KR-20260067573-A - METHOD, COMPUTING DEVICE AND COMPUTER PROGRAM FOR HIGH-PRECISION POSITION ESTIMATION OF ROBOT USING HIERARCHY MAP

KR20260067573AKR 20260067573 AKR20260067573 AKR 20260067573AKR-20260067573-A

Abstract

A method, apparatus, and computer program for high-precision position estimation of a robot using a hierarchical map are provided. The method for high-precision position estimation of a robot using a hierarchical map is performed by a computing device and comprises the steps of generating a map for a predetermined area and performing high-precision position estimation of a target robot using the generated map. The generated map comprises a plurality of map layers having different resolutions for the predetermined area, and the plurality of map layers have a hierarchical structure in which they are sequentially connected according to resolution.

Inventors

정진용

Assignees

리보틱스(주)

Dates

Publication Date: 20260513
Application Date: 20241106

Claims (6)

In a method performed by a computing device, A step of generating a map for a predetermined area; and The method includes the step of performing high-precision position estimation for a target robot using the generated map above, The map generated above is, Characterized by including a plurality of map layers having different resolutions for the aforementioned predetermined area, wherein the plurality of map layers have a hierarchical structure in which they are sequentially connected according to resolution. High-precision robot position estimation method using a hierarchical map.
In paragraph 1, The above plurality of map layers are, It includes a low-resolution map layer having low resolution, a medium-resolution map layer having medium resolution, and a high-resolution map layer having high resolution, The above low-resolution map layer is, It is used for location initialization, consists of a single map containing a map of all areas, and The above medium-resolution map layer is, The entire map is divided into submaps of a certain size for management, and the center points of the submaps are managed by constructing a KD-tree; based on the location estimated using the low-resolution map layer, submaps of a certain distance are loaded and used to correct the location. The above high-resolution map layer is, Characterized by generating a high-resolution map patch for a specified location according to user settings, constructing a KD-tree with the center point of the high-resolution map patch, and using it to correct the position when the target robot enters the patch. High-precision robot position estimation method using a hierarchical map.
In paragraph 1, The step of generating the above map is, A step of creating a low-resolution map layer having low resolution, wherein, in the process of creating the low-resolution map layer, graph data is maintained based on graph SLAM and sensor data is stored in each node; After the low-resolution map layer is generated, a correction is performed on the graph data at the loop closing point, and a medium-resolution map layer having medium resolution is generated based on the corrected graph data and the stored sensor data, wherein the generated medium-resolution map layer is divided into submap units and stored; and When a specific area is designated by a user as an area requiring a high-resolution map based on the low-resolution map layer generated above, the method includes the step of generating a high-resolution map layer having high resolution corresponding to the specific area based on sensor data for the specific area. High-precision robot position estimation method using a hierarchical map.
In paragraph 2, The step of performing high-precision position estimation for the above-mentioned target robot is, A step of estimating the initial position of the target robot using the low-resolution map layer; A step of generating an integrated sub-map by loading a sub-map of a medium-resolution map layer within a certain distance based on the estimated initial position, and first re-estimating the position of the target robot using the generated integrated sub-map; A step of continuously updating the integrated submap by removing submaps that have moved away according to the movement of the target robot and adding newly closer submaps; and If it is determined that the target robot has approached the inside of a patch of the high-resolution map layer using the KD-tree of the high-resolution map layer, the method includes the step of secondarily re-estimating the position of the target robot through the high-resolution map layer using the first re-estimated position as an initial value. High-precision robot position estimation method using a hierarchical map.
processor; Network interface; Memory; and It includes a computer program that is loaded into the memory and executed by the processor, The above computer program is, Instructions for generating a map of a specified area; and It includes instructions for performing high-precision position estimation for a target robot using the map generated above, and The map generated above is, Characterized by including a plurality of map layers having different resolutions for the aforementioned predetermined area, wherein the plurality of map layers have a hierarchical structure in which they are sequentially connected according to resolution. A computing device that performs a high-precision position estimation method for a robot using a hierarchical map.
Combined with a computing device, A step of generating a map for a predetermined area; and A computer program stored on a recording medium readable by a computing device for executing a method for estimating the high-precision position of a robot using a hierarchical map, comprising the step of performing high-precision position estimation for a target robot using the generated map - the generated map includes a plurality of map layers having different resolutions for the predetermined area, and the plurality of map layers have a hierarchical structure in which they are sequentially connected according to resolution.

