Logistics package simulation reconstruction method and device, electronic equipment and storage medium

Abstract

The invention provides a logistics parcel simulation reconstruction method and device, electronic equipment and a storage medium, and relates to the technical field of logistics parcel simulation reconstruction, and the method comprises the steps: obtaining geometric representation data of a logistics parcel, and obtaining a parcel simulation type corresponding to the logistics parcel; based on the package simulation type, converting the geometric representation data into a corresponding physical simulation model; generating a convex hull model based on the physical simulation model; and based on the physical simulation model, the convex hull model and the physical attributes of the logistics parcel, performing simulation in a digital twin environment of a logistics system. According to the method, the overall modeling efficiency can be ensured while the simulation trueness of the logistics parcel in digital twinning is improved.

Description

Technical Field

[0001] This invention relates to the field of logistics parcel simulation reconstruction technology, and in particular to a logistics parcel simulation reconstruction method, apparatus, electronic device and storage medium. Background Technology

[0002] With the rapid development of e-commerce, the throughput of parcels handled by logistics warehousing, sorting, and delivery systems continues to grow, and the complexity of the scenarios is increasing. Against this backdrop, accurate simulation of the dynamic behavior of parcels has become a key means to optimize system design, predict failure modes, and improve operational efficiency. Traditional system development processes typically rely on physical prototype testing or discrete event simulation, which not only has limitations in terms of cost, cycle time, and repeatability, but also struggles to accurately characterize the diversity of parcels in terms of geometry, material properties, and complex contact mechanics.

[0003] To achieve high-fidelity reproduction and system-level reliable verification of the aforementioned multiphysics characteristics, building digital twins of logistics systems has become an important industry trend. Digital twins can reproduce the equipment layout, sensor distribution, and dynamic behavior of packages in real-world warehousing and sorting scenarios within a virtual environment, thus enabling design verification and control loop testing without interfering with online operations. Currently, while mainstream digital twin implementations can build simulation environments based on physics engines and 3D modeling tools, they still face significant challenges when modeling logistics packages. Especially in handling irregular geometries, deformation behavior of soft and hard composite materials, and complex contact friction models, existing methods often struggle to balance simulation accuracy and computational efficiency. This results in a significant deviation between the simulated behavior of logistics packages and the real physical world, consequently affecting the reliability of subsequent simulation-based predictions, optimizations, and decisions.

[0004] Therefore, how to effectively improve the simulation realism of logistics parcels in digital twins while ensuring overall modeling efficiency is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, electronic device, and storage medium for simulating and reconstructing logistics packages, which can improve the simulation realism of logistics packages in digital twins while ensuring overall modeling efficiency.

[0006] This invention provides a method for simulating and reconstructing logistics parcels, comprising: Obtain the geometric representation data of the logistics package and obtain the package simulation type corresponding to the logistics package; Based on the package simulation type, the geometric representation data is converted into the corresponding physical simulation model; Based on the physical simulation model, a convex hull model is generated; Simulations are performed in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics package.

[0007] According to a logistics parcel simulation reconstruction method provided by the present invention, the step of converting the geometric representation data into a corresponding physical simulation model based on the parcel simulation type includes: If the package simulation type is elastic package, then the geometric representation data is converted into a tetrahedral finite element model; If the package simulation type is soft package, then the geometric representation data is converted into a particle system model; If the enclosure simulation type is rigid enclosure, then a surface mesh model is constructed based on the geometric representation data.

[0008] According to a logistics parcel simulation reconstruction method provided by the present invention, the step of converting the geometric representation data into a tetrahedral finite element model includes: The geometric representation data is preprocessed to obtain watertight mesh data; The watertight mesh data is tetrahedralized to generate a tetrahedral mesh model; The tetrahedral finite element model is constructed using the finite element method, based on the tetrahedral mesh model and preset elastic material parameters. The preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0009] According to the logistics package simulation reconstruction method provided by the present invention, before constructing the tetrahedral finite element model using the finite element method based on the tetrahedral mesh model and preset elastic material parameters, the method further includes: The nodes of the tetrahedral mesh model are smoothed. The process of constructing the tetrahedral finite element model using the finite element method, based on the tetrahedral mesh model and preset elastic material parameters, includes: The tetrahedral finite element model is constructed using the finite element method, based on a smoothed tetrahedral mesh model and preset elastic material parameters.

[0010] According to a logistics parcel simulation reconstruction method provided by the present invention, the step of converting the geometric representation data into a particle system model includes: The point cloud data in the geometric representation data is subjected to noise reduction processing; Based on the spatial location of each point in the point cloud data after noise reduction, the position information of each particle in three-dimensional space is determined. By using particle simulation methods, based on preset particle physical properties and the position information, the motion of particles is simulated to construct the particle system model; The preset particle physical properties include at least one of particle mass, particle influence radius, and particle initial velocity.

[0011] According to the logistics package simulation reconstruction method provided by the present invention, before determining the position information of each particle in three-dimensional space based on the spatial position of each point in the noise-reduced point cloud data, the method further includes: The point cloud data after noise reduction is resampled to obtain resampled point cloud data. The determination of the position information of each particle in three-dimensional space based on the spatial position of each point in the point cloud data after noise reduction includes: Based on the spatial position of each point in the resampled point cloud data, the position information of each particle in three-dimensional space is determined.

