Search

KR-20260063153-A - METHOD AND APPARATUS FOR MODELING OF DIGITAL TWIN

KR20260063153AKR 20260063153 AKR20260063153 AKR 20260063153AKR-20260063153-A

Abstract

The present invention relates to a digital twin modeling method and apparatus using actual measurement data. Specifically, it relates to a method and apparatus that generate a digital twin model by collecting 3D scan data, LiDAR data, and actual measurement data for a target space to be modeled as a digital twin using a 3D scan camera, a LiDAR camera, and a laser ruler, thereby ensuring that there is no difference from the actual space and reducing production costs.

Inventors

  • 이석희

Assignees

  • 엑스투알 주식회사

Dates

Publication Date
20260507
Application Date
20241030

Claims (10)

  1. A 3D scan data collection step for collecting 3D scan data obtained by capturing the target space to be modeled as a digital twin using a 3D scan camera; A LiDAR image data collection step for collecting LiDAR image data that captures the plane of the target space through a LiDAR camera; A measurement data collection step for collecting measurement data including actual measurements for a target space, a sub-space separating the target space, and a plurality of objects located in the target space through a measurement device; and A digital twin model generation step comprising generating a digital twin model by matching a digital 3D spatial model generated using the collected 3D scan shooting data with a digital planar model generated using the collected LiDAR shooting data; A digital twin modeling method characterized by applying the actual measurement data to the digital twin model and adjusting the digital twin model through the actual measurement data to generate a digital twin model identical to the target space.
  2. In claim 1, The above digital twin modeling method is, A digital twin modeling method characterized by utilizing the collected 3D scan data as basic data for a digital twin model for a sub-space separating the target space and a plurality of objects located in the target space through matching the digital 3D space model and the digital planar model, and utilizing the collected LiDAR scan data as basic data for a digital twin model for the entire target space.
  3. In claim 1, The above digital twin model generation step is, A digital twin modeling method further comprising: a digital 3D space model generation step of generating the digital 3D space model by repeating the process of setting one of a plurality of first shooting data constituting the above 3D scan shooting data as a first reference shooting data, detecting and aligning corresponding points for an overlapping area between the set first reference shooting data and another first shooting data, and setting the aligned result as a new first reference shooting data to detect and align corresponding points for an overlapping area between another first reference shooting data.
  4. In claim 3, The above digital 3D space model generation step is, A digital twin modeling method characterized by detecting a corresponding point for the overlapping area by detecting, for each cloud point of the first reference shooting data set above, the cloud point among the cloud points of the other first shooting data where the Euclidean distance is the minimum.
  5. In claim 1, The above digital twin model generation step is, A digital twin modeling method further comprising: a digital plane model generation step of generating the digital plane model by repeating the process of setting one of a plurality of second shooting data constituting the above LiDAR shooting data as a second reference shooting data, detecting and aligning corresponding points for an overlapping area between the set second reference shooting data and another second shooting data, and setting the aligned result as a new second reference shooting data and detecting and aligning corresponding points for an overlapping area between another second reference shooting data.
  6. In claim 5, The above digital planar model generation step is, A digital twin modeling method characterized by detecting a corresponding point for the overlapping area by detecting, for each cloud point of the second reference shooting data set above, the cloud point among the cloud points of the other second shooting data where the Euclidean distance is the minimum.
  7. In claim 1, The above digital twin model generation step is, A digital twin modeling method characterized by creating the digital twin model by fitting the plane of the digital 3D space model created above to the digital planar model created above and adjusting the digital 3D space model based on the digital planar model created above.
  8. In claim 1, The above measured data is, A digital twin modeling method characterized by comprising the above-mentioned target space, a sub-space separating the above-mentioned target space, and actual measurements and labels for a plurality of objects or a combination thereof.
  9. In claim 8, The above digital twin model generation step is, A digital twin modeling method further comprising: a step of applying actual measurement data to the digital twin model generated through the matching, wherein the total space, subspace, and multiple objects of the digital twin model generated through the matching correspond to the target space, subspace, multiple objects, or combination thereof of the collected actual measurement data, respectively, are each assigned labels according to the target space, subspace, multiple objects, or combination thereof of the collected actual measurement data, and the size of the digital twin model is adjusted according to the actual measurement corresponding to the assigned labels, thereby applying the actual measurement data to the digital twin model generated through the matching.
  10. A memory storing a program for performing a method according to any one of claims 1 to 9; and A digital twin modeling device using actual measurement data, characterized by including a processor that loads and executes a program stored in the memory.

