JP7581576B2 - How to operate an overhead crane equipped with a human detection system - Google Patents

Description

The present invention relates to a method for operating an overhead crane equipped with a human detection system that prevents accidents such as collisions between a worker and the crane's suspended load in the crane work area.

Conventionally, a crane collision prevention device has been proposed that uses a distance sensor to determine the distance to an object, calculates the object's moving speed based on the change in distance over time, and obtains the relative speed of the crane to perform collision prevention control (see, for example, Patent Document 1).

The applicant of the present application has previously proposed the following devices 1 to 5 using a range finder, i.e., a scanning type light wave distance meter capable of outputting physical shape data of a space, also called "light detection and ranging" or "laser imaging detection and ranging" (LIDAR, an abbreviation for Light Detection and Ranging or Laser Imaging Detection and Ranging. It is a type of remote sensing technology that uses light, and measures scattered light in response to pulsed laser irradiation to analyze the distance to a distant object and the shape and properties of the object.): 1. A device that, when creating a three-dimensional map of a crane work area, determines whether a blind spot has occurred and performs a scanning operation again at a position that compensates for the blind spot to create a three-dimensional map without blind spots (see Patent Document 2).
2. A device that can check the safety of a crane work area by using a simple equipment configuration, taking measures against disturbance light, and reliably extracting the worker without being affected by the crane work environment (see Patent Document 3).
3. A device with a simple equipment configuration that determines the size of the object being handled by the crane when it is hoisted, and changes the deceleration distance depending on that information to prevent collision with objects such as workers (see Patent Document 4). 4. A device with a simple equipment configuration that enables three-dimensional measurements of the object being handled and the object to be performed in real time, determines the size of the object being handled when the crane is hoisted, and decelerates and stops depending on that information to prevent collision with objects including moving objects such as workers (see Patent Document 5).
5. A safety device for a crane that can directly detect the inclination of an object being handled by a crane by making it possible to perform three-dimensional measurement of the object in real time with a simple equipment configuration (see Patent Document 6).

JP 2017-71455 A JP 2019-127372 A JP 2019-127374 A JP 2019-127375 A JP 2021-54584 A JP 2021-54585 A

However, with the above-mentioned conventional crane safety devices, it can be difficult to detect an operator if the camera's field of view is blocked by luggage when the operator is working near the suspended load during hoisting, or if the operator is under the suspended load during transport and is no longer visible on the camera.

In view of the problem with the above-mentioned conventional crane safety devices, particularly the human detection system, that it becomes difficult to detect the worker due to the load being handled by the crane, the present invention aims to provide a method for operating an overhead crane equipped with a human detection system that uses a human detection system composed of simple equipment such as multiple cameras, etc., to prevent accidents such as collisions between the worker and the load suspended by the crane in the crane work area.

In order to solve the above problems, the method of operating an overhead crane equipped with a human detection system of the present invention is characterized in that it combines the fields of view and coordinates of multiple cameras into a reference coordinate centered on the hook, traces people within a range where they can be detected, sets up a security zone in which it is determined that the person is in danger if the person's position information is suddenly lost, and if all cameras tracing a person within the security zone lose sight of the person twice in a row, it is determined that the person has disappeared, and the crane is stopped along with an announcement to that effect.

In this case, the control device that operates the overhead crane from the ground can be equipped with buttons for enabling and disabling the human detection system, inspecting the human detection system, and for slinging operations, allowing the range for human detection to be switched and the human detection system to be inspected depending on whether a load is suspended or not.

The overhead crane can also be equipped with a human detection status indicator light and a number of people indicator that displays the number of people detected by the human detection system, thereby displaying the status of the human detection system.

According to the method of operating an overhead crane equipped with a human detection system of the present invention, after the crane power is turned on, images are taken with a camera at short time intervals of several times per second, for example, every 250 ms, and people are traced in areas where they can be detected. A security zone is set up in the area close to the suspended load, and if all cameras tracing a person in the security zone fail to detect them twice in a row, it is assumed that the person's position information has been lost and an announcement is made, and the crane is stopped. With a simple equipment configuration, collisions with objects such as workers can be prevented when blind spots occur due to the items being handled by the crane.