[0012] According to a logistics parcel simulation reconstruction method provided by the present invention, the simulation is performed in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics parcel, including: The physical simulation model, the convex hull model, and the physical properties of the logistics package are encapsulated into a package simulation asset file; wherein, the physical properties of the logistics package include at least one of package mass, package density, center of mass position, moment of inertia tensor, and physical material; The package simulation asset file is imported into the robot simulation platform for simulation in the digital twin environment of the logistics system.

[0013] The present invention also provides a logistics parcel simulation reconstruction device, comprising: The package data acquisition module is used to acquire the geometric representation data of the logistics package and to acquire the package simulation type corresponding to the logistics package; The first model generation module is used to convert the geometric representation data into a corresponding physical simulation model based on the package simulation type. The second model generation module is used to generate a convex hull model based on the physical simulation model; The logistics parcel simulation module is used to perform simulations in the digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics parcel.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the logistics package simulation reconstruction method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the logistics parcel simulation reconstruction method as described in any of the preceding claims.

[0016] The present invention provides a method, apparatus, electronic device, and storage medium for simulating and reconstructing logistics packages. It acquires the geometric representation data of the logistics packages and their corresponding simulation types. Then, based on the simulation type, it converts the geometric representation data into a corresponding physical simulation model. By constructing differentiated physical simulation models for different logistics packages, targeted high-fidelity modeling is achieved at the system level, effectively solving the problem that traditional methods, due to their single model, cannot handle complex physical behaviors such as deformation of irregularly shaped parts and flow of soft packages. Next, a convex hull model is generated based on the physical simulation model to serve as a collision model, ensuring the real-time performance of large-scale simulations. Finally, simulation is performed in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics packages. In the digital twin environment, physical properties drive and define the dynamic behavior and material response of the physical simulation model, the convex hull model handles rapid collision detection, and the physical model performs detailed deformation calculations based on the physical properties of the logistics packages. The two models work together, significantly improving the realism of logistics package simulations in the digital twin while ensuring overall modeling and simulation efficiency, providing a reliable basis for system design verification and operational optimization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts of the logistics parcel simulation reconstruction method provided by the present invention; Figure 2 This is the second flowchart of the logistics parcel simulation reconstruction method provided by the present invention; Figure 3 This is the third flowchart of the logistics parcel simulation reconstruction method provided by the present invention; Figure 4 This is a schematic diagram of the structure of the logistics parcel simulation reconstruction device provided by the present invention; Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] With the rapid development of e-commerce, the throughput of parcels handled by logistics warehousing, sorting, and delivery systems continues to grow, and the complexity of the scenarios is increasing. Against this backdrop, accurate simulation of the dynamic behavior of parcels has become a key means to optimize system design, predict failure modes, and improve operational efficiency. Traditional system development processes typically rely on physical prototype testing or discrete event simulation, which not only has limitations in terms of cost, cycle time, and repeatability, but also struggles to accurately characterize the diversity of parcels in terms of geometry, material properties, and complex contact mechanics.

[0021] To achieve high-fidelity reproduction and system-level reliability verification of the aforementioned multiphysics characteristics, building digital twins of logistics systems has become a significant industry trend. Digital twins can reproduce the equipment layout, sensor distribution, and dynamic behavior of packages in real-world warehousing and sorting scenarios within a virtual environment, enabling design verification and control loop testing without interfering with online operations. Currently, mainstream digital twin implementations, while capable of building simulation environments based on physics engines and 3D modeling tools, still face significant challenges when modeling logistics packages. Especially in handling irregular geometries, deformation behavior of soft and hard composite materials, and complex contact friction models, existing methods often struggle to balance simulation accuracy and modeling efficiency. This results in a significant deviation between the simulated behavior of logistics packages and the real physical world, consequently affecting the reliability of subsequent simulation-based predictions, optimizations, and decisions.

[0022] Therefore, how to effectively improve the simulation realism of logistics parcels in digital twins while ensuring overall modeling efficiency is a technical problem that urgently needs to be solved.

[0023] In the existing technology, the CAD (Computer-Aided Design) modeling methods for logistics packages generally include the following: (1) Parametric modeling method, which abstracts the package into basic geometric shapes such as cuboids and cylinders, and generates corresponding models by inputting parameters such as length, width, height, and radius of curvature. This method is suitable for modeling common regular cardboard boxes or standard packaging bodies; (2) Modeling method based on point cloud or 3D scanning, which obtains point cloud data by laser infrared scanning or depth camera for irregular parts or packages, and uses reverse engineering software to fit the surface, thereby obtaining a 3D CAD model that is similar to the real object; (3) Simplified modeling method, which simplifies complex geometric shapes into bounding boxes or convex hull models, retaining only the main physical properties such as volume and mass, without pursuing the refinement of the shape.

[0024] Analysis revealed that all of the above methods have certain shortcomings: (1) Parametric modeling can generate models quickly, but it cannot truly reflect the shape characteristics of irregular or soft wrapping; (2) Modeling methods based on point clouds or 3D scanning have high accuracy, but rely on expensive scanning equipment and complex data processing procedures, making it difficult to meet the needs of large-scale applications; (3) Simplified modeling improves simulation efficiency, but excessive abstraction can lead to deviations between the model and the actual wrapping in terms of geometry and physical performance, thus affecting the reliability of subsequent simulation results.

