Method for automatic construction of robot multi-body simulation model and related device

Abstract

The application discloses a kind of automatic construction method of robot multi-body simulation model and related equipment, it is related to robot simulation modeling technical field, the method includes: obtaining robot description file, to obtain intermediate representation data structure by parsing description file;To the physical constraint parameter in intermediate representation data structure and kinematics structure perform validity check, and when identifying repairable defect, to intermediate representation data structure corresponding modification is carried out, obtains standardized intermediate representation;Based on standardized intermediate representation, for each joint corresponding joint component is generated, when joint has physical constraint parameter, activate the physical constraint connection port reserved on joint component, and generate the physical constraint component corresponding to physical constraint parameter, connect physical constraint component to physical constraint connection port;All joint components, physical constraint components and connecting rod components generated based on standardized intermediate representation are connected, and the robot multi-body dynamics simulation model is output.

Description

Technical Field

[0001] This application relates to the field of robot simulation modeling technology, and in particular to an automatic construction method and related equipment for multi-body robot simulation models. Background Technology

[0002] In the robot development process, engineers typically need to build corresponding multibody dynamics simulation models based on robot description files such as the unified robot description format used by the robot operating system, the simulation description format used by the simulator, or the multi-joint contact dynamics format used by the multi-joint contact dynamics simulation platform, in order to verify the robot's kinematic and dynamic behavior.

[0003] In related technologies, when converting the aforementioned robot description file into a multibody dynamics simulation model, often only a simple mapping of the kinematic skeleton is performed, i.e., only the pose of the links and the type of joints are converted, while the physical constraint parameters contained in the description file, such as joint range of motion limits, damping coefficients, and friction coefficients, are completely ignored. The resulting simulation model does not generate stopping forces or damping effects when joints exceed their physical travel, and the simulation results deviate significantly from the behavior of real robots, resulting in insufficient physical fidelity.

[0004] Furthermore, the conversion tools in related technologies do not perform physical validity checks on the input description files. In practical engineering, robot description files often contain problems such as virtual links with zero mass, non-positive definite inertial tensors, and inverted upper and lower limits of joint constraints. The simulation models obtained by direct conversion are prone to initialization failure or divergence during solution processing, resulting in poor reliability of the conversion results. Although some improved solutions attempt to support multi-format input, when dealing with floating bases or planar moving bases, users still need to manually supplement constraint relationships or repeat modeling multiple times, resulting in limited automation and generalization capabilities. Summary of the Invention

[0005] The main purpose of this application is to provide an automatic construction method and related equipment for robot multibody simulation models, aiming to solve the technical problem of poor reliability in simulation model conversion in related technologies.

[0006] To achieve the above objectives, this application proposes an automatic construction method for a robot multibody simulation model, the method comprising: Obtain a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The physical constraint parameters and kinematic structure in the intermediate representation data structure are validated for legality. When a repairable defect of a preset type is identified, the intermediate representation data structure is automatically corrected accordingly to obtain a normalized intermediate representation. Based on the standardized intermediate representation, a corresponding joint component is generated for each joint. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. Connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0007] In one embodiment, the steps of performing legality checks on the physical constraint parameters and kinematic structures in the intermediate representation data structure, and performing corresponding automatic corrections on the intermediate representation data structure when a pre-defined type of repairable defect is identified, to obtain a normalized intermediate representation, include: Based on the link and joint information in the kinematic structure, an adjacency relationship is constructed. The adjacency relationship is traversed to detect loops, isolated nodes, and multiple nodes, and to verify that the parent and child links referenced by each joint are recorded in the link information. For each link in the link information, determine whether its mass is positive, and construct a matrix based on its inertia tensor to determine whether all eigenvalues ​​of the matrix are positive. For the physical constraint parameters, determine whether the upper limit of the rotation joint and the lower limit of the traverse joint are greater than the lower limit; For a follower joint marked with a follower relationship in the joint information, it is determined that the master joint it references exists in the joint information, and tracing along the follower relationship starting from the master joint will not return to the follower joint; When a repairable defect of a preset type is identified, the corresponding automatic correction is performed. The automatic correction includes correcting non-positive mass values ​​to preset small positive values, correcting non-positive definite inertial tensors to positive definite matrices, swapping the inverted upper and lower limits of the limit, and taking the absolute values ​​of negative damping coefficients and friction coefficients to obtain a normalized intermediate representation.

[0008] In one embodiment, the step of generating a corresponding joint component for each joint based on the normalized intermediate representation, activating the reserved physical constraint connection port on the joint component when the joint has the physical constraint parameters, automatically generating a physical constraint component corresponding to the physical constraint parameters, and connecting the physical constraint component to the physical constraint connection port includes: When the physical constraint parameters of the joint include a range of motion limitation, an elastic stop component is generated. The difference between the upper and lower limits of the range of motion limitation is determined as the stroke of the elastic stop component, and the quotient of the preset maximum joint torque and the preset over-limit allowable amount is determined as the stiffness of the elastic stop component. When the physical constraint parameters of the joint include a damping coefficient, a damping component is generated, and the damping coefficient is assigned to the damping parameter of the damping component. When the physical constraint parameters of the joint include the friction coefficient, a friction component is generated, and the friction coefficient is assigned to the friction parameters of the friction component. At least one of the elastic stop assembly, the damping assembly, or the friction assembly is used as the physical constraint assembly and connected to the physical constraint connection port of the joint assembly.

[0009] In one embodiment, the step of obtaining a robot description file in at least one format and parsing the description file to obtain a unified intermediate representation data structure, wherein the intermediate representation data structure includes link and joint information describing the robot's kinematic structure and physical constraint parameters attached to the joints, includes: Read the robot description file and determine the format type based on the file extension or root tag; When the format type is the Unified Robot Description Format, the preset Unified Format Parser is called to extract the mass, center of mass position and inertia tensor from the link label, and the joint type, rotation axis direction, limit parameter and damping friction coefficient from the joint label. When the format type is simulation description format, the preset simulation format adapter is called to convert the six-element pose label into position and attitude parameters, and the limit parameters nested in the axis label and limit label are mapped to a structure consistent with the unified format; When the format type is a multi-joint contact dynamics format, the preset multi-joint format adapter is called to flatten the recursively nested tree-shaped body structure into a link list, create fixed joints for the jointless sub-body, and convert the angle-based limit parameters into radians. The parsed or adapted information is assembled into the unified intermediate representation data structure.

[0010] In one embodiment, after the step of connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: A closed-loop control loop is generated for each active joint. The closed-loop control loop includes a preset target position constant source, an angle or displacement sensor, a feedback differential, a preset proportional-integral-derivative controller, and a torque or force actuator. The mass, center of mass position, and inertia tensor of the linkage driven by the movable joint are obtained. The equivalent inertia of the rotation axis of the movable joint is calculated according to the parallel axis theorem. Based on the equivalent inertia, the proportional gain, integral gain, and derivative gain of the proportional-integral-derivative controller are determined according to the preset target natural frequency and damping ratio. Based on the limiting range of the movable joint, a trajectory planning channel based on a time schedule is automatically generated. The start and end points of the trajectory planning channel are automatically set according to the limiting range, and a smooth trajectory curve is generated using continuous differentiable interpolation. Connect the torque or force actuator in the closed-loop control loop to the corresponding port of the joint assembly, and output a complete robot multibody dynamics simulation model containing the closed-loop control loop.

[0011] In one embodiment, after the step of connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: The intermediate representation data structure contains the number of links, joints, and degrees of freedom, as well as the number of driven joints with a follower relationship. Generate a link parameter mapping table, recording line by line the original name of each link in the robot description file and the corresponding part sub-model name, mass value, centroid coordinates, and inertia tensor in the simulation model; Generate a joint parameter mapping table, recording the original name of each joint and its corresponding joint type, limit range, damping coefficient, friction coefficient, and follower relationship line by line; Based on the kinematic structure, a tree-like topology path diagram starting from the root link is generated, and all verification issues and corresponding automatic repair operation records are summarized. The link parameter mapping table, the joint parameter mapping table, the tree topology path diagram, and the automatic repair operation record are organized into a preset format Hypertext Markup Language conversion report file.