1 is a system configuration diagram showing an embodiment of a control device for implementing a method for operating an overhead crane equipped with a human detection system of the present invention. FIG. FIG. 1 is a diagram showing the field of view of two cameras. This is a diagram of the concept of human disappearance. FIG. FIG. FIG. 13 is a diagram of a human detection count display and an example of the display. FIG. 13 is a diagram showing camera dirt determination. This is a diagram of the concept of suspended load size (circumscribed circle). FIG. 13 is a diagram of a human detection management area. FIG. 1 is a diagram of the safety area of the hoisting operation. FIG. 1 is a diagram of a safe area for traverse motion. FIG. 1 is a diagram of a safe area for running motion. FIG. 1 is a diagram of the safety area for simultaneous lateral travel operations. FIG. 1 is a diagram illustrating the concept of simultaneous traverse and travel operation. This is a diagram of the concept of suspended load transport height. FIG. 13 is a diagram of rope length correction. FIG. 1 is a diagram illustrating the concept of crane speed. This is a diagram showing a normal person disappearance timing chart. This is a diagram of the first person disappearance and second person detection in the person disappearance timing chart. This is a diagram showing the first and second disappearances in the person disappearance timing chart.

Below, an embodiment of the method for operating an overhead crane of the present invention will be explained with reference to the drawings.

Figures 1 to 20 show one embodiment of a method for operating an overhead crane equipped with the human detection system of the present invention.

FIG. 1 is a system configuration diagram showing an embodiment of a control device for implementing a method for operating an overhead crane equipped with a human detection system according to the present invention.
Here, the load is registered in advance as its maximum standard size (length, width, height) in the computer 16 via the touch panel 17, the shape of the load is defined as a circumscribing circle with a radius from the center of the load to the farthest point as shown in Figure 8, and the distance from the end of the load to the person is calculated. During slinging work, while the "sling" button on the wireless controller in Figure 4 is pressed, calculations are made on the distance from the hook to the person.

Overhead cranes equipped with human detection systems are inspected before crane operations begin, and inspection work is carried out to check the number of people detected and to determine whether the human detection system and the workers' helmets are dirty.
The inspection method is to press the “Human Detection” selection button on the wireless controller in Figure 4 to enable or disable the human detection system, and after confirming that it is enabled by the blue light on the human detection status indicator (signal tower) in Figure 5, press the inspection button to start the inspection.
During inspection, the human detection status indicator lights in Figure 5, which show the status of the human detection system, are displayed in order from top to bottom, and after a lamp test check of all the indicator lights is performed, inspection is carried out according to the announcements.
To determine whether the helmet mark is dirty, the images taken by the human detection cameras (human detection camera 1, human detection camera 2) 18, 20 showing the ON/OFF infrared spotlights in the system configuration diagram of Figure 1 are imported into a personal computer, the number detected is compared with the number of helmets installed under the crane, and if the number detected is less than the number installed, it is determined to be dirty.
To determine whether the camera and floodlight are dirty, the crane is moved for 10 seconds, and dirt is detected if a black object is always attached to the same position within the camera's field of view in the video captured by the camera as shown in Figure 7, even though the background changes due to the movement of the crane.When personnel are replaced or increased during work, the number of detected personnel is constantly checked using the person detection number display in Figure 6 as an inspection response only to the helmet condition.

After the camera inspection is completed before the crane begins operation, the human detection system switches to a detectable state, and the area in which humans can be detected is divided into the nine human detection management areas shown in Figure 9. For each operation, the crane operates at a speed that will not collide with the worker, while determining whether or not each area is a safe area.
The safe area is defined as a location where the human detection system judges it safe to detect a human, and differs for each crane operation. For example, the safe area for the lifting operation of a load is set to a location 2000 mm or more away from the end of the load, as shown in Figure 10. Here, "2000 mm from the end of the load" is based on the fact that "How to Proceed with Safe Rigging Operation (2)" published in the April 2017 issue of the monthly magazine "Crane" states that 2000 mm from the end of the load is a safe evacuation position, and is defined as 2000 mm here and can be changed by parameters.