[0025] Furthermore, analysis revealed that in existing solutions, when the modeled packages are imported into the robot simulation platform, the packages exist in the form of physical entities, making it difficult to simulate deformations and other real-world phenomena.

[0026] Based on the above, this invention proposes a method, apparatus, electronic device, and storage medium for simulating and reconstructing logistics parcels. The following is a detailed description... Figures 1-5 Describe it.

[0027] Figure 1 This is one of the flowcharts illustrating the logistics parcel simulation reconstruction method provided by the present invention, such as... Figure 1 As shown, the logistics package simulation reconstruction method includes steps S110, S120, S130 and S140.

[0028] Step S110: Obtain the geometric representation data of the logistics package and obtain the package simulation type corresponding to the logistics package.

[0029] Geometric representation data refers to data describing the three-dimensional shape of logistics packages, including but not limited to point cloud data and surface mesh data. Surface mesh data is provided in STL (Stereolithography) or OBJ (Object) format.

[0030] The methods for obtaining geometric representation data include, but are not limited to: (1) 3D modeling of logistics packages using a CAD modeling platform to obtain surface mesh data, wherein the CAD modeling platform includes, but is not limited to: SolidWorks (a 3D modeling software), FreeCAD (an open source CAD), etc.; (2) scanning logistics packages using a 3D scanner to obtain point cloud data; (3) obtaining package digital assets from an existing package database and converting them into point cloud data or surface mesh data.

[0031] By employing the aforementioned methods for acquiring package models, the difficulty of obtaining package models is greatly reduced. At the same time, by utilizing existing package databases, readily available data can be quickly converted into the required format and loaded into the robot simulation platform, significantly increasing the amount of data and reducing data acquisition time.

[0032] It should be understood that surface mesh data and point cloud data can be converted to each other.

[0033] Parcel simulation type refers to a classification based on the physical deformation characteristics of logistics parcels, rather than a classification of logistics business.

[0034] Based on simulation requirements analysis, package simulation types are divided into two main categories: rigid packages and flexible packages. Flexible packages are further divided into elastic packages and soft packages. Rigid packages indicate that the package's shape remains essentially unchanged after being subjected to force; flexible packages indicate that the package's shape changes significantly after being subjected to force; elastic packages indicate that the package undergoes recoverable (complete or mostly recovered) deformation after being subjected to force; and soft packages indicate that the package undergoes large, irreversible deformation or flow after being subjected to force.

[0035] The methods for obtaining package simulation types include, but are not limited to: (1) obtaining them through manual marking; and (2) determining them based on preset rules. Specifically, parameters such as the weight, size, and surface hardness of the logistics package are obtained and then compared with preset rules to determine the package simulation type corresponding to each logistics package. For example, standard cardboard boxes and hard plastic boxes are rigid packages, soft packages filled with bubble wrap and foam boxes are elastic packages, and clothing bags and puffed food bags are soft packages.

[0036] Step S120: Based on the package simulation type, convert the geometric representation data into the corresponding physical simulation model.

[0037] A physical simulation model is a high-fidelity model built for a package that can be used to calculate its internal stress or particle motion.

[0038] If the package simulation type is elastic package, then the geometric representation data will be converted into a tetrahedral finite element model; If the package simulation type is soft package, then the geometric representation data will be converted into a particle system model; If the enclosure simulation type is rigid enclosure, then a surface mesh model is constructed based on the geometric representation data.

[0039] Step S130: Generate a convex hull model based on the physical simulation model.

[0040] First, determine the target convex hull creation tool or algorithm corresponding to the physical simulation model. Then, input the physical simulation model into the corresponding target convex hull creation tool or algorithm to create the convex hull and obtain the convex hull model.

[0041] In one embodiment, for tetrahedral finite element models and particle system models, the convex hull model can be calculated using a fast convex hull algorithm (e.g., Quickhull). In another embodiment, for surface mesh models, they can be made into convex hull models using modeling software plugins (such as SolidWorks, Blender, etc.).

[0042] The convex hull model is used to represent collision meshes, which allows the physical hull to interact physically with other objects in the simulation platform after being imported into the simulator.

[0043] Step S140: Based on the physical simulation model, the convex hull model, and the physical properties of the logistics package, perform simulation in the digital twin environment of the logistics system.

[0044] The physical properties of logistics parcels include, but are not limited to: parcel mass, parcel density, center of mass position, moment of inertia tensor, and physical materials.

[0045] In one implementation, the physical simulation model, convex hull model, and physical attributes of the logistics package are encapsulated into a single package simulation asset file. This package simulation asset file is then imported into a robot simulation platform for simulation within the digital twin environment of the logistics system.

[0046] In another implementation, a dedicated folder is created for each logistics package to store the physical simulation model, convex hull model, and physical properties of the package. The entire folder is then imported into a robot simulation platform for simulation within the digital twin environment of the logistics system.

[0047] In the first embodiment described above, the physical simulation model, convex hull model, and physical attributes of the logistics package are encapsulated into a single package simulation asset file. Compared with the folder storage method in the second embodiment described above, this method can ensure the integrity and consistency of the data, avoid data loss or mismatch that may be caused by multiple files being scattered, and simplify the deployment of the digital twin environment.