[0012] In one embodiment, the step of generating a corresponding joint component for each joint based on the normalized intermediate representation further includes: When the type of the joint is identified as a floating joint or a free joint, a six-degree-of-freedom spatial motion component is generated, which includes three translational degrees of freedom and three rotational degrees of freedom. When the type of the joint is identified as a planar motion joint, a planar motion constraint component is generated, which restricts the planar motion constraint component to two translational degrees of freedom and one rotational degree of freedom in the plane. Based on the initial pose of the floating joint or free joint in the normalized intermediate representation, set the initial condition parameters of the six-degree-of-freedom spatial motion component or the planar motion constraint component; The six-degree-of-freedom spatial motion component or the planar motion constraint component is used as the joint component, replacing the default fixed base connection.

[0013] Furthermore, to achieve the above objectives, this application also proposes an automatic construction device for a robot multi-body simulation model, the automatic construction device for the robot multi-body simulation model comprising: The acquisition module is used to acquire a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The correction module is used to perform legality verification on the physical constraint parameters and kinematic structure in the intermediate representation data structure, and when a repairable defect of a preset type is identified, to perform corresponding automatic correction on the intermediate representation data structure to obtain a normalized intermediate representation. The constraint module is used to generate a corresponding joint component for each joint based on the normalized intermediate representation. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. The output module is used to connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0014] In addition, to achieve the above objectives, this application also proposes an automatic construction device for a robot multibody simulation model, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic construction method for the robot multibody simulation model as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the automatic construction method for the robot multibody simulation model as described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: Compared to related technologies where conversion tools fail to perform physical validity checks on input files and only perform simple mapping of kinematic skeletons, resulting in poor reliability of conversion results, this application obtains a robot description file in at least one format, parses the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure includes link and joint information describing the robot's kinematic structure, and physical constraint parameters attached to the joints; performs validity checks on the physical constraint parameters and kinematic structure in the intermediate representation data structure, and when a pre-defined type of repairable defect is identified, performs corresponding automatic correction on the intermediate representation data structure to obtain a standardized intermediate representation; based on the standardized intermediate representation, generates a corresponding joint component for each joint; when the joint has the physical constraint parameters, activates the reserved physical constraint connection port on the joint component, and automatically generates a physical constraint component corresponding to the physical constraint parameters, connecting the physical constraint component to the physical constraint connection port; connects all joint components, physical constraint components, and link components generated based on the standardized intermediate representation to output a complete robot multibody dynamics simulation model. Understandably, when robot description files in various formats are obtained, the input data is standardized by uniformly parsing it into an intermediate representation data structure containing physical constraint parameters. This enables automatic adaptation of multi-format inputs. Therefore, based on this standardized intermediate representation, physical validity checks and automatic corrections can be uniformly performed, thereby determining that the input data before conversion meets the physical rationality requirements. Finally, a high-fidelity simulation model containing complete physical constraints is automatically constructed, solving the technical problem of poor conversion reliability in related technologies. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

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

[0019] Figure 1 A flowchart illustrating an embodiment of the automatic construction method for a robot multibody simulation model according to this application; Figure 2 A flowchart illustrating the second embodiment of the automatic construction method for the robot multibody simulation model of this application; Figure 3A flowchart illustrating the third embodiment of the automatic construction method for robot multibody simulation models in this application; Figure 4 This is a schematic diagram of the module structure of the automatic construction device for robot multibody simulation models according to an embodiment of this application; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the automatic construction method of the robot multibody simulation model in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution in this application embodiment is: Obtain a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The physical constraint parameters and kinematic structure in the intermediate representation data structure are validated for legality. When a repairable defect of a preset type is identified, the intermediate representation data structure is automatically corrected accordingly to obtain a normalized intermediate representation. Based on the standardized intermediate representation, a corresponding joint component is generated for each joint. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. Connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0024] In this embodiment, the automatic construction device for robot multi-body simulation model is used as the execution subject. For ease of description, it will be referred to as "device" in detail below.

[0025] Because the conversion tools in related technologies do not perform physical validity checks on the input files, and only retain the kinematic skeleton while discarding physical constraint parameters such as joint limits, damping, and friction during processing, the physical fidelity of the generated simulation model is insufficient. The solver is prone to initialization failure or divergence when encountering common engineering defects such as zero-mass links, non-positive definite inertial tensors, or inverted limits, resulting in poor reliability of the conversion results.

[0026] This application provides a solution that establishes a closed-loop processing chain for the conversion of robot description files, encompassing "format parsing and standardization—physical validity verification and automatic repair—high-fidelity physical constraint mapping—automatic assembly of multibody models." In the format parsing stage, description files from different sources are uniformly extracted into intermediate representations containing kinematic structures and physical constraint parameters, eliminating the impact of format differences on subsequent processing. In the verification stage, multi-dimensional validity checks are performed on the topology, inertial parameters, joint limits, and follower dependencies in the intermediate representation, and repairable defects are automatically corrected to ensure that the intermediate representation entering the generation stage is physically self-consistent, thereby preventing solution failures from the outset. In the mapping stage, corresponding elastic stop components, damping components, and friction components are automatically generated for the limit, damping, and friction parameters of each joint, and the reserved physical constraint connection ports on the joint components are activated to complete the component connection, enabling the simulation model to realistically reproduce the stopping force at the joint's extreme positions and the damping friction effect during motion. In the assembly stage, all link components, joint components, and physical constraint components are automatically connected according to kinematic topological relationships, and a complete model is output. Therefore, this application automates all the steps that originally required repeated manual verification and configuration, eliminating format barriers while significantly improving the reliability of the conversion results and the physical fidelity of the generated model.

[0027] Based on this, embodiments of this application provide an automatic construction method for a robot multibody simulation model, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the automatic construction method for robot multibody simulation models according to this application.

[0028] In this embodiment, the automatic construction method of the robot multi-body simulation model includes steps S10 to S40: Step S10: Obtain a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; It should be noted that the robot description file includes, but is not limited to, structured description files using a unified robot description format, simulation description format, or multi-joint contact dynamics format. The format refers to the type of tag system and data organization within the file, which can be distinguished by the file extension or root tag. Parsing refers to the process of extracting tag attribute values ​​from the description file and converting them into structured data. The unified intermediate representation data structure is a predefined in-memory data model containing at least one list of link information and at least one list of joint information. The link information includes, but is not limited to, link name, mass, center of mass position, and inertia tensor. The joint information includes, but is not limited to, joint name, type, parent link identifier, child link identifier, rotation axis direction, coordinate transformation parameters, and physical constraint parameters. The kinematic structure refers to the robot topological connection relationship jointly described by the link information and joint information. The physical constraint parameters attached to the joints include, but are not limited to, range of motion limits, maximum torque, maximum velocity, damping coefficient, and friction coefficient.

[0029] Understandably, this step automatically identifies and uniformly parses robot description files of different formats, converting the originally heterogeneous input data into a structurally consistent intermediate representation, eliminating the impact of format differences on subsequent processing, and providing a standardized data foundation for subsequent verification and model generation.

[0030] Step S20: Perform a legality check on the physical constraint parameters and kinematic structure in the intermediate representation data structure, and when a repairable defect of a preset type is identified, perform a corresponding automatic correction on the intermediate representation data structure to obtain a normalized intermediate representation; It should be noted that the legality verification includes, but is not limited to, loop detection, isolated node detection, multi-node detection, and existence verification of parent and child links referenced by joints in the kinematic structure; determining whether the mass of each link is positive; determining whether all eigenvalues ​​of the matrix constructed by the inertia tensor are positive; determining whether the upper limit of the limit of the rotational and translating joints is greater than the lower limit; and determining whether the master joint referenced by the driven joint marked with a follower relationship exists and is free from cyclic dependence. The preset types of repairable defects include, but are not limited to, non-positive mass, non-positive definite inertia tensor, inverted upper and lower limits of the limit, and negative damping or friction coefficients. The automatic correction refers to automatically changing the identified repairable defects according to a preset correction strategy. The preset correction strategy includes, but is not limited to, setting non-positive mass to a preset small positive value, taking the absolute value of the diagonal elements of the non-positive definite inertia tensor and ensuring it is not less than a preset lower limit, swapping inverted upper and lower limits of the limit, and taking the absolute value of negative damping and friction coefficients. The standardized intermediate representation refers to the intermediate representation data structure in which all physical constraint parameters and kinematic structures meet the preset physical rationality conditions after the legality verification and automatic correction.

