AI-based Robot for Grout Width Detection and Automatic Removal

An AI-based robot with computer vision and dust removal capabilities addresses worker health and tile damage issues by precisely measuring grout width and stopping operations at safety thresholds, enhancing safety and efficiency in tile grout removal.

KR103040904B1Active Publication Date: 2026-10-30김재훈
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
KR1020250186062
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-11-28
Publication Date
2026-10-30
Estimated Expiration
2045-11-28

AI Technical Summary

Technical Problem

Existing methods for tile grout removal cause worker fatigue and health issues due to dust exposure, and existing machines fail to detect irregular grout widths, leading to tile damage.

Method used

An AI-based robot that uses computer vision to distinguish between tiles and grout, measures grout width in real time, and stops or alarms upon exceeding safety thresholds, equipped with a suction unit to remove dust and maintain stability on inclined surfaces.

Benefits of technology

Prevents tile damage by detecting irregular grout widths and reducing dust exposure, ensuring high recognition rates and eco-friendly operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112025134591824-PAT00001_ABST
    Figure 112025134591824-PAT00001_ABST
Patent Text Reader

Abstract

The present invention relates to a robot device that automatically removes tile grout in a construction finishing process. Regarding this, clearly distinguishing 'tiles' and 'grout' through artificial intelligence (AI) image recognition It distinguishes and detects the joint width in real time, and sets an alarm in case of an abnormality. By providing this, tile damage is prevented at the source, and the damage occurring simultaneously with the work We provide a robot that protects the health of workers by immediately inhaling fine dust.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a robotic device for automatically removing tile grout in a construction finishing process, and more specifically, to an AI-based robot that uses computer vision technology to distinguish between tiles and grout, detects the width of the grout in real time, and provides an alarm in the event of an abnormality to prevent tile damage. Background Technology

[0003] Traditionally, tile grout removal was performed by workers using manual labor or power tools. However, this results in high worker fatigue and poses serious health problems, such as respiratory diseases caused by large amounts of airborne dust.

[0006] Above all, existing semi-automatic machines merely followed the grout lines and failed to detect sections where the width suddenly narrowed due to poor installation or other factors. Consequently, accidents frequently occurred where machines forcibly entered areas, scratching or breaking the surfaces of expensive tiles, creating an urgent need for intelligent safety technology to resolve this issue. The problem to be solved

[0008] The present invention is intended to solve the aforementioned problems,

[0011] First, it clearly distinguishes between 'tiles' and 'grout' through artificial intelligence (AI) image recognition, and precisely measures the width of the grout in real time; by immediately sounding an alarm or stopping if it exceeds the set range, it fundamentally prevents tile damage,

[0014] Second, the purpose is to provide a robot that protects the health of workers by immediately inhaling fine dust generated simultaneously with work. means of solving the problem

[0016] The present invention for achieving the above objective comprises: a main body; a joint removal unit coupled to the main body and including a rotary cutting blade; a driving module responsible for moving the main body; a sensor module for photographing the floor surface; a suction unit for collecting dust; and a control unit.

[0019] The control unit recognizes joint areas in captured images using an AI algorithm and calculates the width data of the recognized joints in real time. If the calculated width exceeds a preset safety range, the control unit immediately stops the operation and generates a warning alarm; it controls the removal process by tracking the center path of the joint only when the width is within the normal range. Effects of the invention

[0021] According to the present invention,

[0024] First, by monitoring the width of the joints in real time, tile breakage accidents that may occur in sections with irregular widths can be prevented in advance.

[0027] Second, by ensuring visibility with a blower nozzle and removing airborne dust with a rear suction device, a high recognition rate and eco-friendly operation are possible.

[0030] Third, stable posture control and driving are possible even on sloped surfaces such as bathroom floors through the tilt detection sensor. Brief explanation of the drawing

[0032] FIG. 1 is a perspective view showing the overall configuration of a robot according to the present invention. FIG. 2 is a cross-sectional view showing the detailed structure of a joint removal unit according to one embodiment of the present invention. FIG. 3 is a side view illustrating the arrangement of the driving module and the inclination detection sensor of the present invention. Specific details for implementing the invention

[0033] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings.

[0036] Referring to FIG. 1, the robot (10) of the present invention comprises a main body (20), a joint removal unit (30), a sensor module (40), a control unit (50), a driving module (70), and a dust suction unit (80).

[0039] The driving module (70) includes a multi-axis drive wheel (41) and is capable of moving forward, backward, left, and right and rotating, and is equipped with a tilt detection sensor (43) inside the main body to detect the tilt of the robot. The control unit (50) receives this signal and corrects the driving so that the robot does not deviate from the path even on an inclined surface.

[0042] The joint removal unit (30) includes a rotary cutting blade (31) and a cutting depth adjustment device (32). A blower nozzle (33) is positioned in front of the cutting blade in the direction of travel to secure a field of view for the sensor, and a dust suction unit (80) is provided at the bottom or rear of the main body to vacuum-suck and collect dust generated during cutting.

[0045] The most important function of the present invention is the safety algorithm of the control unit (50).

[0048] The control unit (50) detects the boundary line between the tile and the joint through pixel analysis in the image input from the sensor module (40) and calculates the 'joint width,' which is the distance between the two boundary lines, in real time.

[0051] If an abnormal section is detected where the joint width becomes narrower than the thickness of the cutting blade (31) or becomes too wide due to poor construction or other reasons, the control unit (50) determines this as a danger signal. At this time, the control unit (50) immediately stops the driving module (70) and the joint removal unit (30) and generates a warning alarm (sound or light) so that the operator can recognize it. Explanation of the symbols

[0053] 10: Robot body 20: Body case 30: Joint removal unit 31: Rotary cutting blade 32: Cutting depth adjustment device 33: Blower nozzle 40: Sensor module 41: Multi-axis drive wheel 43: Inclination detection sensor 50: Control unit 70: Driving Module 80: Dust Intake Unit

Claims

Claim 1 An AI-based joint width detection and automatic tile joint removal robot for a construction finishing process comprises: a main body (20); a tile joint removal unit (30) coupled to the main body and equipped with a rotary cutting blade (31); a driving module (70) for moving the main body; a sensor module (40) for acquiring an image of a work area; and a control unit (50) for processing information from the sensor module (40) to control operation. The control unit (50) is characterized by using an AI image recognition algorithm to identify a 'joint area' in an input image, calculating the width of the identified joint area in real time, immediately stopping the operation or generating a warning alarm if the calculated width deviates from a preset normal range, and controlling the joint removal unit (30) to perform a removal operation along the center path of the joint area only when the calculated width is within the normal range. Claim 2 An AI-based joint width detection and automatic tile joint removal robot according to claim 1, wherein the robot further includes a tilt detection sensor (43) for detecting the tilt of the robot, and the control unit (50) prevents deviation from the path on an inclined surface by correcting the steering of the driving module (70) based on the signal of the tilt detection sensor (43). Claim 3 An AI-based joint width detection and automatic tile joint removal robot according to claim 1, characterized in that a blower nozzle (33) is provided in front of the tile joint removal unit (30) to remove foreign matter from the joint area with air pressure before the sensor module (40) acquires an image, thereby increasing recognition accuracy. Claim 4 The AI-based joint width detection and automatic tile joint removal robot according to claim 1, characterized in that a dust suction unit (80) is provided at the lower or rear of the main body to vacuum suction and collect joint powder and flying dust crushed by the rotary cutting blade (31).

Citation Information

Patent Citations

  • Tile constructing method and apparatus for thereof

    KR1020240041113A