[0048] During simulation, the robot simulation platform's physics engine uses a convex hull model for rapid collision detection, determining whether a package is in contact with equipment or other packages. Once a collision is detected, a corresponding high-fidelity physics simulation model is triggered to perform precise deformation calculations based on the physical properties of the logistics package, such as the degree of denting of a foam box or the folded posture of a clothing bag.

[0049] Furthermore, while importing the package simulation asset files into the robot simulation platform, simulation environment parameters such as lighting can be configured within the platform's scene. This allows for a high-fidelity, multi-dimensional recreation of the real logistics sorting scenario at both the physical behavior and visual perception levels. This enables digital twins not only to verify the dynamic behavior of packages but also to simultaneously test vision-dependent recognition algorithms and evaluate operational reliability under different lighting conditions, thereby achieving closed-loop verification and optimization of the entire system from physical interaction to environmental perception.

[0050] The logistics parcel simulation reconstruction method provided in this invention acquires the geometric representation data of the logistics parcel and the corresponding parcel simulation type. Then, based on the parcel simulation type, the geometric representation data is converted into a corresponding physical simulation model. By constructing corresponding physical simulation models for different logistics parcels, targeted high-fidelity modeling is achieved at the system level, effectively solving the problem that traditional methods cannot take into account complex physical behaviors such as the deformation of irregularly shaped parts and the flow of soft parcels due to the single model. Next, a convex hull model is generated based on the physical simulation model as a collision model, ensuring the real-time performance of large-scale simulations. Finally, simulation is performed in the digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics parcel. When simulating in the digital twin environment of the logistics system, the physical properties are used to drive and define the dynamic behavior and material response of the physical simulation model, the convex hull model is responsible for rapid collision detection, and the physical model performs fine deformation calculations based on the physical properties of the logistics parcel. The two models work together, which can significantly improve the simulation realism of logistics parcels in the digital twin while ensuring the overall modeling and simulation efficiency, providing a reliable basis for system design verification and operation optimization.

[0051] Based on any of the above embodiments, step S120 includes: step S121, step S122 and step S123.

[0052] If the package simulation type is elastic package, then step S121 is executed: convert the geometric representation data into a tetrahedral finite element model.

[0053] If the package simulation type is elastic package, then the geometric representation data will be converted into a tetrahedral finite element model.

[0054] In one embodiment, the geometric representation data is preprocessed to obtain watertight mesh data; the watertight mesh data is tetrahedralized to generate a tetrahedral mesh model; and a tetrahedral finite element model is constructed based on the tetrahedral mesh model and preset elastic material parameters using the finite element method; wherein the preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0055] In another embodiment, the geometric representation data is preprocessed to obtain watertight mesh data; the watertight mesh data is tetrahedralized to generate a tetrahedral mesh model; the nodes of the tetrahedral mesh model are smoothed; and a tetrahedral finite element model is constructed based on the smoothed tetrahedral mesh model and preset elastic material parameters using the finite element method; wherein the preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0056] If the package simulation type is soft package, then step S122 is executed: convert the geometric representation data into a particle system model.

[0057] If the wrapping simulation type is soft wrapping, then the geometric representation data will be converted into a particle system model.

[0058] The point cloud data in the geometric representation data is subjected to noise reduction and / or resampling; based on the spatial position of each point in the noise-reduced point cloud data, the position information of each particle in three-dimensional space is determined; through particle simulation method, based on preset particle physical properties and position information, the interaction force between particles is simulated to construct a particle system model; wherein, the preset particle physical properties include at least one of particle mass, particle influence radius, particle velocity and resultant force acting on the particle.

[0059] If the enclosure simulation type is rigid enclosure, then step S123 is executed: construct a surface mesh model based on the geometric representation data.

[0060] If the enclosure simulation type is rigid enclosure, then a surface mesh model is constructed based on the geometric representation data.

[0061] A surface mesh model is constructed based on the surface mesh data in the geometric representation data. Since rigid wrapping does not require consideration of deformation, the surface mesh model is directly used as the physical simulation model.

[0062] The logistics parcel simulation reconstruction method provided in this invention matches the most suitable mathematical model for the physical nature of different types of parcels. This differentiated modeling approach optimizes simulation efficiency and, with limited computing power, can support larger-scale and more complex logistics parcel simulations. Furthermore, a tetrahedral finite element model is used for elastic parcels, while a particle system model is used for flexible parcels. This not only preserves the appearance of the logistics parcels but also simulates the deformation of flexible parcels under real-world conditions, thus greatly recreating the parcel's behavior in real life within the simulation platform.

[0063] Based on any of the above embodiments Figure 2 This is the second flowchart of the logistics parcel simulation reconstruction method provided by the present invention, as shown below. Figure 2 As shown, step S121 includes: step S1211, step S1212 and step S1213.

[0064] Step S1211: Preprocess the geometric representation data to obtain watertight mesh data.

[0065] Watertight mesh data: refers to the mesh data corresponding to a triangular mesh model without any holes and with completely closed boundaries. This is a prerequisite for tetrahedral meshing.

