Method and device for ensuring worker safety

Abstract

A method for ensuring worker safety is disclosed. The method comprises the steps of: displaying key elements working in a specific work environment on a two-dimensional plane; generating a distribution map indicating the distribution location of each key element displayed on the two-dimensional plane; updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element; converting the updated distribution map into a binary map indicating the overlap of the distribution locations of the key elements using a conversion code; and evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for ensuring worker safety, and more specifically, to a method and apparatus for ensuring worker safety by utilizing a distribution map of key elements or major elements placed in a workspace. Background Technology

[0003] When workers perform tasks near robots or other equipment, there is a risk of collision between the person and the equipment, and such collisions can cause damage to the equipment and serious injury to the worker.

[0004] Figure 1 is a drawing intended to explain the problems of the prior art.

[0005] Referring to FIG. 1, in order to ensure the safety of a worker, conventional technology is applied such as (A) in FIG. 1, which measures the distance between a worker and a robot using a distance sensor, reduces the speed of the robot when the worker approaches a first area (Z1), and stops the robot when the user approaches a second area (Z2) that is closer to the robot. However, this technology has limitations in that it is difficult to prevent collisions between multiple robots and it determines safety based solely on the distance between the user and the robot without considering the user's movement information.

[0006] As shown in (B) of Fig. 1, technology has been developed to automatically stop the robot when it collides with a worker, but this technology has limitations in that it only operates after a collision has occurred and damage or injury has been sustained, thus limiting the safety guarantee effect.

[0007] Since collisions can occur not only between workers and equipment but also with the surrounding environment, such as walls or structures, and can be caused by movement as well as the relative distance between workers and equipment, a safety strategy capable of considering the location and movement of various elements in the work environment is required. The problem to be solved

[0009] The present invention, aimed at solving the aforementioned problems, is intended to provide a method and device for ensuring worker safety that can guarantee collision safety for workers in an environment where various equipment and workers coexist. means of solving the problem

[0011] A method for ensuring worker safety according to one aspect of the present invention for achieving the above-described purpose comprises: a step of displaying key elements working in a specific work environment on a two-dimensional plane; a step of generating a distribution map indicating the distribution location of each key element displayed on the two-dimensional plane; a step of updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element; a step of converting the updated distribution map into a binary map indicating the overlap of the distribution locations of the key elements using a conversion code; and a step of evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map.

[0012] A computing device according to another aspect of the present invention is a device for performing a method for ensuring worker safety, comprising a processor and a display device, wherein the processor processes a process of displaying key elements working in a specific work environment on a two-dimensional plane, a process of generating a distribution map indicating the distribution location of each key element displayed on the two-dimensional plane, a process of updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element, a process of converting the updated distribution map into a binary map indicating the overlap of the distribution locations of the key elements using a conversion code, and a process of evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map, and the display device visually displays the risk level of the evaluated key element. Effects of the invention

[0014] According to the present invention, collision safety judgment can be performed by considering all various key elements of the work environment. In addition, collision safety judgment can be performed by considering all relative distances and movements (e.g., speed) of multiple pieces of equipment and workers. Brief explanation of the drawing

[0016] Figure 1 is a drawing intended to explain the problems of the prior art. FIG. 2 is a drawing illustrating an exemplary work environment according to an embodiment of the present invention and a worker safety strategy in the exemplary work environment. Figure 3 is a diagram showing the main elements detected in the working environment of Figure 2 on a two-dimensional plane. FIG. 4 is a diagram showing a distribution map (A) according to an embodiment of the present invention and a graph (B) showing a change in distribution according to variable adjustment. Figures 5 and 6 are diagrams illustrating the movement reflection distribution update of Figure 2. FIG. 7 is a flowchart illustrating a method for ensuring worker safety according to an embodiment of the present invention. Figure 8 is a configuration diagram of a computing device for performing the method of Figure 7. Specific details for implementing the invention

[0017] The terms used in this specification are used merely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as “comprising” or “having” are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0019] FIG. 2 is a drawing illustrating an exemplary work environment according to an embodiment of the present invention and a worker safety strategy in the exemplary work environment.