The human disappearance function for lifting and lowering is as shown in Figure 2. Two cameras are installed diagonally above the club, each with its own field of view and coordinates, and in order to accurately measure the relative positions of people, the coordinates of each are combined into one reference coordinate. In the case of a crane, when a person moves around the load, the helmet mark is blocked by the wire rope or hook and goes undetected with one camera, so two cameras are installed diagonally above the club to eliminate blind spots. In coordinate combination, people within 750 mm + β are considered to be the same person, and 750 mm is assumed to be the shoulder width of a person. β is set to -250 to 250 mm. The human disappearance function sets up a warning area around the load that can be judged as dangerous if a person disappears, as shown in Figure 3, and after the crane power is turned on, two cameras take pictures at short time intervals of several times per second, for example, every 250 ms. The human disappearance function detects the three states shown in Figures 18 to 20, and when the state shown in Figure 20 is reached, the person disappears.
Specifically, the timing chart in Fig. 18 shows a normal state in which the person disappearance function is not operating. After the power to the crane is turned on, images are taken every 250 ms with cameras 1 and 2 when the floodlights are on and off, and the coordinates are synthesized. Next, people whose positions in the first and second shots are close to each other are considered to be the same person, and under normal circumstances, people are traced by comparing the distance from the previous position in the order of the shots taken from the second shot onwards.
The timing chart in Fig. 19 shows the state of person disappearance in the first shot and person detection in the second shot. The second and third shots are compared, and if there is no person nearby, it is determined that person disappearance occurred in the first shot. In the fourth shot, the distance between the second and fourth shots, in which a person was traced, is compared, and if a person is detected nearby, tracing of the person continues.
The timing chart in Figure 20 shows the first and second human disappearance states. The second and third shots are compared, and if there is no person nearby, it is determined that the first human disappearance has occurred. In the fourth shot, the distance between the second and fourth shots, which were tracing the person, is compared, and if there is no person nearby, it is determined that the second human disappearance has occurred, and the computer notifies the PLC control panel to stop the crane. Even if a person is detected in the fifth shot, the crane will be stopped.
The condition for a person disappearance is when, while tracing a person within the security area as shown in Table 1, neither camera 1 nor camera 2 detects a person twice in a row.
This makes it possible to prevent false detections caused by noise in the images captured by the human detection system while ensuring safety.
In this embodiment, the case where the required field of view can be covered with two cameras has been described, but if the required field of view cannot be covered with two cameras, the number of cameras can be three or more. In this case, too, if all cameras tracing a person within the security area lose sight of the person twice in a row, it is determined that the person has disappeared.

In the traverse alone operation of FIG. 11, the system designates the management areas "(1), (3), (4), (6), (7), (8), (9) and part of (5)" as the safe area.
The stop should be within a range of 2000 mm in front of both the left and right ends of the load in the direction of travel, and should also include the area behind the load, taking into consideration the possibility of people jumping out from the side or entering under the load from the rear.
In Fig. 11, if a person is detected in area (2) ahead in the direction of travel, it is determined that there is a possibility of danger to the person, so the crane switches to a slow speed and the yellow lamp of the person detection status indicator in Fig. 5 flashes. If a person is still detected in area (5), the crane stops and the yellow lamp lights up.
In the case of driving alone as shown in Fig. 12, if the safety area changes and a person is detected ahead in area (4) in the direction of travel, it is determined that there is a possibility of danger to the person, so the crane switches to a low speed and the yellow lamp of the person detection status indicator in Fig. 5 flashes. If a person is still detected in area (5), the crane stops and the yellow lamp lights up.

In the case of simultaneous traverse and travel operations, the management areas "(3), (6), (7), (8), (9) and part of (5)" in FIG. 13 are designated as the safety areas.
In Fig. 13, if a person is detected in areas (1), (2), or (4) ahead in the direction of travel, it is determined that there is a possibility of danger to the person, so the crane switches to a slow speed and the yellow lamp of the person detection status indicator in Fig. 5 flashes. If a person is still detected in area (5), the crane stops and the yellow lamp lights up.
When a safety area is set as in the management area of FIG. 13 for simultaneous traverse and travel operations, the safety area setting remains as in FIG. 13 for simultaneous traverse and travel operations until the movement ends so that the machine can reliably detect a person and stop the machine even if the operator releases his or her hand from either the traverse or travel button during simultaneous traverse operations as in FIG. 14.

The transport height at which a person will not collide with a suspended load when hoisting the load can be calculated as follows, assuming that half the height of the suspended load registered on the touch panel as shown in FIG. 15 is h/2, the hook rope length calculated from the hoisting speed is Lf, the sling rope length set in advance is Ls, and the distance from the pendulum fulcrum position to the floor surface is H:
Conveying height = H - (h / 2 + Ls + Lf)
If the transport height does not reach 2000 mm, a height at which the crane will not hit a person, the crane will operate at a low speed, and if the transport height reaches 2000 mm, the white lamp of the human detection status indicator in Figure 5 will light up, allowing high-speed operation.
For example, if the height h of the suspended load is set to 2000 mm, half the height of the suspended load (h/2) is 1000 mm. If the hook rope length Lf calculated from the hoisting speed is 8000 mm, the sling rope length Ls is 2000 mm, and the distance H from the pendulum fulcrum position to the floor surface is 12000 mm, then the transport height is
Conveying height = 12000 mm - (1000 mm + 2000 mm + 8000 mm) = 1000 mm
Since this is lower than 2000 mm, the crane operates at a slow speed.
When the hook rope length Lf is increased to 6000 mm by winding from this state, the conveying height is:
Conveying height = 12000 mm - (1000 mm + 2000 mm + 6000 mm) = 3000 mm
Since this is higher than 2000 mm, the crane operates at a high speed.
The transport height of the suspended load is corrected based on the upper limit operating position or the height at which the load is detected by the load cell as shown in FIG.