[0066] In one embodiment, when the geometric representation data is point cloud data obtained through 3D scanning or other methods, considering that the original point cloud may contain noise, outliers, and missing data regions (holes), it needs to be preprocessed first. The preprocessing steps may include point cloud denoising, outlier removal, and surface reconstruction and hole repair based on point cloud reconstruction algorithms (such as Poisson reconstruction and rolling sphere method), ultimately generating a continuous, closed triangular surface mesh, i.e., watertight mesh data.

[0067] In another implementation, when the geometric representation data is surface mesh data exported from CAD software or an existing surface mesh model, it may already be a watertight mesh. In this case, the surface mesh data is first verified. If the verification passes, the surface mesh data can be directly used as watertight mesh data; if defects such as holes or non-manifold edges are found in the mesh, a mesh repair tool needs to be used to repair the surface mesh data to obtain watertight mesh data, ensuring that it meets the watertightness requirements.

[0068] Step S1212: Tetrahedral subdivision of the watertight mesh data to generate a tetrahedral mesh model.

[0069] Then, the watertight mesh data is tetrahedralized to generate a tetrahedral mesh model.

[0070] Using watertight mesh data as the constraint boundary input, the CDT (Constrained Delaunay Tetrahedralization) algorithm generates a series of tetrahedra inside the boundary, thus obtaining an initial tetrahedral mesh model. It should be understood that the surface triangles of these tetrahedra are completely identical to the input mesh, thereby accurately preserving the original geometry.

[0071] Step S1213: Using the finite element method, the tetrahedral finite element model is constructed based on the tetrahedral mesh model and preset elastic material parameters.

[0072] The preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0073] A tetrahedral finite element model is constructed using the finite element method (FEM) based on a tetrahedral mesh model and preset elastic material parameters. The specific construction process is as follows: Based on the actual constituent materials of the elastic package, preset elastic material parameters are configured for it. These preset elastic material parameters include, but are not limited to, Young's modulus, Poisson's ratio, shear modulus, and bulk modulus. It should be noted that the preset elastic material parameters must at least include Young's modulus and Poisson's ratio to define the linear elastic constitutive relationship of the material. Shear modulus and bulk modulus can both be calculated from Young's modulus and Poisson's ratio. Of course, depending on engineering requirements, the preset elastic material parameters may also directly include Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0074] Import the tetrahedral mesh model into a finite element solver, assign parameter values ​​to each tetrahedral element in the tetrahedral mesh model based on preset elastic material parameters, and then perform a finite element solution to construct the tetrahedral finite element model. Finite element solvers include, but are not limited to, Abaqus, ANSYS, or the FEM kernel.

[0075] The logistics package simulation reconstruction method provided in this invention assigns elastic material parameters to a tetrahedral mesh model, and then constructs a tetrahedral finite element model using the finite element method, thereby achieving quantitative and high-precision prediction of the deformation behavior of elastic packages.

[0076] Based on any of the above embodiments, before step S1213, the method further includes: Step S1214: Smooth the nodes of the tetrahedral mesh model.

[0077] At this point, step S1213 includes: constructing the tetrahedral finite element model using the finite element method, based on the smoothed tetrahedral mesh model and preset elastic material parameters.

[0078] Considering that the initial tetrahedral mesh model obtained by directly converting watertight mesh data may contain a large number of low-quality elements, these low-quality elements are very likely to cause numerical instability, convergence difficulties or even calculation failures in subsequent finite element analysis.

[0079] The Laplacian Smoothing algorithm is applied to the initial tetrahedral mesh. This algorithm optimizes the cell shape by iteratively adjusting the positions of the internal nodes, avoiding tetrahedrons that are too thin, flat, or have small interior angles.

[0080] Then, using the finite element method, a tetrahedral finite element model is constructed based on the smoothed tetrahedral mesh model and preset elastic material parameters.

[0081] The logistics package simulation reconstruction method provided in this embodiment of the invention performs mesh smoothing optimization after tetrahedral subdivision of the watertight mesh data. This results in better element quality and more regular shape of the processed tetrahedral mesh model, which can significantly improve the numerical stability, calculation accuracy and convergence speed of subsequent finite element simulation, and ensure the reliability of the simulation results of elastic package deformation and stress.

[0082] Based on any of the above embodiments Figure 3 This is the third flowchart of the logistics parcel simulation reconstruction method provided by the present invention, as shown below. Figure 3 As shown, step S122 includes: step S1221, step S1222 and step S1223.

[0083] Step S1221: Denoise reduction processing is performed on the point cloud data in the geometric representation data.

[0084] The noise reduction algorithms used in the noise reduction process include, but are not limited to: statistical outlier removal algorithm, radius filtering, etc.

[0085] The point cloud data is denoised to obtain denoised point cloud data. This denoising process filters out floating noise points that are significantly far from the main object, caused by scanning errors.

[0086] Step S1222: Based on the spatial position of each point in the point cloud data after noise reduction, determine the position information of each particle in three-dimensional space.

[0087] The spatial position of each point in the noise-reduced point cloud data is determined as the initial position of each particle in three-dimensional space, and its position information in three-dimensional space is the three-dimensional coordinate of the corresponding point cloud data point.