[0031] Understandably, this step, by embedding active verification and automatic repair mechanisms in the conversion process, can identify and correct physical inconsistencies in the input data before generating the simulation model, making the intermediate representations entering the subsequent generation stage physically self-consistent. This effectively avoids solver initialization failures or simulation divergence caused by input data defects, significantly improving the reliability of the conversion results.

[0032] Step S30: Based on the normalized intermediate representation, generate a corresponding joint component for each joint. When the joint has the physical constraint parameters, activate the reserved physical constraint connection port on the joint component, automatically generate a physical constraint component corresponding to the physical constraint parameters, and connect the physical constraint component to the physical constraint connection port. It should be noted that the joint assembly refers to a standard dynamic component corresponding to the joint type in the joint information, including but not limited to rotational joint assemblies and translational joint assemblies. The physical constraint connection port refers to a pre-set interface on the joint assembly for connecting one-dimensional mechanical components, which is activated when a physical constraint is detected as needing to be added. The physical constraint components include, but are not limited to, elastic stop components, damping components, and friction components. The elastic stop component is used to provide elastic restoring torque or restoring force when the joint angle or displacement exceeds the motion range limit. Its stroke parameter is determined based on the difference between the upper and lower limits of the motion range limit, and its stiffness parameter is determined based on the quotient of a pre-set maximum joint torque and a pre-set allowable over-limit amount. The damping parameter of the damping component is assigned a value based on the damping coefficient. The friction parameter of the friction component is assigned a value based on the friction coefficient.

[0033] Understandably, this step automatically identifies the physical constraint parameters carried by the joint, automatically generates the corresponding physical constraint components and connects them to the joint's mechanical interface, so that the final simulation model can realistically reproduce the stopping effect of the joint when it reaches the limit and the damping and friction effects during the movement, thereby greatly improving the physical fidelity of the simulation model.

[0034] Step S40: Connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0035] It should be noted that connecting all joint components, physical constraint components, and link components refers to automatically generating connection statements between the interfaces of each component in the model assembly environment based on the parent-child relationships of the joints in the standardized intermediate representation, so that all components form an overall model consistent with the kinematic structure. The complete robot multibody dynamics simulation model refers to a model package containing all link components, joint components, and physical constraint components, which can be directly loaded by the multibody dynamics solver and used for simulation calculations.

[0036] Understandably, this step assembles all the components produced in the previous steps into a ready-to-use simulation model by automatically completing component instantiation, layout arrangement, and connection generation. This eliminates the tedious process of manually creating components, setting parameters, and connecting them one by one, and realizes fully automated construction from robot description files in various formats to a solvable complete multibody dynamics simulation model.

[0037] This embodiment provides an automatic construction method for a robot multibody simulation model. When multiple robot description files in different formats are obtained, the input data is standardized by uniformly parsing it into an intermediate representation data structure containing physical constraint parameters. This enables automatic adaptation of multi-format inputs. Therefore, based on this standardized intermediate representation, physical validity verification and automatic correction can be uniformly performed to determine whether the input data before conversion meets the physical rationality requirements. Finally, an automatic construction of a high-fidelity simulation model containing complete physical constraints is completed, solving the technical problem of poor conversion reliability in related technologies.

[0038] In one feasible implementation, the steps of performing legality checks on the physical constraint parameters and kinematic structures in the intermediate representation data structure, and automatically correcting the intermediate representation data structure when a pre-defined type of repairable defect is identified to obtain a normalized intermediate representation include: Based on the link and joint information in the kinematic structure, an adjacency relationship is constructed. The adjacency relationship is traversed to detect loops, isolated nodes, and multiple nodes, and to verify that the parent and child links referenced by each joint are recorded in the link information. For each link in the link information, determine whether its mass is positive, and construct a matrix based on its inertia tensor to determine whether all eigenvalues ​​of the matrix are positive. For the physical constraint parameters, determine whether the upper limit of the rotation joint and the lower limit of the traverse joint are greater than the lower limit; For a follower joint marked with a follower relationship in the joint information, it is determined that the master joint it references exists in the joint information, and tracing along the follower relationship starting from the master joint will not return to the follower joint; When a repairable defect of a preset type is identified, the corresponding automatic correction is performed. The automatic correction includes correcting non-positive mass values ​​to preset small positive values, correcting non-positive definite inertial tensors to positive definite matrices, swapping the inverted upper and lower limits of the limit, and taking the absolute values ​​of negative damping coefficients and friction coefficients to obtain a normalized intermediate representation.

[0039] It should be noted that the adjacency relationship refers to a graph structure constructed with links as nodes and joints as edges, used to characterize the connection relationship between links in the kinematic structure. A loop refers to a closed path that starts from a link, traverses along the adjacency relationship, and returns to that link. An isolated node refers to a link that has no joint connection with any other link. Multiple root nodes refer to at least two root links without parent links in the adjacency relationship. The follower relationship refers to the joint motion following attribute marked in the joint information, used to describe the preset proportional and offset relationship between the angle or displacement of the follower joint and the angle or displacement of the parent joint. The follower relationship includes, but is not limited to, joint following constraints defined by the mimicry tag in the unified robot description format. A cyclic dependency refers to a closed dependency link formed by starting from a follower joint, tracing the parent joint along the follower relationship, and returning to the follower joint after at least one tracing. The preset small positive value includes, but is not limited to, 10 to the power of -6 kilograms. The positive definite matrix refers to a symmetric matrix where all eigenvalues ​​are positive.

[0040] It is understandable that this implementation method can systematically cover common physical inconsistencies in input data by performing targeted verifications on four dimensions: kinematic topology, inertial parameters, joint constraints, and follower relationships. Instead of simply reporting errors and stopping the process after a problem is detected, the method automatically triggers a repair process. This allows the entire conversion process to continue running and produce a usable, standardized intermediate representation even when faced with input defects that are common in engineering practice, providing a physically self-consistent data foundation for subsequent steps.

[0041] In one feasible implementation, the step of generating a corresponding joint component for each joint based on the normalized intermediate representation, activating the reserved physical constraint connection port on the joint component when the joint has the physical constraint parameters, automatically generating a physical constraint component corresponding to the physical constraint parameters, and connecting the physical constraint component to the physical constraint connection port includes: When the physical constraint parameters of the joint include a range of motion limitation, an elastic stop component is generated. The difference between the upper and lower limits of the range of motion limitation is determined as the stroke of the elastic stop component, and the quotient of the preset maximum joint torque and the preset over-limit allowable amount is determined as the stiffness of the elastic stop component. When the physical constraint parameters of the joint include a damping coefficient, a damping component is generated, and the damping coefficient is assigned to the damping parameter of the damping component. When the physical constraint parameters of the joint include the friction coefficient, a friction component is generated, and the friction coefficient is assigned to the friction parameters of the friction component. At least one of the elastic stop assembly, the damping assembly, or the friction assembly is used as the physical constraint assembly and connected to the physical constraint connection port of the joint assembly.

[0042] It should be noted that the elastic stop component refers to a physical constraint component that generates a restoring torque or restoring force proportional to the excess amount when the joint angle or displacement exceeds a preset range. Its stroke parameter refers to the range of free joint movement, determined by the difference between the upper and lower limits of the movement range. Its stiffness parameter refers to the magnitude of the restoring torque or restoring force corresponding to a unit excess amount, determined by the quotient of a preset maximum joint torque and a preset excess allowable amount. The preset maximum joint torque includes, but is not limited to, the maximum torque value extracted from the limit label in the robot description file. The preset excess allowable amount includes, but is not limited to, 0.01 radians or 0.01 meters. The damping component refers to a physical constraint component that generates a resistance torque or resistance proportional to the joint movement speed. Its damping parameter includes, but is not limited to, the damping coefficient. The friction component refers to a physical constraint component that generates a constant friction torque or friction force related to the direction of movement but independent of the speed magnitude. Its friction parameter includes, but is not limited to, the Coulomb coefficient of friction. The assignment refers to directly writing the corresponding parameter values ​​extracted from the normalized intermediate representation into the corresponding attribute fields of the generated component.

[0043] It is understood that this implementation generates dedicated physical constraint components for different types of physical constraint parameters, and sets the working parameters of each component according to the actual parameter values ​​in the standardized intermediate representation. This ensures that the limiting and stopping characteristics, damping characteristics, and friction characteristics of each joint in the final model correspond one-to-one with the definitions in the original description file. As a result, the physical behavior of the robot under various working conditions is realistically reproduced in the simulation calculation, avoiding deviations between the simulation results and reality due to the neglect of physical constraints, and improving the physical fidelity and credibility of the simulation model.