[0020] Referring to FIG. 2, an exemplary worker environment (20) according to an embodiment of the present invention includes a two-degree-of-freedom robot (21) moving in a plane and one worker (22). That is, the worker safety securing method according to an embodiment of the present invention takes into account a two-degree-of-freedom robot (21) and a worker (22) on one side.

[0021] However, this is merely to limit the environment in which the worker safety securing method of the present invention is applied to an environment including a 2-degree-of-freedom robot manipulator (21) and one worker (22) in order to aid in understanding the explanation. Therefore, the present invention can be applied in the same way to an environment including multiple robot manipulators with higher degrees of freedom and multiple workers, which will be fully understood by those skilled in the art through the following explanation.

[0022] The method for ensuring worker safety according to an embodiment of the present invention proceeds in the order of detecting key elements, distributing and arranging distributions, updating distributions reflecting movement, integrating distributions, and evaluating risk (safety).

[0024] Figure 3 is a diagram showing the main elements detected in the working environment of Figure 2 on a two-dimensional plane.

[0025] Referring to FIG. 3, the key elements detected in the working environment of FIG. 2 can be represented by specific shapes indicating the positions of the 2-degree-of-freedom robot and the worker (22) in a two-dimensional plane. For example, the worker (22) can be represented by a circle (C), and the 2-degree-of-freedom robot can be represented by a straight line. Here, the key elements of the 2-degree-of-freedom robot represented by the straight line include the circle (C) representing the worker (22), the first joint (A) which can cause a collision, and the end portion (B) of the straight line extending from the first joint (A). Thus, the detection of key elements means the detection of the first joint (A), the end portion (B), and the circle (C).

[0026] To detect these key elements, methods of detection using markers or color tags, such as Apriltags attached to the 2-degree-of-freedom robot (21) and the worker (22), or methods of detection using deep learning may be used. For example, first, a marker of a specific shape (e.g., QR code, circular marker) or a tag of a specific color is attached to the worker and the 2-degree-of-freedom robot. Then, a camera is installed at a fixed position to recognize the positions of the worker and the 2-degree-of-freedom robot in a two-dimensional plane. The camera collects images in real time and tracks the position of the object by recognizing the color tag or marker. Then, the marker or color tag is recognized in the images collected by the camera through image processing technology. For example, a library such as OpenCV can be used to filter pixels of a specific color and apply an algorithm that recognizes a specific shape (e.g., an object recognition algorithm). At this time, to visualize the worker as a circle and the 2-degree-of-freedom robot as a straight line, the center of the corresponding position is calculated and displayed as a circle or straight line graphic. Afterwards, the recognized coordinate data is mapped onto a two-dimensional plane, and the operator's position is represented as a circle, while the 2-degree-of-freedom robot is represented as a straight line. The operator's center coordinates are set to the center of the circle, and the robot's coordinates are represented as straight lines according to the joint angles.

[0028] FIG. 4 is a diagram showing a distribution map (a) according to an embodiment of the present invention and a graph (b) showing a change in distribution according to variable adjustment.

[0029] Referring to Fig. 4, a distribution map is generated that represents the distribution locations (position density) of the key elements displayed on a two-dimensional plane detected in the previous step. At this time, a Gaussian distribution may be used, but is not limited thereto, and other forms of distribution may be used. For example, the distribution map may model the position density by applying a Gaussian distribution function to the position information (e.g., two-dimensional coordinate data) of each of the detected key elements. Here, position density represents the position density, or distribution, of the key elements distributed at specific locations in a two-dimensional plane or three-dimensional space.

[0030] A Gaussian distribution (Gaussian density) is applied to each location of the key elements (1st joint (A), end part (B), worker (C)) in the distribution map to create a circle that becomes fainter as it moves away from the center. This allows the density around a specific point to be visualized as a circle, as shown in Figure 4 (a).