[0088] Step S1223: Using a particle simulation method, based on preset particle physical properties and the position information, the motion of the particles is simulated to construct the particle system model.

[0089] The preset particle physical properties include at least one of particle mass, particle influence radius, and particle initial velocity.

[0090] Particle mass, the mass of each particle, can be determined based on the package mass and the number of point cloud data points. For example, if there are 50,000 point cloud data points and the package mass is 0.5 kg, then the particle mass can be set to 0.00001 kg.

[0091] The particle influence radius refers to the maximum distance between particle interactions, which can be set according to the average spacing of the point cloud and the simulation accuracy requirements.

[0092] The initial velocity of a particle refers to the initial velocity vector of the particle in three-dimensional space, which is usually set to (0,0,0).

[0093] After particles are created and assigned the aforementioned preset particle physical properties, their motion is driven using particle simulation methods to calculate the particle's motion state, including its position, velocity, and the net force acting on it. The net force is the vector sum of all forces acting on a single particle. It is the cause of changes in the particle's motion state and is composed of internal inter-particle interaction forces and external forces (such as gravity and contact forces).

[0094] Among them, particle simulation methods may include, but are not limited to: SPH (Smoothed Particle Hydrodynamics) and PBD (Position-Based Dynamics).

[0095] In one implementation, the interacting neighbors of each particle are identified based on its location information and radius of influence. According to the neighbor distribution, the internal forces acting on the current particle (including pressure gradient forces and viscous forces) are calculated using the SPH particle simulation method, and combined with external forces, to determine the instantaneous resultant force acting on each particle. Subsequently, the particle's velocity and position are updated based on this resultant force. By continuing this process, a complete particle system model is constructed to simulate the deformation and motion of flexible logistics packages.

[0096] In another implementation, the interacting neighbors of each particle are identified based on its location information and influence radius. Density and distance constraints are then constructed according to the neighbor distribution. These constraints are iteratively solved using the SPH particle simulation method to update the particle positions, and the particle velocities are updated based on the modified positions. By continuing this process, a complete particle system model is constructed to simulate the deformation and movement of flexible logistics packages.

[0097] The logistics package simulation reconstruction method provided in this invention discretizes continuous, flexible logistics packages—which are difficult to describe using traditional meshes—into an interacting set of particles, thus constructing a particle system model. This allows for the realistic simulation of the extremely nonlinear deformation and pseudo-fluid behavior of flexible packages. This is something that traditional rigid or elastic body models cannot achieve at all, and is crucial for understanding the unique fault modes of flexible logistics packages, such as jamming and entanglement, in automated sorting.

[0098] Based on any of the above embodiments, before step S1222, step S1224 is also included.

[0099] Step S1224: Resample the noise-reduced point cloud data to obtain resampled point cloud data.

[0100] At this point, step S1222 includes: determining the position information of each particle in three-dimensional space based on the spatial position of each point in the resampled point cloud data.

[0101] Considering the need to balance simulation accuracy and computational efficiency, the density of point cloud data directly determines the number and distribution quality of particles in the particle system model, thus affecting the computational cost and physical realism of the simulation.

[0102] Therefore, after denoising the point cloud data, the denoised point cloud data is further resampled to obtain resampled point cloud data.

[0103] In one implementation, if the point cloud data is determined to be too dense based on the point cloud data, the point cloud data is downsampled by voxel grid filtering to obtain a smaller number of particles.

[0104] In another implementation, if the point cloud data is determined to be too sparse, the point cloud data is upsampled by interpolation to obtain a smoother surface representation.

[0105] The logistics package simulation reconstruction method provided in this embodiment of the invention optimizes the quality of source data used to generate particle system models by resampling the point cloud data after noise reduction. This effectively controls the computational scale and improves the overall simulation efficiency and stability while ensuring the realism of deformation simulation.

[0106] Based on any of the above embodiments, step S140 includes: step S141 and step S142.

[0107] Step S141: Encapsulate the physical simulation model, the convex hull model, and the physical attributes of the logistics package into a package simulation asset file.

[0108] The physical properties of the logistics package include at least one of the following: package mass, package density, center of mass position, moment of inertia tensor, and physical material.

[0109] For each logistics package, its physical simulation model, convex hull model, and physical attributes are encapsulated into a package simulation asset file and exported. The file can be in USD (Universal Scene Description), STL, OBJ, or other formats.

[0110] This single package simulation asset file contains all hierarchical relationships, references, and custom attributes. It is natively supported by robot simulation platforms such as Isaac Sim (NVIDIA's Isaac simulation platform) and enables more complex scene assembly and real-time collaboration.

[0111] Step S142: Import the package simulation asset file into the robot simulation platform for simulation in the digital twin environment of the logistics system.

[0112] Robot simulation platforms, including but not limited to: Isaac, SimMujoco (Multi-Joint dynamics with Contact), Gazebo (robot simulator), and other simulation platforms.

[0113] Importing the package simulation asset file into the robot simulation platform allows the platform to fully identify its hierarchical structure, multi-model representation, and physical attributes, enabling high-fidelity simulation of complex scenarios. Specifically, after parsing the package simulation asset file, the platform automatically performs the following configurations: setting the convex hull model as the collision model for the logistics package, binding the physical simulation model to the built-in FEM solver plugin, and loading the physical attributes of the logistics package. This completes the one-click deployment of the package in the simulation environment for simulation within the digital twin environment of the logistics system.