[0044] In one feasible implementation, the steps of obtaining a robot description file in at least one format, parsing the description file to obtain a unified intermediate representation data structure, wherein the intermediate representation data structure includes link and joint information describing the robot's kinematic structure, and physical constraint parameters attached to the joints, include: Read the robot description file and determine the format type based on the file extension or root tag; When the format type is the Unified Robot Description Format, the preset Unified Format Parser is called to extract the mass, center of mass position and inertia tensor from the link label, and the joint type, rotation axis direction, limit parameter and damping friction coefficient from the joint label. When the format type is simulation description format, the preset simulation format adapter is called to convert the six-element pose label into position and attitude parameters, and the limit parameters nested in the axis label and limit label are mapped to a structure consistent with the unified format; When the format type is a multi-joint contact dynamics format, the preset multi-joint format adapter is called to flatten the recursively nested tree-shaped body structure into a link list, create fixed joints for the jointless sub-body, and convert the angle-based limit parameters into radians. The parsed or adapted information is assembled into the unified intermediate representation data structure.

[0045] It should be noted that the file extension refers to the suffix string after the last period in the robot description file name, and the root tag refers to the document root element name defined by the outermost pair of start and end tags of the robot description file. The unified robot description format is the Extensible Markup Language (XML) format used in robot operating systems to describe the kinematic structure and physical properties of robots. The simulation description format is the XML format used in simulators. The multi-joint contact dynamics format is the XML format used in multi-joint contact dynamics simulation platforms. The six-element pose tag refers to a pose description field containing six space-separated values, where the first three values ​​represent translational position and the last three values ​​represent rotational attitude. The position and attitude parameters refer to pose information represented as translation vectors and rotation Euler angles or quaternions, respectively. The recursively nested tree-like body structure refers to a tree-like hierarchical organization in the multi-joint contact dynamics format where body tags can be nested layer by layer into child body tags. The flattening refers to expanding the tree-like body structure into a single, parallel list of links while preserving the parent-child relationships between bodies in the original tree structure. The angle-limiting parameter refers to the angle value of the joint movement range in degrees, and the radian value refers to the radian value obtained by multiplying the angle value by the ratio of pi to 180 degrees.

[0046] Understandably, this implementation achieves automatic identification of multiple formats through a two-layer recognition mechanism using file extensions and root tags, and configures a dedicated parsing or adaptation module for each format. This enables the unified conversion of input data from different sources and with different organization methods into a consistent intermediate representation data structure. For the split mapping of the six-element pose label in the simulation description format, it resolves the difference in pose representation between this format and the unified robot description format. For the flattening of the tree-like body structure and the conversion from angle to radian in the multi-joint contact dynamics format, it resolves the incompatibility issues between its nesting hierarchy and unit system with other formats. Therefore, subsequent verification and generation processing logic does not need to be aware of the differences in the original input format and can all be executed based on the same intermediate representation data structure, thus achieving generalized support for multi-format inputs.

[0047] In one feasible implementation, after the step of connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: A closed-loop control loop is generated for each active joint. The closed-loop control loop includes a preset target position constant source, an angle or displacement sensor, a feedback differential, a preset proportional-integral-derivative controller, and a torque or force actuator. The mass, center of mass position, and inertia tensor of the linkage driven by the movable joint are obtained. The equivalent inertia of the rotation axis of the movable joint is calculated according to the parallel axis theorem. Based on the equivalent inertia, the proportional gain, integral gain, and derivative gain of the proportional-integral-derivative controller are determined according to the preset target natural frequency and damping ratio. Based on the limiting range of the movable joint, a trajectory planning channel based on a time schedule is automatically generated. The start and end points of the trajectory planning channel are automatically set according to the limiting range, and a smooth trajectory curve is generated using continuous differentiable interpolation. Connect the torque or force actuator in the closed-loop control loop to the corresponding port of the joint assembly, and output a complete robot multibody dynamics simulation model containing the closed-loop control loop.

[0048] It should be noted that the "movable joint" refers to the independent drive joint remaining after excluding fixed joints and driven joints with a follower relationship. The "closed-loop control loop" refers to a feedback control system composed of sensors, controllers, and actuators, used to make the actual position or angle of the joint track a preset target position. The "preset target position constant source" refers to a signal source component that outputs a constant target position signal. The "angle or displacement sensor" refers to a sensing component used to measure the current angle or displacement of the joint in real time and feed the measured value back to the differential. The "feedback differential" refers to a computational component used to calculate the difference between the target position and the current actual position fed back by the sensor. The "preset proportional-integral-derivative controller" refers to a linear controller component containing three adjustable parameters: proportional gain, integral gain, and derivative gain, whose output is torque or force command. The "torque or force actuator" refers to a drive component that receives the torque or force command output by the controller and applies it to the joint. The parallel axis theorem states that the moment of inertia of a rigid body about any axis parallel to its center of mass is equal to the sum of its moment of inertia about the center of mass and the mass of the rigid body multiplied by the square of the distance between the two axes. The equivalent inertia refers to the equivalent moment of inertia obtained by converting the inertia tensor and center of mass position of the link onto the joint axis using the parallel axis theorem. The preset target natural frequency refers to the natural frequency of the closed-loop system preset according to the desired response speed of the robot joint. The damping ratio refers to the desired damping characteristic ratio of the closed-loop system. The determination based on the equivalent inertia refers to calculating the proportional gain, integral gain, and differential gain by substituting the equivalent inertia, the target natural frequency, and the damping ratio into a preset parameter tuning formula. The trajectory planning channel based on the time schedule refers to the signal generation link that generates corresponding position commands according to a preset time node sequence. The continuous differentiable interpolation refers to the interpolation algorithm that is continuous in both position and velocity between adjacent time nodes. The smooth trajectory curve refers to a motion trajectory without abrupt changes in position, velocity, and acceleration.

[0049] It is understood that this implementation method automatically generates a complete closed-loop control loop for each independent moving joint after the model is assembled, and automatically tunes the controller gain according to the inertial parameters of the link. This results in a simulation model that not only includes the mechanical body and physical constraints, but also has directly operable joint servo control capabilities and automatically planned reference trajectories. Users can directly perform closed-loop dynamic simulation verification without having to manually design the controller structure and repeatedly debug the gain parameters, which greatly shortens the time cycle from model import to simulation readiness.

[0050] In one feasible implementation, after the step of connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: The intermediate representation data structure contains the number of links, joints, and degrees of freedom, as well as the number of driven joints with a follower relationship. Generate a link parameter mapping table, recording line by line the original name of each link in the robot description file and the corresponding part sub-model name, mass value, centroid coordinates, and inertia tensor in the simulation model; Generate a joint parameter mapping table, recording the original name of each joint and its corresponding joint type, limit range, damping coefficient, friction coefficient, and follower relationship line by line; Based on the kinematic structure, a tree-like topology path diagram starting from the root link is generated, and all verification issues and corresponding automatic repair operation records are summarized. The link parameter mapping table, the joint parameter mapping table, the tree topology path diagram, and the automatic repair operation record are organized into a preset format Hypertext Markup Language conversion report file.

[0051] It should be noted that the number of degrees of freedom refers to the total number of independent motion directions of all movable joints after excluding fixed joints. The driven joint refers to a joint whose motion is constrained by other joints through a driven relationship, and its motion is determined by the motion of the master joint according to a preset ratio and offset. The link parameter mapping table refers to a structured record in tabular form that establishes the parameter correspondence between the original links in the input file and the part sub-models in the output model. Each row of the record includes, but is not limited to, the original name, the corresponding part sub-model name, the mass value, the components of the center of mass on the three coordinate axes, and the component values ​​of the inertia tensor. The joint parameter mapping table refers to a structured record in tabular form that establishes the parameter correspondence between the original joints in the input file and the joint components in the output model. Each row of the record includes, but is not limited to, the original name, the corresponding joint type, the lower limit value, the upper limit value, the damping coefficient, the friction coefficient, and the name of the referenced master joint. The tree-like topology path diagram refers to the hierarchical path text representation formed by expanding layer by layer to all end links according to the parent-child connection relationship of the joints in the kinematic structure, with the root link as the starting node. The root link refers to a link that does not have a parent joint in the kinematic structure. The verification and repair record refers to a summary of all issues identified in the verification steps and their corresponding automatic repair operations. Each record includes, but is not limited to, the verification dimension to which the issue belongs, the issue description, and the repair strategy executed. The preset format Hypertext Markup Language conversion report file refers to an interactive report document that can be directly viewed in a browser, organized according to a predefined page structure and style specifications. This report document organizes the link parameter mapping table, the joint parameter mapping table, the tree topology path diagram, and the automatic repair operation record in their respective page areas.

