution of the robot scene simulation method applied to the digital robot industry chain described in the present invention, the steps of dynamically correcting the morphological deviation parameters and clamping force compensation parameters in the fixture database based on the comparison results and updating the fixture digital twin model are as follows:

[0059] Compare visual inspection and error, measure part position deviation d based on visual sensor v And calculate the correction value:

[0060] Δd v =d v -d0,

[0061] Among them, d v is the part position deviation measured by vision, d0 is the simulation reference value, that is, the expected part position in the simulation calculation, Δd v is the error correction value, which represents the difference between the actual deviation and the simulated reference value,

[0062] Update the shape deviation parameter Δd and the clamping force compensation parameter F c , the update formula is:

[0063] Δd'=Δd+λΔd v ,

[0064] F' c =F c +μΔd v ,

[0065] Among them, Δd' is the updated morphological deviation parameter, Δd is the original morphological deviation parameter, λ is the updated weight coefficient, F' c is the updated clamping force compensation parameter, F c is the original clamping force compensation parameter, and μ is the clamping force compensation adjustment coefficient.

[0066] The beneficial effects of the present invention are as follows: the present invention obtains the three-dimensional morphological data of the fixture in the actual operating environment, and combines the geometric structure information for data fusion correction, effectively compensating for the morphological deviation caused by manufacturing tolerances, wear and mechanical errors; in the robot grasping modeling process, the part posture changes and grasping errors in the clamping process are calculated through simulation analysis, and the rotation transformation matrix and stiffness matrix are used to analyze the offset of the parts in the clamping process, dynamically adjust the grasping points, and effectively reduce the assembly deviation caused by the clamping error.

[0067] In the assembly path optimization stage, the present invention adjusts the path of the robot performing the assembly task through the correction parameters in the fixture database, and optimizes the robot's insertion force and assembly trajectory by comprehensively considering the grasping error and fixture shape deviation, so that the robot can adaptively adjust the assembly strategy and improve the assembly accuracy.

[0068] The present invention uses a visual sensor to collect the position deviation of parts in the actual assembly process after the assembly simulation is executed, compares it with the simulation data, calculates the error correction value, and dynamically updates the morphological deviation parameters and clamping force compensation parameters in the fixture database.

[0069] In summary, compared with traditional simulation methods that only rely on CAD models or fixed clamping forces for approximate simulation, it can more effectively compensate for virtual-reality synchronization errors, improve the simulation system's adaptability to fixture mechanical errors, part morphological deformation, and clamping force fluctuations, enable robots to have higher precision and stability in complex assembly tasks, and significantly reduce the assembly failure rate caused by the accumulation of clamping errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0071] Figure 1 It is a flow chart of the robot scene simulation method applied to the digital robot industry chain of the present invention. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0073] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0074] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0075] Example 1, reference Figure 1 This embodiment provides a robot scene simulation method applied to a digital robot industry chain, comprising the following steps:

[0076] Step S1: Based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment, a digital twin model of the robot and the fixture is constructed;

[0077] In step S1, the point cloud data or visual measurement data of the fixture in the actual operating environment is obtained, the geometric structure information is corrected, and the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters are stored in the fixture database;

[0078] Based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment, the steps to build the digital twin model of the robot and the fixture are as follows:

[0079] Use laser scanning or visual measurement equipment to obtain the three-dimensional point cloud data of the fixture in the actual operating environment, and define the point cloud set as P:

[0080] P={p i |p i =(x i ,y i ,z i )},i=1,2,…,N,

[0081] Among them, P represents the point cloud data set of the fixture, p i represents the i-th sampling point, x i ,y i ,z i Represents point p i The coordinate value in the three-dimensional coordinate system, N represents the total number of sampling points,

[0082] Based on the point cloud data P, the least squares method or surface fitting method is used to reconstruct the geometric surface of the fixture. The reconstruction process is expressed as:

[0083]

[0084] Among them, S(x,y) represents the fixture surface equation after fitting, M represents the highest order of fitting, and a mn is the fitting coefficient to be determined, m, n represent the power exponent of the fitting equation, and control the order of the surface shape, x m ,y n Respectively represent the calculated values ​​of coordinates x and y at different powers,

[0085] By comparing the point cloud data P with the ideal fixture geometry model S0(x,y), the morphological deviation is calculated using the following formula:

[0086]

[0087] Where Δd i Indicates the morphological deviation value of the i-th point, xi,0 ,y i,0 ,z i,0 is the coordinate value of the corresponding point of the ideal geometric model of the fixture,

[0088] Construct a morphological deviation compensation model and correct the geometric model of the fixture. The model formula is:

[0089]

[0090] in, represents the geometry model of the fixture after compensation, S(x,y) is the uncorrected fitting surface equation, and w i is the weight coefficient, Δd i Indicates the morphological deviation correction value;

[0091] Specifically, the fixture's point cloud data is acquired and fitted to construct a fixture geometry model based on real measurements. The fixture's geometry is corrected by calculating morphological deviations. A morphological deviation compensation model is introduced to enable dynamic adjustment of the fixture's digital twin model, improving modeling accuracy.