[0114] It should be understood that while importing the package simulation asset file into the robot simulation platform, simulation environment parameters such as lighting can also be configured within the simulation platform's scene. This allows for a high-fidelity, multi-dimensional recreation of the real logistics sorting scenario at both the physical behavior and visual perception levels. This enables digital twins not only to verify the dynamic behavior of packages but also to simultaneously test vision-dependent recognition algorithms and evaluate operational reliability under different lighting conditions, thereby achieving closed-loop verification and optimization of the entire system from physical interaction to environmental perception.

[0115] The logistics parcel simulation reconstruction method provided in this invention simplifies the deployment content compared to traditional logistics simulation methods by using the above-mentioned parcel simulation asset encapsulation and automated deployment process. It achieves standardized management and intelligent configuration of simulation assets, and significantly reduces the difficulty and time cost of constructing complex physical simulation scenarios.

[0116] The logistics parcel simulation reconstruction device provided in the embodiments of this application is described below. The logistics parcel simulation reconstruction device described below can be referred to in correspondence with the logistics parcel simulation reconstruction method described above.

[0117] Figure 4 This is a schematic diagram of the logistics parcel simulation reconstruction device provided by the present invention, as shown below. Figure 4 As shown, the device includes a package data acquisition module 410, a first model generation module 420, a second model generation module 430, and a logistics package simulation module 440; wherein: The package data acquisition module 410 is used to acquire the geometric representation data of the logistics package and acquire the package simulation type corresponding to the logistics package; The first model generation module 420 is used to convert the geometric representation data into a corresponding physical simulation model based on the package simulation type. The second model generation module 430 is used to generate a convex hull model based on the physical simulation model; The logistics package simulation module 440 is used to perform simulation in the digital twin environment of the logistics system based on the physical simulation model, the convex hull model and the physical properties of the logistics package.

[0118] The logistics parcel simulation reconstruction device provided in this invention acquires the geometric representation data of the logistics parcels and their corresponding parcel simulation types. Then, based on the parcel simulation type, it converts the geometric representation data into corresponding physical simulation models. By constructing corresponding physical simulation models for different logistics parcels, targeted high-fidelity modeling is achieved at the system level, effectively solving the problem that traditional methods, due to their single model, cannot handle complex physical behaviors such as deformation of irregularly shaped parts and flow of soft parcels. Next, a convex hull model is generated based on the physical simulation model to serve as a collision model, ensuring the real-time performance of large-scale simulations. Finally, simulation is performed in the digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics parcels. During simulation in the digital twin environment of the logistics system, physical properties drive and define the dynamic behavior and material response of the physical simulation model, the convex hull model handles rapid collision detection, and the physical model performs detailed deformation calculations based on the physical properties of the logistics parcels. The two models work together, significantly improving the simulation realism of logistics parcels in the digital twin while ensuring overall modeling and simulation efficiency, providing a reliable basis for system design verification and operational optimization.

[0119] According to the present invention, a logistics parcel simulation reconstruction device is provided, wherein the first model generation module 420 includes: The first model generation unit is used to: if the package simulation type is elastic package, convert the geometric representation data into a tetrahedral finite element model; The second model generation unit is used to: if the package simulation type is soft package, convert the geometric representation data into a particle system model; The third model generation unit is used to: if the package simulation type is rigid package, then construct a surface mesh model based on the geometric representation data.

[0120] According to the logistics parcel simulation reconstruction device provided by the present invention, the first model generation unit is specifically used for: The geometric representation data is preprocessed to obtain watertight mesh data; The watertight mesh data is tetrahedralized to generate a tetrahedral mesh model; The tetrahedral finite element model is constructed using the finite element method, based on the tetrahedral mesh model and preset elastic material parameters. The preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

[0121] According to the logistics parcel simulation reconstruction device provided by the present invention, the first model generation unit is further specifically used for: The nodes of the tetrahedral mesh model are smoothed. The tetrahedral finite element model is constructed using the finite element method, based on a smoothed tetrahedral mesh model and preset elastic material parameters.

[0122] According to the logistics parcel simulation reconstruction device provided by the present invention, the second model generation unit is specifically used for: The point cloud data in the geometric representation data is subjected to noise reduction processing; Based on the spatial location of each point in the point cloud data after noise reduction, the position information of each particle in three-dimensional space is determined. By using particle simulation methods, based on preset particle physical properties and the position information, the motion of particles is simulated to construct the particle system model; The preset particle physical properties include at least one of particle mass, particle influence radius, and particle initial velocity.

[0123] According to the logistics package simulation reconstruction device provided by the present invention, the second model generation unit is further specifically used for: The point cloud data after noise reduction is resampled to obtain resampled point cloud data. The determination of the position information of each particle in three-dimensional space based on the spatial position of each point in the point cloud data after noise reduction includes: Based on the spatial position of each point in the resampled point cloud data, the position information of each particle in three-dimensional space is determined.