Description

Method and apparatus for modeling of digital twin The present invention relates to a digital twin modeling method and apparatus using actual measurement data, and more specifically, to a method and apparatus that generates a digital twin model by collecting 3D scan shooting data, LiDAR shooting data, and actual measurement data of a target space for modeling a digital twin using a 3D scan camera, LiDAR camera, and laser ruler, thereby ensuring that there is no difference from the actual space and reducing production costs. A digital twin refers to a technology that models a real physical space (such as a city or smart factory) containing objects (such as devices or things) into a virtual space, which is a digital space. These digital twins offer the advantage of providing various information through a virtual space to monitor the state of physical objects in real time or optimize performance by predicting object problems through simulation, and are being applied in various fields such as smart factories, smart cities, and healthcare. To efficiently perform monitoring, problem prediction, and optimization through digital twins, it is crucial to model the actual physical target space into a digital space to create a digital twin model that is indistinguishable from the actual physical target space. However, conventional technology for generating digital twin models has the problem that it takes a long time to produce because experts manually model various objects one by one, and the shape, form, and measurements inevitably differ from the actual physical object space. Accordingly, the present invention proposes a method for generating and providing a digital twin model identical to the actual physical target space. In addition, the present invention proposes a method for generating a digital twin model of the actual target space by superimposing the actual target space using a 3D scan camera to collect 3D scan shooting data and LiDAR shooting data superimposed on the plane of the actual space using a LiDAR camera, generating a digital 3D space model using the 3D scan shooting data, generating a digital plane model using the LiDAR shooting data, and then matching the digital 3D space model to the digital plane model. Furthermore, the present invention proposes a method for finally generating a digital twin model identical to the target space by collecting actual measurement data for the target space and a plurality of objects included in each target space and applying it to the generated digital twin model. Next, we will briefly explain the prior art existing in the technical field of the present invention, and then describe the technical details that the present invention aims to achieve differently from the said prior art. First, Korean Published Patent No. 2023-0138139 (October 5, 2023) relates to a digital twin modeling device for predicting airflow in a plant factory, which replicates the shape constituting the actual system of a plant factory including sensors and equipment used for plant cultivation, builds the spatial information of the plant factory into a DB, and performs 3D modeling utilizing the spatial information of the plant factory. In other words, Korean Patent Publication No. 2023-0138139 describes only the extent of replicating the shape constituting the actual system of a physical factory, and does not utilize the 3D scan data, LiDAR data, and actual measurement data of the actual target space and a plurality of objects provided in the target space proposed in the present invention, nor does it describe at all the method of modeling and generating a digital twin model identical to the target space using the 3D scan data, LiDAR data, and actual measurement data. Therefore, there are significant differences between the two inventions in their technical configuration and effects. In addition, Korean Published Patent No. 2020-0013297 (January 26, 2023) relates to a method and apparatus for automatically identifying changes to facilities based on digital twins and video. Specifically, it concerns a method and apparatus for automatically identifying changes to facilities based on digital twins and video that constructs a digital twin model that models a space, analyzes CCTV footage acquired from a CCTV camera capturing the space to recognize objects within the space, and then tracks changes in the location of the recognized objects through the digital twin model. In other words, Korean Published Patent No. 2020-0013297 merely describes the creation of a digital twin model of space, but does not describe at all how the digital twin model is created. On the other hand, the present invention generates a digital twin model that matches (is identical to) the target space using 3D scan data and LiDAR shooting data of the actual target space and actual measurement data of the target space and a plurality of objects located in the target space, and Korean Published Patent No. 2020-0013297 does not descri