[0092] The steps of correcting the geometric structure information and storing the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters in the fixture database are:

[0093] Based on the morphological deviation compensation model, the geometric structure information of the fixture is updated, and the error distribution function is introduced. The function formula is:

[0094]

[0095] Among them, E(x,y) represents the mean square deviation of the morphological error, is the average morphological deviation of all sampling points,

[0096] By correcting the morphological error, the clamping force compensation parameters are calculated. The calculation formula is:

[0097]

[0098] Among them, F c represents the clamping force after correction, F0 is the clamping force under ideal conditions, k is the influencing factor of the morphological deviation,

[0099] The corrected fixture geometry model Morphological deviation parameter Δd i And clamping force compensation parameter F c Store in fixture database DB:

[0100]

[0101] Among them, DB represents the fixture database, Store the corrected geometric model data, Δd stores the morphological deviation data, F c Storing clamping force compensation parameters;

[0102] Specifically, based on the morphological deviation calculation results, the geometric model of the fixture is corrected, and the error distribution is further calculated to optimize the clamping force compensation parameters. The corrected data is stored in the database, which can truly reflect the actual working conditions in the digital twin environment.

[0103] Step S2: Based on the digital twin model of the fixture in step S1, the robot grasping process is modeled, and the interaction between the robot actuator and the fixture is simulated in a simulation environment. The part posture changes and grasping errors during the gripping process are calculated, and the grasping points of the robot actuator are adjusted based on the simulation analysis.

[0104] In step S2, the corrected gripping point data, clamping force compensation parameters and simulation correction data are stored in the fixture database;

[0105] The robot grasping process is modeled, and the interaction between the robot's actuator and the fixture is simulated in a simulation environment. The steps for calculating the part posture changes and grasping errors during the gripping process and adjusting the grasping points of the robot actuator based on the simulation analysis are as follows:

[0106] Based on the digital twin model of the fixture, define the grasping equation of the robot actuator:

[0107]

[0108] Among them, G(x,y,z) represents the grasping position of the robot actuator, R(θ) is the rotation transformation matrix, and θ is the rotation angle. is the geometric model of the compensated fixture, T is the translation vector, which represents the displacement of the robot actuator relative to the fixture,

[0109] Calculate the change in part posture during the clamping process using the following formula:

[0110] Φ=J -1 F c ,

[0111] Where Φ is the angular offset vector of the part, J is the clamping stiffness matrix, which represents the elastic deformation relationship between the fixture and the part, -1 represents the inverse operation of the matrix, and F c is the corrected clamping force,

[0112] Calculate the gripping error e based on simulation analysis and adjust the gripping point:

[0113] e=||G-G0||,

[0114] G'=G-αe,

[0115] Wherein, e is the grasping error, which represents the Euclidean distance between the actual grasping point G and the target grasping point G0, G0 is the target grasping point, which represents the grasping position of the robot in the ideal state, G' is the corrected grasping point, and α is the adjustment coefficient;

[0116] Specifically, based on the digital twin model, the interaction between the robot actuator and the fixture is modeled, and the part posture changes and grasping errors are calculated. The rotation transformation matrix and stiffness matrix are used to analyze the offset of the part during the robot clamping process, and the grasping point is adjusted based on the error analysis. The corrected grasping point ensures that the robot can grasp the part accurately and reduces assembly deviations caused by clamping errors.

[0117] Step S3, based on the grasping error calculated in step S2, obtain correction parameters from the fixture database and adjust the path of the robot performing the assembly task;

[0118] During the path adjustment process, the robot insertion force and assembly trajectory are optimized based on the corrected grasping point data, clamping force compensation parameters and fixture shape deviation;

[0119] The steps to obtain correction parameters from the fixture database and adjust the path of the robot to perform the assembly task are as follows:

[0120] Based on the grasping error e, the assembly path is corrected. The correction formula is:

[0121] P'=P-βe,

[0122] Among them, P is the original assembly path, which represents the initial trajectory of the robot performing the assembly task, P' is the corrected assembly path, β is the path correction coefficient,

[0123] Combined with the fixture shape deviation, the insertion force F a Optimize, the optimization formula is:

[0124]

[0125] Among them, F' a is the optimized insertion force, F a is the initial insertion force, which indicates the initial force value when the robot performs the assembly task, γ is the influence factor of the morphological deviation, and w i is the weight coefficient, which indicates the influence of different morphological deviations on the insertion force, Δd i is the morphological deviation correction value;

[0126] Specifically, the robot assembly path is corrected based on the grasping error to reduce the cumulative deviation during the assembly process. By calculating the impact of the error on the path and adjusting the insertion force, the assembly task is made more accurate, allowing the robot to adapt to the fixture shape deviation and grasping error when performing the assembly task, thereby improving the assembly accuracy.