Description

Method, Computing Device and Computer Program for High-Precision Position Estimation of Robot Using Hierarchy Map Various embodiments of the present disclosure relate to a method, apparatus, and computer program for high-precision position estimation of a robot using a hierarchical map. More specifically, the present disclosure relates to a method, apparatus, and computer program for high-precision position estimation of a robot using a hierarchical map, which provides a method for high-precision position estimation of a robot using a hierarchical map to overcome the limitations of the Grid Map method and enable accurate position estimation of the robot. The Grid Map method (e.g., FIGS. 1 to 3) is a method that divides a space into several fixed grids and represents a map based on the occupancy status of each grid. This method is frequently used for robots to perceive the environment and estimate their position. In the Grid Map method, sensor measurements, such as data from a LiDAR (Light Detection and Ranging) sensor, are matched with the occupancy data of the grid map, and the robot's position is estimated by calculating an occupancy score based on this. A characteristic of this Grid Map is that the accuracy of position estimation varies depending on the resolution of the grids that constitute the environment. However, the Grid Map method has several limitations. First, if the grid size is too small relative to the spatial area, the number of grid cells becomes excessive, leading to a larger map data size. As the map data size increases, the processing time and memory usage also rise, placing a burden on system performance. Conversely, as the grid size increases, a problem may arise where the precision of location estimation decreases. In particular, for applications requiring precise location estimation, it is sometimes difficult to ensure sufficient accuracy using only the Grid Map method. For example, in situations requiring high positional accuracy, such as when a robot performs docking operations, relying solely on grid map-based position estimation methods may encounter limitations. In such cases, auxiliary technologies like April Tags or QR codes are sometimes applied alongside the grid map method to supplement positional accuracy. Position correction methods based on April Tags or QR codes guarantee high positional precision at specific points, offering the advantage of overcoming the limitations of the grid map approach. The aforementioned background technology is one that the inventor possessed or acquired in the process of deriving the content of the present disclosure, and it cannot be considered as prior art disclosed to the general public prior to the filing of this application. The following drawings attached to this specification illustrate preferred embodiments of the present disclosure and serve to further enhance understanding of the technical concept of the present disclosure together with the detailed description of the invention; therefore, the present disclosure should not be interpreted as being limited only to the matters described in such drawings. FIGS. 1 to 3 are drawings illustrating a conventional Grid Map method for estimating location. FIG. 4 is a diagram illustrating, exemplarily, the structure of a map generated according to various embodiments of the present disclosure. FIG. 5 is a diagram illustrating a low-resolution map layer on which a pose graph of a robot is displayed in various embodiments. FIG. 6 is a diagram illustrating a high-resolution area selected by a user through a low-resolution map layer in various embodiments. FIG. 7 is a drawing illustrating a completed high-resolution map patch in various embodiments. FIG. 8 is a diagram illustrating, in various embodiments, an exemplary kd-tree of a submap for fast searching. Figure 9 is a diagram illustrating a conventional 2D Localization algorithm. Figure 10 is a diagram showing a cost map generated by transforming a 2D occupancy map. FIG. 11 is a drawing illustrating the hardware configuration of a computing device according to various embodiments of the present disclosure. The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure is complete and to fully inform those skilled in the art of the scope of the present disclosure, and the present disclosure is defined only by the scope of the claims. The terms used herein are for describing the embodiments and are not intended to limit the disclosure. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and/or "comprising" do not exclu