[0031] As shown in FIG. 4 (b), the distribution can be adjusted not only in position but also in size (e.g., distribution area) and intensity (e.g., brightness) through variable adjustment. In this embodiment, the distribution (z1) of the robot's first joint (A) has a small distribution area on the distribution map because there is a relatively low risk of collision with the worker, and the distribution (z2) of the end portion (B) of the straight line extending from the first joint (A) has a large distribution area because there is a relatively high risk of collision with the worker.

[0032] A distribution represents the location (or location density) of each major element. Therefore, if multiple distributions overlap, it may be determined that there is a high risk of collision at that location. The safety strategy proposed in the embodiment of the present invention first places distributions representing the respective locations of each major element on a distribution map and aggregates them to generate a single distribution map as shown in FIG. 4 (a). At this time, the distribution map (z) can be generated as shown in the following equation (1).

[0033]

[0034] Here, z i is the distribution (values) representing the positions (values) of each principal element, and n is the total number of principal elements. Here, each distribution z i It is normalized to have a size of 0 to 1, and thus the entire distribution z calculated through Equation (1) also has a size of 0 to 1.

[0035] Using the distribution z generated in this way, a binary map b representing the overlap of distributions (distribution locations) can be generated through the exemplary code (e.g., pseudo code) below.

[0037] input: width, height, threshold, z

[0038] output: b

[0039] for i = 0:width

[0040] for j = 0:height

[0041] if z[i][j] > threshold: b[i][j] = 1

[0042] else: b[i][j] = 0

[0043] return b

[0045] In the code above, width is the width of the binary map (number of pixels in the horizontal direction), height is the height of the binary map (number of pixels in the vertical direction), threshold is a threshold value that determines the value of the binary map based on whether it is greater or smaller than this value, z is a distribution (value) representing the location of the principal elements and includes the value for each pixel. b is the binary map to be generated (a 2D array consisting of 0s or 1s).

[0046] In the code above, “for i = 0:width” and “for j = 0:height” are double loops that iterate through each pixel. Here, i represents the horizontal direction (width) and j represents the vertical direction (height).

[0047] In the code above, “if z[i][j] > threshold: b[i][j] = 1” and “else: b[i][j] = 0” are conditional statements that compare whether the value at position i, j of z is greater than the given threshold. If the value is greater than the threshold, 1 is stored in b[i][j], and otherwise, 0 is stored in b[i][j].

[0048] In the code above, “return b” means that after examining all pixels, it returns the binary map b.

[0049] In this way, the above code generates a binary map b consisting of 1 or 0 based on the element value (pixel value) corresponding to the distribution location of each major element in the distribution map, depending on whether the element value (pixel value) of each major element exceeds a specific threshold.

[0050] Meanwhile, the sensitivity of the safety diagnosis can be evaluated by changing the threshold of the above code, and finally, the size of the binary map b obtained using the summation function is calculated (distribution integration in Fig. 2), and the presence or absence of safety can be detected by comparing it with a preset safety diagnosis threshold (risk (safety) evaluation in Fig. 2), where a larger sum value indicates a higher risk.

[0051] Here, the size of binary map b may refer to the number of elements (or pixels) satisfying a specific threshold condition in binary map b. In this case, the number of elements (pixels) satisfying a specific condition can be obtained by summing the number of specific values ​​satisfying the condition, for example, 1, within binary map b using a summing function. To explain this in relation to risk, if the size of binary map b—that is, the sum of the number of elements (or pixels) satisfying a specific threshold condition—is greater than a preset safety diagnosis threshold, it is determined that the risk is high.

[0053] Figures 5 and 6 are diagrams illustrating the movement reflection distribution update of Figure 2.

[0054] Referring to FIGS. 5 and 6, FIG. 4(a) illustrates a distribution map considering only the relative positions between each major element; however, if only the relative positions between each major element are considered, the safety guarantee effect is limited. Accordingly,

[0055] As shown in Fig. 5(a), the 2-degree-of-freedom robot maintains a certain distance from the operator, but depending on the movement speed of the 2-degree-of-freedom robot (V, arrow in Fig. 5(a)), the manipulator of the 2-degree-of-freedom robot may approach the operator quickly and a collision may occur soon.