[0052] Understandably, this implementation method automatically summarizes and outputs a conversion report containing parameter mapping details, topology view, and a complete record of verification and repair after the model is generated. This allows users to fully trace the correspondence between every parameter and component from the original robot description file to the final simulation model. It provides auditable documentation for engineering verification, model review, and problem troubleshooting, eliminating the problem of unclear parameter sources due to lack of records in traditional conversion methods, and improving the traceability and credibility of the model conversion process.

[0053] In one feasible implementation, the step of generating a corresponding joint component for each joint based on the normalized intermediate representation further includes: When the type of the joint is identified as a floating joint or a free joint, a six-degree-of-freedom spatial motion component is generated, which includes three translational degrees of freedom and three rotational degrees of freedom. When the type of the joint is identified as a planar motion joint, a planar motion constraint component is generated, which restricts the planar motion constraint component to two translational degrees of freedom and one rotational degree of freedom in the plane. Based on the initial pose of the floating joint or free joint in the normalized intermediate representation, set the initial condition parameters of the six-degree-of-freedom spatial motion component or the planar motion constraint component; The six-degree-of-freedom spatial motion component or the planar motion constraint component is used as the joint component, replacing the default fixed base connection.

[0054] It should be noted that the floating joint or free joint refers to the base joint type marked as having no fixed constraints in the robot description file, including but not limited to floating joints in the Unified Robot Description Format and free joints in the Multi-Joint Contact Dynamics Format. Such joints allow their connected links to perform six degrees of freedom motion in three-dimensional space. The six-degree-of-freedom spatial motion component refers to a mechanical component used in a simulation environment to describe an object completely unrestrained by kinematics in space. This component provides translational degrees of freedom along three orthogonal coordinate axes and rotational degrees of freedom about three orthogonal coordinate axes, and its pose is determined by the translational displacement and rotational angle. The planar motion constraint component refers to a constraint component that restricts the motion of an object to a preset plane. This component provides translational degrees of freedom along two orthogonal directions within the preset plane and one rotational degree of freedom about the normal direction of the preset plane, for a total of three degrees of freedom. The initial condition parameters include, but are not limited to, the components of the initial position on the three coordinate axes, the Euler angles or quaternion representation of the initial attitude, and the initial velocity and initial angular velocity. The default fixed base connection refers to the method of connecting the robot's root link to the origin of the world coordinate system in the simulation environment through fixed constraints. This method constrains all six degrees of freedom of the root link to zero.

[0055] Understandably, this implementation method automatically generates spatial motion components or planar motion constraint components that match the motion characteristics of the joints representing unconstrained bases in the input file, rather than indiscriminately using fixed base connections. This ensures that the generated simulation model can accurately represent the kinematic characteristics of floating-base robots and planar mobile robots. For applications with floating bases, such as humanoid robots or quadruped robots, the introduction of six-DOF spatial motion components allows the robot as a whole to move freely in space. For applications of planar mobile robots, such as automated guided vehicles, planar motion constraint components retain the chassis's in-plane motion capability while constraining irrelevant out-of-plane degrees of freedom, avoiding deviations from physical reality during simulation. Thus, this implementation method significantly expands the generalization support capability of the conversion method for different types of robot platforms.

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0057] Reference Figure 2 , Figure 2 This is a system architecture diagram provided for embodiments of this application. The system architecture adopts a top-down layered process structure: the top layer is the input source layer, containing three parallel input channels: a unified robot description format file, a simulation description format file, and a multi-joint contact dynamics format file; the second layer is the automatic format detection and adapter layer, which distributes input files to corresponding format parsers through dual determination of file extension and Extensible Markup Language (XML) root tags, with the outputs of each parser converging into a unified intermediate representation data structure; the third layer is the semantic verification and repair layer, containing five sub-modules: kinematic topology verification, inertial parameter verification, joint limit verification, follower relationship verification, and geometric path verification, with the verification engine connected to the automatic repair module; the fourth layer is the model generation layer, containing three parallel branches: a link sub-model generator, a joint and physical constraint generator, and a follower control chain generator, as well as a lower-level intelligent layout engine; the fifth layer is the controller generation layer; and the bottom layer is the output layer, outputting the model simulation model package and the XML conversion report. Solid arrows between modules in the diagram indicate the order of data flow, while dashed arrows indicate traceability information flow paths.

[0058] This embodiment provides an automatic construction method for robot multibody simulation models. The method includes four stages in its overall process: acquisition and parsing, verification and repair, physical constraint mapping and generation, and model assembly and output. The specific implementation of each stage is described below.

[0059] During the acquisition and parsing phase, the system reads the robot description file and automatically determines the format type based on the file extension or root tag: files with the suffix ".urdf" or those with the root tag "".urdf". <robot>The criteria for "" are based on the Unified Robot Description Format (URDF), which has the suffix ".sdf" or the root tag "". <sdf>The determination is that it belongs to Simulation Description Format (SDF), and the root tag is "". <mujoco>The determination is that it is a multi-joint dynamics with contact (MJCF) format. For the URDF format, the parser extracts... <link> The link name in the label <inertial>The sub-label contains six components: mass, centroid position, and inertia tensor. <visual>Visual geometric information in sub-tags and <collision>Collision geometry information in sub-tags, and <joint>The tags should include the joint name, type (supporting revolve, prismatic, continuous, fixed, floating, and planar), parent-child links, coordinate transformation, and rotation axis direction. <limit>The upper and lower limits of the limit switch, the maximum torque, and the maximum speed are all within the range. <dynamics>The damping coefficient and friction coefficient in <mimic>The follower relationship parameters in the code. For SDF format, the adapter will... <pose>The six-element pose string of the label is split into two sets of parameters: position and orientation, and then nested within... <axis>Within the label <limit>Limit parameters are mapped to a structure consistent with URDF. For the MJCF format, the adapter traverses the recursively nested tree structure and flattens it into a list of links. Fixed joints are created for jointless sub-bodies, hinge types are mapped to revolute joints, slide types to prismatic joints, and free types to floating joints. Angle-based limit parameters are converted to radians. After all format parsing or adaptation is complete, the extracted information is assembled into a unified RobotIR data structure. This data structure contains a list of links and a list of joints. Each element in the links list contains an inertial attribute, a list of visualizations, and a list of collisions. Each element in the joints list contains a limit attribute, a dynamics attribute, and a mimic attribute.

[0060] During the verification and repair phase, the system performs multi-dimensional physical and semantic verification on the intermediate representation data structure. Kinematic topology verification uses a depth-first search algorithm to traverse the adjacency relationships with links as nodes and joints as edges, detecting loops and reporting errors, marking isolated nodes, and automatically selecting the first link without a parent joint when the number of root links is zero. When the number of root links is greater than one, the excess root links are connected to the world coordinate system through fixed joints. Simultaneously, it verifies that the parent and child links referenced by each joint exist in the link list. Inertial parameter verification checks whether the mass of each link is greater than zero. It constructs a 3x3 symmetric matrix from the six components of the inertial tensor given by URDF (ixx, iyy, izz, ixy, ixz, iyz) and calculates the eigenvalues. When all eigenvalues ​​are positive, it is considered positive definite. Triangle inequality constraints are also verified. Joint limit verification checks whether rotational and translative joints are defined. <limit>Whether the label and lower limit are less than the upper limit. Follower relationship validation targets... <mimic>The tag-defined driven joints check the existence of their referenced master joints and detect circular dependencies through chain tracing. Geometric path verification checks the validity of resource location prefixes and mesh file paths. When a repairable defect is identified, the automatic repair module performs corrections according to a preset strategy: for masses less than or equal to zero, the correction is 10 to the power of negative six kilograms; for inertia tensors that are not positive definite, the absolute values ​​of diagonal elements are taken and ensured to be not less than 10 to the power of negative nine; missing... <limit>The rotational joints of the label are supplemented with the default limit of negative to positive pi radians, and the traverse joints are supplemented with the default limit of negative one meter to positive one meter; the upper and lower limits are interchanged if inverted, and if equal, each is increased or decreased by 0.01; the absolute value is taken for damping coefficients or friction coefficients that are negative. For links with visual geometric definitions but no inertial parameters, a small mass and a unit inertia matrix are supplemented.