[0127] Step S4, performing assembly simulation according to the path optimized in step S3, and collecting part position deviations during the actual assembly process through a visual sensor, and comparing them with the simulation data;

[0128] In step S4, based on the comparison results, the morphological deviation parameters and clamping force compensation parameters in the fixture database are dynamically corrected, and the fixture digital twin model is updated;

[0129] Based on the comparison results, the steps of dynamically correcting the morphological deviation parameters and clamping force compensation parameters in the fixture database and updating the fixture digital twin model are as follows:

[0130] Compare visual inspection and error, measure part position deviation d based on visual sensor v And calculate the correction value:

[0131] Δd v =d v -d0,

[0132] Among them, d v is the part position deviation measured by vision, d0 is the simulation reference value, that is, the expected part position in the simulation calculation, Δd v is the error correction value, which represents the difference between the actual deviation and the simulated reference value,

[0133] Update the shape deviation parameter Δd and the clamping force compensation parameter F c , the update formula is:

[0134] Δd'=Δd+λΔd v ,

[0135] F' c =F c +μΔd v ,

[0136] Among them, Δd' is the updated morphological deviation parameter, Δd is the original morphological deviation parameter, λ is the updated weight coefficient, F' c is the updated clamping force compensation parameter, F c is the original clamping force compensation parameter, μ is the clamping force compensation adjustment coefficient;

[0137] Specifically, by comparing the actual part positions with assembly simulation and visual sensors, the morphological deviation parameters and clamping force compensation parameters in the fixture database are dynamically adjusted to achieve continuous optimization of the digital twin model.

[0138] By updating the fixture database in real time, the robot can adaptively optimize the grasping and assembly process in a changing environment, improving the accuracy and stability of long-term operation.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A robot scene simulation method applied to the digital robot industry chain, characterized by: include, Step S1: Based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment, a digital twin model of the robot and the fixture is constructed; Step S2: Based on the digital twin model of the robot and fixture from step S1, the robot grasping process is modeled, and the interaction between the robot actuator and the fixture is simulated in a simulation environment. The part posture changes and grasping errors during the grasping process are calculated, and the grasping point of the robot actuator is adjusted based on the simulation analysis. Step S3, based on the grasping error calculated in step S2, obtain correction parameters from the fixture database and adjust the path of the robot performing the assembly task; Step S4, performing assembly simulation according to the path optimized in step S3, and collecting part position deviations during the actual assembly process through a visual sensor, and comparing them with the simulation data; The step S2 comprises: Based on the digital twin model of the robot and fixture, the grasping equation of the robot actuator is defined: Among them, G(x,y,z) represents the grasping position of the robot actuator, R(θ) is the rotation transformation matrix, and θ is the rotation angle. is the geometric model of the compensated fixture, T is the translation vector, which represents the displacement of the robot actuator relative to the fixture, Calculate the change in part posture during the clamping process using the following formula: Φ=J -1 F c , Where Φ is the angular offset vector of the part, J is the clamping stiffness matrix, which represents the elastic deformation relationship between the fixture and the part, -1 represents the inverse operation of the matrix, and F c is the corrected clamping force, Calculate the gripping error e based on simulation analysis and adjust the gripping point: e=||G-G0||, G'=G-αe, Wherein, e is the grasping error, which represents the Euclidean distance between the actual grasping point G and the target grasping point G0, G0 is the target grasping point, which represents the grasping position of the robot in the ideal state, G' is the corrected grasping point, and α is the adjustment coefficient; The step S3 comprises: Based on the grasping error e, the assembly path is corrected. The correction formula is: P'=P-βe, Among them, P is the original assembly path, which represents the initial trajectory of the robot performing the assembly task, P' is the corrected assembly path, β is the path correction coefficient, Combined with the fixture shape deviation, the insertion force F a Optimize, the optimization formula is: Among them, F' a is the optimized insertion force, F a is the initial insertion force, which indicates the initial force value when the robot performs the assembly task, γ is the influence factor of the morphological deviation, and w i is the weight coefficient, which indicates the influence of different morphological deviations on the insertion force, Δd i is the morphological deviation correction value.