[0056] When the safety of a worker is evaluated based on a distribution map that reflects only relative positions as in Fig. 4 (a), the two distributions consisting of the worker and the end portion of the 2-DOF robot manipulator are far apart as in Fig. 5 (b), so no overlap occurs in the binary map as in Fig. 5 (c). Therefore, the sum value representing the size of the binary map becomes 0, which may lead to a misjudgment that the current state is safe.

[0057] In this embodiment, to improve safety evaluation performance for such situations, a motion-reflecting distribution update (update of the distribution map) is further added to adjust the distribution positions of the operator and the end portion of the 2-DOF robot manipulator on the distribution map as follows. Equation 2 below represents the adjustment of the distribution positions.

[0059]

[0060] Here, α is the gain value, and v x , v y is the velocity value of the corresponding key element, and, z x and z y is the location (distribution location) of the corresponding major element. Figure 6 shows the safety evaluation result calculated using the above equation (2).

[0061] As can be seen in Fig. 6(b), the distribution corresponding to the end of the manipulator of the 2-DOF robot moves toward the operator in reflection of the robot's movement (manipulator's velocity), causing overlap, and accordingly, the overlap is displayed on the binary map in Fig. 6(c). Accordingly, a sum value exceeding a preset safety diagnostic threshold occurs, allowing for successful detection of risk factors.

[0062] In this way, using the method of the present invention, not only the magnitude of the danger (binary map summation size) but also the predicted location of the danger can be inferred through the binary map. Meanwhile, velocity information of key elements (e.g., the velocity of the end section of a manipulator of a 2-degree-of-freedom robot) can be easily obtained through the system's joint sensors and kinematics, or external vision sensors.

[0064] FIG. 7 is a flowchart illustrating a method for ensuring worker safety according to an embodiment of the present invention.

[0065] Referring to FIG. 7, in step 710, a step is performed to display the key elements working in a specific work environment on a two-dimensional plane.

[0066] Next, in step 720, a step of generating a distribution map indicating the distribution location of each major element displayed in the two-dimensional plane is performed.

[0067] Next, in step 730, a step of updating the distribution map is performed by adjusting the distribution position of each major element based on the movement speed of each major element.

[0068] Next, in step 740, the updated distribution map is converted into a binary map representing the overlap of the distribution locations of the major elements using a conversion code.

[0069] Next, in step 750, a step is performed to evaluate the risk level of the key element corresponding to the worker among the key elements based on the sum of the pixel values ​​included in the converted binary map.

[0070] In an embodiment, step 710 includes the step of recognizing the location of a marker or a tag of a specific color attached to the main elements using a vision sensor, and the step of displaying the result recognized by the vision sensor as a specific shape on the two-dimensional plane. Here, the operator may be represented by a circle, and the robot manipulator may be represented by points representing joints and straight lines connecting the points.

[0071] In an embodiment, step 720 includes the step of generating a distribution map representing the distribution location of each major element by applying a Gaussian distribution function to the location information of each major element displayed in the two-dimensional plane.

[0072] In the embodiment, step 730 is the distribution location (z) of each major element. x , z y The velocity value (v) of each major element in ) x , v y It includes a step of adjusting the distribution position of each of the above major elements by summing the values.

[0073] In the embodiment, the velocity value of each key element is obtained through a joint sensor or an external vision sensor attached to each key element.

[0074] In an embodiment, step 740 may be a step of converting the pixel value corresponding to the distribution location of each major element in the updated distribution map into a binary map composed of 1 or 0 depending on whether it exceeds a specific threshold.

[0075] In an embodiment, step 750 may be a step of evaluating the risk level by comparing the sum value with a preset safety diagnosis threshold.

[0077] Figure 8 is a configuration diagram of a computing device for performing the method of Figure 7.

[0078] Referring to FIG. 8, the computing device (200) includes a processor (210), memory (220), input interface (230), display device (240), communication device (250), storage device (260), and system bus (270).

[0079] The processor (210) is a device that processes operations performed in each step as the subject of steps S710 to S750 in the method of FIG. 7, and includes, for example, a CPU and / or a GPU.