[0061] In the physical constraint mapping and generation phase, the system generates corresponding standard library components for each joint based on the normalized intermediate representation: revolute and continuous joints generate Revolute components, prismatic joints generate Prismatic components, and fixed joints generate Fixed components. The rotation axis direction and coordinate transformation of the joints are respectively derived from... <axis>and <origin>Tag extraction and setting. When a joint is defined. <limit>When tagging, an ElastoBacklash elastic stop component is generated. Its stroke parameter 'b' is set to the upper limit minus the lower limit, and its stiffness parameter 'c' is calculated according to the formula c = effort divided by the preset over-limit angle, where effort is the maximum torque value and the preset over-limit angle is 0.01 radians. The two ends of this component are connected to the joint's axis (one-dimensional mechanical port) and support port, respectively. When the joint is defined... <dynamics>When the tag includes a damping coefficient, a damper component is generated; when it includes a friction coefficient, a BearingFriction Coulomb friction component is generated. When the joint has limits, damping, friction parameters, or mimic follower relationships, the `useAxisFlange` parameter of the joint component is automatically set to true, exposing the two one-dimensional mechanical ports: axis and support. For defined... <mimic>For constrained joints, a closed-loop following control chain is generated, consisting of angular displacement sensors, adders, constant sources, and position actuators. This achieves a following relationship where the driven joint angle equals the multiplier transmission ratio multiplied by the master joint angle plus the offset. For links, an encapsulated sub-model is generated for each link, named "Parts" (link name). This sub-model contains a "Frame_a" reference interface, a "Body" component, and several "FixedTranslation" and "FixedRotation" components. Collision geometry is recorded using structured annotations. In complex base recognition and adaptation, when the joint type is floating or MJCF-mapped free type, a "FreeMotion" six-DOF spatial motion component is generated, providing translational and rotational degrees of freedom along the X, Y, and Z axes. When the joint type is planar, a planar motion constraint component is generated, restricting the robot to two translational degrees of freedom in the XY plane and one rotational degree of freedom around the Z axis. This is suitable for scenarios such as automated guided vehicles and autonomous mobile robots.

[0062] Reference Figure 3 , Figure 3 This diagram illustrates the intelligent layout engine strategy selection provided in this application embodiment. The diagram uses a decision tree structure: the left starting frame receives the robot topology analysis results, and the judgment diamond frame determines whether there are multiple symmetrical branches. If not, a serpentine layout strategy is adopted, placing components sequentially along the canvas direction according to the preorder traversal order of the kinematic tree, forming a serpentine chain. If so, a symmetrical layout strategy centered on the pelvis or base is adopted, symmetrically extending multiple branches from the central component to the left and right sides, with the upper body branches extending upwards and the lower body branches extending downwards. Specifically, for a typical industrial robotic arm structure, a serpentine layout is used to arrange the components sequentially along the X direction of the canvas; for a typical humanoid robot or quadruped robot structure, a vertically symmetrical layout centered on the pelvis is used. The calculated coordinates of each component on the canvas are embedded in the generated code as annotations.

[0063] During the model assembly and output phase, the system generates the Robot master model file, declares World component instances, all part sub-model instances, and all joint component instances at the top level, generates connect statements in the equation section based on the parent-child relationship of the joints, connects the World component and the root link through the root joint, connects all joints in series with each link sub-model according to the parent-child relationship, connects all physical constraint components to the physical constraint connection ports of the corresponding joints, and connects all driven follower control chains to the corresponding master joints and driven joints.

[0064] After the model assembly is completed, the system automatically generates a closed-loop controller. It iterates through all moving joints, excluding fixed joints and driven joints with mimic relationships, and generates a proportional-integral-derivative (PID) control loop for each independent degree of freedom. This loop includes a target position constant source, angle or displacement sensors, a feedback differential, a PID controller, and a torque or force actuator. The controller gain parameters are self-tuned as follows: The mass m, center of mass position, and inertia tensor of the link driven by the moving joint are obtained. The equivalent inertia J_eq is calculated according to the parallel axis theorem, i.e., J_eq equals the rotational inertia J_cm about the center of mass axis plus m multiplied by the square of the distance d from the center of mass to the axis of rotation. Based on the preset target natural frequency omega_n and damping ratio zeta, the PID parameters are calculated using the formulas: Kp equals the square of omega_n multiplied by J_eq, Kd equals 2 multiplied by zeta multiplied by omega_n multiplied by J_eq, and Ki equals Kp divided by the preset time constant. Simultaneously, a time-based trajectory planning channel is generated for each active joint. The start and end points are automatically set according to the limit range, and a smooth trajectory curve is generated using a fifth-order polynomial or higher-order continuously differentiable interpolation algorithm. Taking the H1-2 humanoid robot with fifty-eight links and thirty-nine degrees of freedom as an example, this method automatically generated a complete simulation model package containing fifty-one ElastoBacklash limit components, twelve Mimic control chains, and thirty-nine PID controllers in actual tests. The entire conversion process replaced the manual modeling work that originally required several days.

[0065] After the model is output, the system automatically generates a conversion report in HyperText Markup Language (HTML) format, which includes basic robot information statistics, link parameter mapping tables, joint parameter mapping tables, a tree-structured topology path diagram, and verification and repair records. The link parameter mapping table records line by line the name of each original link and its corresponding part sub-model name, mass value, center of mass coordinates, and inertia tensor; the joint parameter mapping table records line by line the original name of each joint and its corresponding joint type, limit range, damping coefficient, friction coefficient, and follower relationship; the tree-structured topology path diagram expands layer by layer from the root link according to the parent-child relationship to all end links; the verification and repair records summarize the problems found in each verification step and the automatic repair operations.

[0066] Regarding the technical solution of this application, there is also an alternative implementation method based on co-simulation using an external physics engine: instead of directly converting the robot description file into a native multibody component, the robot's dynamic equations are packaged and compiled into a standard Functional Mock-up Unit (FMU) using the original simulation environment. This FMU is then imported as a black-box component into the target simulation environment, exposing only the torque input port and sensor data output port, and co-simulating is performed through the Functional Mock-up Interface (FMI) standard. The disadvantage of this alternative is that it loses the white-box modeling characteristics; simulation engineers cannot directly view and modify link mass parameters, adjust structural topology, or observe 3D component details, and it significantly increases the dependency threshold of the toolchain.

[0067] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the automatic construction method of the robot multibody simulation model of this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0068] This application also provides an automatic construction device for robot multibody simulation models; please refer to [reference needed]. Figure 4 The automatic construction device for the robot multi-body simulation model includes: The acquisition module 10 is used to acquire a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The correction module 20 is used to perform legality verification on the physical constraint parameters and kinematic structure in the intermediate representation data structure, and when a repairable defect of a preset type is identified, to perform corresponding automatic correction on the intermediate representation data structure to obtain a normalized intermediate representation. The constraint module 30 is used to generate a corresponding joint component for each joint based on the normalized intermediate representation. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and the physical constraint component is connected to the physical constraint connection port. Output module 40 is used to connect all joint components, physical constraint components and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0069] And / or, the automatic construction device for the robot multibody simulation model includes: The first construction module is used to construct adjacency relationships based on the link and joint information in the kinematic structure, traverse the adjacency relationships to detect loops, isolated nodes and multiple nodes, and verify that the parent link and child link referenced by each joint are recorded in the link information. The first determination module is used to determine whether the mass of each link in the link information is positive, and to construct a matrix based on its inertia tensor to determine whether all eigenvalues ​​of the matrix are positive. The second determination module is used to determine whether the upper limit of the rotation joint and the lower limit of the locating joint are greater than the lower limit of the physical constraint parameters. The third determination module is used to determine, for the driven joints marked with a driven relationship in the joint information, whether the master joint they refer to exists in the joint information, and whether tracing along the driven relationship starting from the master joint will not return to the driven joint; The first correction module is used to perform corresponding automatic correction when a repairable defect of a preset type is identified. The automatic correction includes correcting non-positive mass values ​​to preset small positive values, correcting non-positive definite inertial tensors to positive definite matrices, swapping the inverted upper and lower limits of the limit, and taking the absolute values ​​of negative damping coefficients and friction coefficients to obtain a normalized intermediate representation.