[0124] According to the present invention, a logistics package simulation reconstruction device is provided, wherein the logistics package simulation module 440 is specifically used for: The physical simulation model, the convex hull model, and the physical properties of the logistics package are encapsulated into a package simulation asset file; wherein, the physical properties of the logistics package include at least one of package mass, package density, center of mass position, moment of inertia tensor, and physical material; The package simulation asset file is imported into the robot simulation platform for simulation in the digital twin environment of the logistics system.

[0125] It should be noted that the logistics package simulation reconstruction device provided in this embodiment of the invention can realize all the method steps implemented in the above logistics package simulation reconstruction method embodiment and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0126] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5As shown, the electronic device may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute the logistics package simulation reconstruction method provided in the above embodiments. This logistics package simulation reconstruction method includes: acquiring geometric representation data of the logistics package and acquiring the package simulation type corresponding to the logistics package; converting the geometric representation data into a corresponding physical simulation model based on the package simulation type; generating a convex hull model based on the physical simulation model; and performing simulation in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics package.

[0127] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the logistics package simulation reconstruction method provided in the above embodiments. The logistics package simulation reconstruction method includes: acquiring geometric representation data of a logistics package and acquiring the package simulation type corresponding to the logistics package; converting the geometric representation data into a corresponding physical simulation model based on the package simulation type; generating a convex hull model based on the physical simulation model; and performing simulation in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics package.

[0129] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0130] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus the necessary high-resource hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating and reconstructing logistics parcels, characterized in that, include: Obtain the geometric representation data of the logistics package and obtain the package simulation type corresponding to the logistics package; Based on the package simulation type, the geometric representation data is converted into the corresponding physical simulation model; Based on the physical simulation model, a convex hull model is generated; Simulations are performed in a digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics package.

2. The logistics parcel simulation reconstruction method according to claim 1, characterized in that, The step of converting the geometric representation data into a corresponding physical simulation model based on the package simulation type includes: If the package simulation type is elastic package, then the geometric representation data is converted into a tetrahedral finite element model; If the package simulation type is soft package, then the geometric representation data is converted into a particle system model; If the enclosure simulation type is rigid enclosure, then a surface mesh model is constructed based on the geometric representation data.

3. The logistics parcel simulation reconstruction method according to claim 2, characterized in that, The process of converting the geometric representation data into a tetrahedral finite element model includes: The geometric representation data is preprocessed to obtain watertight mesh data; The watertight mesh data is tetrahedralized to generate a tetrahedral mesh model; The tetrahedral finite element model is constructed using the finite element method, based on the tetrahedral mesh model and preset elastic material parameters. The preset elastic material parameters include at least one of Young's modulus, Poisson's ratio, shear modulus, and bulk modulus.

4. The logistics parcel simulation reconstruction method according to claim 3, characterized in that, Before constructing the tetrahedral finite element model using the finite element method based on the tetrahedral mesh model and preset elastic material parameters, the process further includes: The nodes of the tetrahedral mesh model are smoothed. The process of constructing the tetrahedral finite element model using the finite element method, based on the tetrahedral mesh model and preset elastic material parameters, includes: The tetrahedral finite element model is constructed using the finite element method, based on a smoothed tetrahedral mesh model and preset elastic material parameters.

5. The logistics parcel simulation reconstruction method according to claim 2, characterized in that, The step of converting the geometric representation data into a particle system model includes: The point cloud data in the geometric representation data is subjected to noise reduction processing; Based on the spatial location of each point in the point cloud data after noise reduction, the position information of each particle in three-dimensional space is determined. By using particle simulation methods, based on preset particle physical properties and the position information, the motion of particles is simulated to construct the particle system model; The preset particle physical properties include at least one of particle mass, particle influence radius, and particle initial velocity.

6. The logistics parcel simulation reconstruction method according to claim 5, characterized in that, Before determining the position information of each particle in three-dimensional space based on the spatial position of each point in the point cloud data after noise reduction, the method further includes: The point cloud data after noise reduction is resampled to obtain resampled point cloud data. The determination of the position information of each particle in three-dimensional space based on the spatial position of each point in the point cloud data after noise reduction includes: Based on the spatial position of each point in the resampled point cloud data, the position information of each particle in three-dimensional space is determined.

7. The logistics parcel simulation reconstruction method according to any one of claims 1 to 6, characterized in that, The simulation, based on the physical simulation model, the convex hull model, and the physical properties of the logistics package, is performed in a digital twin environment of the logistics system, including: The physical simulation model, the convex hull model, and the physical properties of the logistics package are encapsulated into a package simulation asset file; wherein, the physical properties of the logistics package include at least one of package mass, package density, center of mass position, moment of inertia tensor, and physical material; The package simulation asset file is imported into the robot simulation platform for simulation in the digital twin environment of the logistics system.

8. A logistics parcel simulation reconstruction device, characterized in that, include: The package data acquisition module is used to acquire the geometric representation data of the logistics package and to acquire the package simulation type corresponding to the logistics package; The first model generation module is used to convert the geometric representation data into a corresponding physical simulation model based on the package simulation type. The second model generation module is used to generate a convex hull model based on the physical simulation model; The logistics parcel simulation module is used to perform simulations in the digital twin environment of the logistics system based on the physical simulation model, the convex hull model, and the physical properties of the logistics parcel.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the logistics parcel simulation reconstruction method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the logistics parcel simulation reconstruction method as described in any one of claims 1 to 7.

Images