[0080] In an embodiment, the processor (210) processes the process of displaying key elements working in a specific work environment on a two-dimensional plane, the process of generating a distribution map showing the distribution location of each key element displayed on the two-dimensional plane, the process of updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element, the process of converting the updated distribution map into a binary map showing the overlap of the distribution locations of the key elements using a conversion code, and the process of evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map.

[0081] The memory (220) is a device for temporarily and / or permanently storing program instructions for performing steps S710 to S750 in the method of FIG. 7, and includes volatile memory and / or non-volatile memory.

[0082] The input interface (230) is a device that transmits user input to a processor to perform steps S710 to 750 in the method of FIG. 7, and includes a keyboard, a touch screen, etc.

[0083] The display device (240) is a device that visually displays intermediate data and / or result data processed at each step in the method of FIG. 7, and, for example, can visually display the risk level of a key element corresponding to a worker evaluated by the processor (210).

[0084] The communication device (250) may be a device that transmits intermediate data and / or result data processed at each step of the method of FIG. 7 to an external device using wired or wireless communication. Here, wireless communication includes Wi-Fi, Bluetooth, internet communication, mobile communication, etc.

[0085] The storage device (260) may be a device that permanently holds various programs and an operating system for executing these programs to perform steps S710 to S750 in the method of FIG. 7, such as a hard disk.

[0086] The system bus (270) serves as a data communication path between the above components (210 to 260).

[0087] The embodiments disclosed in this specification should be considered in an exemplary sense for the sake of illustration rather than in a limiting sense. The scope of the invention is defined by the claims, not by the foregoing description, and all variations within the scope of the claims should be interpreted as being included in the invention.

Claims

Claim 1 A method for ensuring worker safety, comprising: a step of displaying key elements working in a specific work environment on a two-dimensional plane; a step of generating a distribution map indicating the distribution location of each key element displayed on the two-dimensional plane; a step of updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element; a step of converting the updated distribution map into a binary map indicating the overlap of the distribution locations of the key elements using a conversion code; and a step of evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map. Claim 2 A method for ensuring worker safety according to claim 1, wherein the step of displaying the main elements on a two-dimensional plane comprises: a step of recognizing the position of a marker or a tag of a specific color attached to the main elements using a vision sensor; and a step of displaying the result recognized by the vision sensor as a specific shape on the two-dimensional plane. Claim 3 A method for ensuring worker safety according to claim 1, wherein the step of generating the distribution map comprises the step of generating a distribution map representing the distribution location of each major element by applying a Gaussian distribution function to the location information of each major element displayed in the two-dimensional plane. Claim 4 A method for ensuring worker safety according to claim 1, wherein the step of updating the distribution map includes the step of adjusting the distribution position of each major element by summing the velocity value of each major element at the distribution position of each major element. Claim 5 A method for ensuring worker safety in paragraph 4, wherein the speed value of each major element is obtained through an articulation sensor or an external vision sensor attached to each major element. Claim 6 A method for ensuring worker safety according to claim 1, wherein the step of converting to a binary map is a step of converting to a binary map composed of 1 or 0 depending on whether the pixel value corresponding to the distribution position of each major element in the updated distribution map exceeds a specific threshold. Claim 7 A method for ensuring worker safety according to claim 1, wherein the step of evaluating the risk level is a step of evaluating the risk level by comparing the sum value with a preset safety diagnosis threshold value. Claim 8 A computing device for performing a method to ensure worker safety, wherein the computing device includes a processor and a display device, and the processor processes a process of displaying key elements working in a specific work environment on a two-dimensional plane; a process of generating a distribution map indicating the distribution location of each key element displayed on the two-dimensional plane; a process of updating the distribution map by adjusting the distribution location of each key element based on the movement speed of each key element; a process of converting the updated distribution map into a binary map indicating the overlap of the distribution locations of the key elements using a conversion code; and a process of evaluating the risk level of a key element corresponding to a worker among the key elements based on the sum of pixel values ​​included in the converted binary map, and the display device visually displays the risk level of the evaluated key element.

Images