[0070] And / or, the automatic construction device for the robot multibody simulation model includes: The first generation module is used to generate an elastic stop component when the physical constraint parameters of the joint include a range of motion limitation, determine the difference between the upper and lower limits of the range of motion limitation as the stroke of the elastic stop component, and determine the ratio of the preset maximum joint torque and the preset over-limit allowable amount as the stiffness of the elastic stop component. The second generation module is used to generate a damping component when the physical constraint parameters of the joint include a damping coefficient, and to assign the damping coefficient to the damping parameter of the damping component. The third generation module is used to generate a friction component when the physical constraint parameters of the joint include the friction coefficient, and to assign the friction coefficient to the friction parameters of the friction component. The first connection module is used to connect at least one of the elastic stop component, the damping component, or the friction component as the physical constraint component to the physical constraint connection port of the joint component.

[0071] And / or, the automatic construction device for the robot multibody simulation model includes: The reading module is used to read the robot description file and determine the format type based on the file extension or root tag. The first calling module is used to call a preset unified format parser to extract the mass, center of mass position and inertia tensor from the link label, and the joint type, rotation axis direction, limit parameter and damping friction coefficient from the joint label when the format type is a unified robot description format. The second calling module is used to call a preset simulation format adapter to convert the six-element pose label into position and attitude parameters when the format type is a simulation description format, and to map the limit parameters nested in the axis label and limit label into a structure consistent with the unified format. The third calling module is used to call a preset multi-joint format adapter to flatten the recursively nested tree-shaped body structure into a link list when the format type is a multi-joint contact dynamics format, create fixed joints for the non-joint sub-body, and convert the angle limit parameters into radians. An assembly module is used to assemble the parsed or adapted information into the unified intermediate representation data structure.

[0072] And / or, the automatic construction device for the robot multibody simulation model includes: The fourth generation module is used to generate a closed-loop control loop for each active joint. The closed-loop control loop includes a preset target position constant source, an angle or displacement sensor, a feedback differential, a preset proportional-integral-derivative controller, and a torque or force actuator. The first calculation module is used to obtain the mass, center of mass position and inertia tensor of the linkage driven by the movable joint, calculate the equivalent inertia of the rotation axis of the movable joint according to the parallel axis theorem, and determine the proportional gain, integral gain and derivative gain of the proportional-integral-derivative controller based on the equivalent inertia according to the preset target natural frequency and damping ratio. The fifth generation module is used to automatically generate a trajectory planning channel based on a time schedule according to the limit range of the active joint. The start and end points of the trajectory planning channel are automatically set according to the limit range, and a smooth trajectory curve is generated by continuous differentiable interpolation. The second output module is used to connect the torque or force actuator in the closed-loop control loop to the corresponding port of the joint component, and output a complete robot multibody dynamics simulation model containing the closed-loop control loop.

[0073] And / or, the automatic construction device for the robot multibody simulation model includes: The statistics module is used to count the number of links, joints, and degrees of freedom in the intermediate representation data structure, as well as the number of driven joints with a driven relationship. The first mapping module is used to generate a link parameter mapping table, which records the original name of each link in the robot description file and the corresponding part sub-model name, mass value, centroid coordinates and inertia tensor in the simulation model line by line. The second mapping module is used to generate a joint parameter mapping table, which records the original name of each joint and its corresponding joint type, limit range, damping coefficient, friction coefficient and follower relationship line by line. The third generation module is used to generate a tree-like topology path diagram starting from the root link based on the kinematic structure, and to summarize all verification issues and corresponding automatic repair operation records. The organization module is used to organize the link parameter mapping table, the joint parameter mapping table, the tree topology path diagram, and the automatic repair operation record into a preset format Hypertext Markup Language conversion report file.

[0074] And / or, the automatic construction device for the robot multibody simulation model includes: The sixth generation module is used to generate a six-degree-of-freedom spatial motion component when the type of the joint is identified as a floating joint or a free joint. The six-degree-of-freedom spatial motion component includes three translational degrees of freedom and three rotational degrees of freedom. The seventh generation module is used to generate a planar motion constraint component when the type of the joint is identified as a planar motion joint. The planar motion constraint component restricts the joint to two translational degrees of freedom and one rotational degree of freedom in a plane. The setting module is used to set the initial condition parameters of the six-degree-of-freedom spatial motion component or the planar motion constraint component based on the initial pose of the floating joint or free joint in the normalized intermediate representation. An alternative module is used to replace the default fixed base connection with the six-degree-of-freedom spatial motion component or the planar motion constraint component as the joint component.

[0075] The automatic construction device for robot multi-body simulation models provided in this application, employing the automatic construction method for robot multi-body simulation models in the above embodiments, can solve the technical problem of poor reliability in simulation model conversion in related technologies. Compared with the prior art, the beneficial effects of the automatic construction device for robot multi-body simulation models provided in this application are the same as those of the automatic construction method for robot multi-body simulation models provided in the above embodiments, and other technical features in the automatic construction device for robot multi-body simulation models are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0076] This application provides an automatic construction device for robot multi-body simulation models. The automatic construction device for robot multi-body simulation models includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the automatic construction method for robot multi-body simulation models in the above embodiment 1.

[0077] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an automated construction device suitable for implementing robot multi-body simulation models according to embodiments of this application. The automated construction device for robot multi-body simulation models in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, tablets, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital televisions and desktop computers. Figure 5 The automated construction device for robot multibody simulation models shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0078] like Figure 5 As shown, the automated construction device for robot multibody simulation models may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the automated construction device for robot multibody simulation models. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the automated construction device for robot multi-body simulation models to exchange data wirelessly or via wired communication with other devices. Although an automated construction device for robot multi-body simulation models with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0079] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0080] The automatic construction device for robot multi-body simulation models provided in this application, employing the automatic construction method for robot multi-body simulation models in the above embodiments, can solve the technical problem of poor reliability in simulation model conversion in related technologies. Compared with the prior art, the beneficial effects of the automatic construction device for robot multi-body simulation models provided in this application are the same as the beneficial effects of the automatic construction method for robot multi-body simulation models provided in the above embodiments, and other technical features in the automatic construction device for robot multi-body simulation models are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0081] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0082] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0083] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the automatic construction method of the robot multibody simulation model in the above embodiments.

[0084] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0085] The aforementioned computer-readable storage medium may be included in an automated construction device for robot multibody simulation models; or it may exist independently and not be assembled into an automated construction device for robot multibody simulation models.

[0086] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the automated construction device for the robot multibody simulation model, cause the automated construction device for the robot multibody simulation model to: Obtain a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The physical constraint parameters and kinematic structure in the intermediate representation data structure are validated for legality. When a repairable defect of a preset type is identified, the intermediate representation data structure is automatically corrected accordingly to obtain a normalized intermediate representation. Based on the standardized intermediate representation, a corresponding joint component is generated for each joint. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. Connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model.

[0087] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0088] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0089] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0090] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described automatic construction method for robot multi-body simulation models, thereby solving the technical problem of poor reliability in simulation model conversion in related technologies. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the automatic construction method for robot multi-body simulation models provided in the above embodiments, and will not be repeated here.

[0091] All acquisition of signals, information, or actions in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization of the owner of the relevant device.