2. A robot scene simulation method applied to a digital robot industry chain according to claim 1, characterized in that: In step S1, the point cloud data or visual measurement data of the fixture in the actual operating environment is obtained, the geometric structure information is corrected, and the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters are stored in the fixture database.

3. A robot scene simulation method applied to a digital robot industry chain according to claim 2, characterized in that: The steps of constructing a digital twin model of the robot and the fixture based on the geometric structure information of the fixture and the morphological deviation of the fixture in the actual operating environment are as follows: Use laser scanning or visual measurement equipment to obtain the three-dimensional point cloud data of the fixture in the actual operating environment, and define the point cloud set as P: P={p i |p i =(x i ,y i ,z i )},i=1,2,…,N, Among them, P represents the point cloud data set of the fixture, p i represents the i-th sampling point, x i ,y i ,z i Represents point p i The coordinate value in the three-dimensional coordinate system, N represents the total number of sampling points, Based on the point cloud data P, the least squares method or surface fitting method is used to reconstruct the geometric surface of the fixture. The reconstruction process is expressed as: Among them, S(x,y) represents the fixture surface equation after fitting, M represents the highest order of fitting, and a mn is the fitting coefficient to be determined, m, n represent the power exponent of the fitting equation, and control the order of the surface shape, x m ,y n Respectively represent the calculated values ​​of coordinates x and y at different powers, By comparing the point cloud data P with the ideal fixture geometry model S0(x,y), the morphological deviation is calculated using the following formula: Where Δd i Indicates the morphological deviation value of the i-th point, x i,0 ,y i,0 ,z i,0 is the coordinate value of the corresponding point of the ideal geometric model of the fixture, Construct a morphological deviation compensation model and correct the geometric model of the fixture. The model formula is: in, represents the geometry model of the fixture after compensation, S(x,y) is the uncorrected fitting surface equation, and w i is the weight coefficient, Δd i Indicates the morphological deviation correction value.

4. The robot scene simulation method applied to the digital robot industry chain according to claim 3, characterized in that: The step of correcting the geometric structure information and storing the corrected fixture geometric model, morphological deviation parameters and clamping force compensation parameters in the fixture database is: Based on the morphological deviation compensation model, the geometric structure information of the fixture is updated, and the error distribution function is introduced. The function formula is: Among them, E(x,y) represents the mean square deviation of the morphological error, is the average morphological deviation of all sampling points, By correcting the morphological error, the clamping force compensation parameters are calculated. The calculation formula is: Among them, F c represents the clamping force after correction, F0 is the clamping force under ideal conditions, k is the influencing factor of the morphological deviation, The corrected fixture geometry model Morphological deviation parameter Δd i And clamping force compensation parameter F c Store in fixture database DB: Among them, DB represents the fixture database, Store the corrected geometric model data, Δd stores the morphological deviation data, F c Stores the gripping force compensation parameters.

5. The robot scene simulation method applied to the digital robot industry chain according to claim 4, characterized in that: In step S2, the corrected grasping point data, clamping force compensation parameters and simulation correction data are stored in the fixture database.

6. The robot scene simulation method applied to the digital robot industry chain according to claim 5, characterized in that: During the path adjustment process, the robot insertion force and assembly trajectory are optimized based on the corrected grasping point data, clamping force compensation parameters and fixture shape deviation.

7. The robot scene simulation method applied to the digital robot industry chain according to claim 6, characterized in that: In step S4, based on the comparison results, the morphological deviation parameters and clamping force compensation parameters in the fixture database are dynamically corrected, and the digital twin models of the robot and the fixture are updated.

8. The robot scene simulation method applied to the digital robot industry chain according to claim 7, characterized in that: The steps of dynamically correcting the morphological deviation parameters and clamping force compensation parameters in the fixture database based on the comparison results and updating the digital twin model of the robot and the fixture are as follows: Compare visual inspection and error, measure part position deviation d based on visual sensor v And calculate the correction value: Δd v =d v -d0, Among them, d v is the part position deviation measured by vision, d0 is the simulation reference value, that is, the expected part position in the simulation calculation, Δd v is the error correction value, which represents the difference between the actual deviation and the simulated reference value, Update the shape deviation parameter Δd and the clamping force compensation parameter F c , the update formula is: Δd'=Δd+λΔd v , F' c =F c +μΔd v , Among them, Δd' is the updated morphological deviation parameter, Δd is the original morphological deviation parameter, λ is the updated weight coefficient, F' c is the updated clamping force compensation parameter, F c is the original clamping force compensation parameter, and μ is the clamping force compensation adjustment coefficient.

Images