[0092] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.< / mimic> < / dynamics> < / limit> < / origin> < / axis> < / limit> < / mimic> < / limit> < / limit> < / axis> < / pose> < / mimic> < / dynamics> < / limit> < / joint> < / collision> < / visual> < / inertial> < / mujoco> < / sdf> < / robot>

Claims

1. An automatic construction method for a multi-body simulation model of a robot, characterized in that, The method includes: Obtain a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The physical constraint parameters and kinematic structure in the intermediate representation data structure are validated for legality. When a repairable defect of a preset type is identified, the intermediate representation data structure is automatically corrected accordingly to obtain a normalized intermediate representation. Based on the standardized intermediate representation, a corresponding joint component is generated for each joint. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. Connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model; The steps of performing legality checks on the physical constraint parameters and kinematic structures in the intermediate representation data structure, and automatically correcting the intermediate representation data structure when a pre-defined type of repairable defect is identified to obtain a standardized intermediate representation include: Based on the link and joint information in the kinematic structure, an adjacency relationship is constructed. The adjacency relationship is traversed to detect loops, isolated nodes, and multiple nodes, and to verify that the parent and child links referenced by each joint are recorded in the link information. For each link in the link information, determine whether its mass is positive, and construct a matrix based on its inertia tensor to determine whether all eigenvalues ​​of the matrix are positive. For the physical constraint parameters, determine whether the upper limit of the rotation joint and the lower limit of the traverse joint are greater than the lower limit; For a follower joint marked with a follower relationship in the joint information, it is determined that the master joint it references exists in the joint information, and tracing along the follower relationship starting from the master joint will not return to the follower joint; When a repairable defect of a preset type is identified, the corresponding automatic correction is performed. The automatic correction includes correcting the non-positive mass to a preset small positive value, correcting the non-positive definite inertial tensor to a positive definite matrix, swapping the inverted upper and lower limits of the limit, and taking the absolute value of the negative damping coefficient and friction coefficient to obtain a normalized intermediate representation. The step of generating a corresponding joint component for each joint based on the normalized intermediate representation, activating the reserved physical constraint connection port on the joint component when the joint has the physical constraint parameters, automatically generating a physical constraint component corresponding to the physical constraint parameters, and connecting the physical constraint component to the physical constraint connection port includes: When the physical constraint parameters of the joint include a range of motion limitation, an elastic stop component is generated. The difference between the upper and lower limits of the range of motion limitation is determined as the stroke of the elastic stop component, and the quotient of the preset maximum joint torque and the preset over-limit allowable amount is determined as the stiffness of the elastic stop component. When the physical constraint parameters of the joint include a damping coefficient, a damping component is generated, and the damping coefficient is assigned to the damping parameter of the damping component. When the physical constraint parameters of the joint include the friction coefficient, a friction component is generated, and the friction coefficient is assigned to the friction parameters of the friction component. At least one of the elastic stop assembly, the damping assembly, or the friction assembly is used as the physical constraint assembly and connected to the physical constraint connection port of the joint assembly.

2. The method as described in claim 1, characterized in that, The steps of obtaining a robot description file in at least one format, parsing the description file to obtain a unified intermediate representation data structure, wherein the intermediate representation data structure contains link and joint information describing the robot's kinematic structure, and physical constraint parameters attached to the joints, include: Read the robot description file and determine the format type based on the file extension or root tag; When the format type is the Unified Robot Description Format, the preset Unified Format Parser is called to extract the mass, center of mass position and inertia tensor from the link label, and the joint type, rotation axis direction, limit parameter and damping friction coefficient from the joint label. When the format type is simulation description format, the preset simulation format adapter is called to convert the six-element pose label into position and attitude parameters, and the limit parameters nested in the axis label and limit label are mapped to a structure consistent with the unified format; When the format type is a multi-joint contact dynamics format, the preset multi-joint format adapter is called to flatten the recursively nested tree-shaped body structure into a link list, create fixed joints for the jointless sub-body, and convert the angle-based limit parameters into radians. The parsed or adapted information is assembled into the unified intermediate representation data structure.

3. The method as described in claim 1, characterized in that, After connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: A closed-loop control loop is generated for each active joint. The closed-loop control loop includes a preset target position constant source, an angle or displacement sensor, a feedback differential, a preset proportional-integral-derivative controller, and a torque or force actuator. The mass, center of mass position, and inertia tensor of the linkage driven by the movable joint are obtained. The equivalent inertia of the rotation axis of the movable joint is calculated according to the parallel axis theorem. Based on the equivalent inertia, the proportional gain, integral gain, and derivative gain of the proportional-integral-derivative controller are determined according to the preset target natural frequency and damping ratio. Based on the limiting range of the movable joint, a trajectory planning channel based on a time schedule is automatically generated. The start and end points of the trajectory planning channel are automatically set according to the limiting range, and a smooth trajectory curve is generated using continuous differentiable interpolation. Connect the torque or force actuator in the closed-loop control loop to the corresponding port of the joint assembly, and output a complete robot multibody dynamics simulation model containing the closed-loop control loop.

4. The method as described in claim 1, characterized in that, After connecting all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model, the method further includes: The intermediate representation data structure contains the number of links, joints, and degrees of freedom, as well as the number of driven joints with a follower relationship. Generate a link parameter mapping table, recording line by line the original name of each link in the robot description file and the corresponding part sub-model name, mass value, centroid coordinates, and inertia tensor in the simulation model; Generate a joint parameter mapping table, recording the original name of each joint and its corresponding joint type, limit range, damping coefficient, friction coefficient, and follower relationship line by line; Based on the kinematic structure, a tree-like topology path diagram starting from the root link is generated, and all verification issues and corresponding automatic repair operation records are summarized. The link parameter mapping table, the joint parameter mapping table, the tree topology path diagram, and the automatic repair operation record are organized into a preset format Hypertext Markup Language conversion report file.

5. The method as described in claim 1, characterized in that, The step of generating a corresponding joint component for each joint based on the normalized intermediate representation further includes: When the type of the joint is identified as a floating joint or a free joint, a six-degree-of-freedom spatial motion component is generated, which includes three translational degrees of freedom and three rotational degrees of freedom. When the type of the joint is identified as a planar motion joint, a planar motion constraint component is generated, which restricts the planar motion constraint component to two translational degrees of freedom and one rotational degree of freedom in the plane. Based on the initial pose of the floating joint or free joint in the normalized intermediate representation, set the initial condition parameters of the six-degree-of-freedom spatial motion component or the planar motion constraint component; The six-degree-of-freedom spatial motion component or the planar motion constraint component is used as the joint component, replacing the default fixed base connection.

6. An automatic construction device for a multi-body simulation model of a robot, characterized in that, The device includes: The acquisition module is used to acquire a robot description file in at least one format, parse the description file to obtain a unified intermediate representation data structure, the intermediate representation data structure containing link and joint information describing the robot's kinematic structure, as well as physical constraint parameters attached to the joints; The correction module is used to perform legality verification on the physical constraint parameters and kinematic structure in the intermediate representation data structure, and when a repairable defect of a preset type is identified, to perform corresponding automatic correction on the intermediate representation data structure to obtain a normalized intermediate representation. The constraint module is used to generate a corresponding joint component for each joint based on the normalized intermediate representation. When the joint has the physical constraint parameters, the reserved physical constraint connection port on the joint component is activated, and a physical constraint component corresponding to the physical constraint parameters is automatically generated and connected to the physical constraint connection port. The output module is used to connect all joint components, physical constraint components, and link components generated based on the normalized intermediate representation to output a complete robot multibody dynamics simulation model. The correction module is further configured to: construct adjacency relationships based on the link and joint information in the kinematic structure; traverse the adjacency relationships to detect loops, isolated nodes, and multiple nodes; verify that the parent and child links referenced by each joint are recorded in the link information; determine whether the mass of each link in the link information is positive; construct a matrix based on its inertia tensor to determine whether all eigenvalues ​​of the matrix are positive; determine whether the upper limit of the rotational and traversal joints is greater than the lower limit of the physical constraint parameters; determine whether the referenced master joint exists in the joint information for a slave joint marked with a follower relationship, and ensure that tracing along the follower relationship starting from the master joint will not return to the slave joint; and perform corresponding automatic correction for identified repairable defects of a preset type, including correcting non-positive mass to a preset small positive value, correcting non-positive definite inertia tensors to positive definite matrices, swapping inverted upper and lower limit values, and taking the absolute values ​​of negative damping coefficients and friction coefficients to obtain a normalized intermediate representation. The constraint module is further configured to: generate an elastic stop component for joints whose physical constraint parameters include a range of motion limitation; determine the stroke of the elastic stop component as the difference between the upper and lower limits of the range of motion limitation; and determine the stiffness of the elastic stop component as the quotient of a preset maximum joint torque and a preset over-limit allowable amount. For joints whose physical constraint parameters include a damping coefficient, a damping component is generated, and the damping coefficient is assigned to the damping parameter of the damping component. For joints whose physical constraint parameters include a friction coefficient, a friction component is generated, and the friction coefficient is assigned to the friction parameter of the friction component. At least one of the elastic stop component, the damping component, or the friction component is used as the physical constraint component and connected to the physical constraint connection port of the joint component.

7. An automatic construction device for multi-body simulation models of robots, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic construction method for a robot multibody simulation model as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the automatic construction method for a robot multibody simulation model as described in any one of claims 1 to 5.

Images