The travel and traverse speeds at which a person will not collide with a suspended load are calculated based on the information on the maximum suspended load size registered on the touch panel, and the distances (1) to (6) from the camera field of view to the person are calculated as shown in Figure 17, and the speed at which the person will not collide with the suspended load is determined as the stopping distance of the crane. This is the speed at which the person will not collide with the suspended load even when the crane is operating at high speed, regardless of the transport height of 2000 mm.
Crane stopping distance = [(1) Half the field of view of one camera] + [(2) Camera installation distance in the direction of movement] - ([(3) Radius from the center of the load to the farthest point] + [(4) Distance traveled due to human detection processing time] + [(5) Average human height (similar)] + [(6) Safety distance from the person to the end of the load])
For example, if (1) half the field of view of one camera is 7,500 mm, (2) the camera mounting distance in the direction of movement is 1,500 mm, (3) the radius from the center of the load to the farthest point is 2,500 mm, (4) the movement distance due to the human detection processing time is 1,000 mm, (5) the average height of a person (similar) is 1,700 mm, and (6) the safety distance from a person to the end of the load is 2,000 mm, then the crane stopping distance is:
Crane stopping distance = [(1) 7500 mm] + [(2) 1500 mm] - ([(3) 2500 mm] + [(4) 1000 mm] + [(5) 1700 mm] + [(6) 2000 mm]) = 1800 mm
It becomes.
If the rated speed of a crane is 1 m/sec, the deceleration time is 6 seconds. Speed and deceleration time are proportional to each other, and the crane stopping distance is calculated as 1/2 * [crane speed] * [deceleration time].
Therefore, the relationship between the crane speed, deceleration time, and crane stopping distance is shown in Table 2.

In Table 2, the speed at which the crane stopping distance is 1,800 mm or less is the speed at which a person will not be hit by the suspended load, so the speed at which a person will not be hit by the suspended load is 0.7 m/s.

The above describes the method of operating an overhead crane equipped with a human detection system of the present invention based on its embodiment, but the present invention is not limited to the configuration described in the above embodiment, and the configuration can be changed as appropriate within the scope of the spirit of the present invention.

The method of operating an overhead crane equipped with a human detection system of the present invention has a simple equipment configuration and can stop the crane to prevent a worker from being crushed under a suspended load if the camera's field of view is blocked by luggage when working near a suspended load when hoisting the object, or if the worker goes under the suspended load and cannot be seen by the camera, making it suitable for use in overhead cranes and can also be used, for example, in construction machinery that uses cranes.

01 Main winding motor 02 Main winding inverter 03 Auxiliary winding motor 04 Auxiliary winding inverter 05 Traverse motor 06 Traverse inverter 07 Travel motor 08 Travel inverter 09 PLC control panel 10 Person detection status indicator 11 Person detection number indicator 12 Wireless receiver 13 Load cell 14 Upper limit switch 15 Audio loudspeaker 16 Personal computer 17 Touch panel 18 Person detection camera 1
19 Person detection floodlight 1
20 Human detection camera 2
21 Person detection floodlight 2
22 Wireless transmitter (wireless control device)

Claims (3)

A method for operating an overhead crane equipped with a human detection system, comprising the steps of: combining the fields of view and coordinates of multiple cameras into a reference coordinate centered on the hook; tracing an area in which a person can be detected; establishing a warning area in which it is determined that the person will be in danger if the person's position information is suddenly lost; and if, while tracing a person within the warning area, the human detection means determines that a person has not been detected twice consecutively in images taken by all of the cameras at specified time intervals, it is determined that the person has disappeared, and stopping the crane along with an announcement to that effect. The method for operating an overhead crane equipped with a human detection system according to claim 1, characterized in that the control device for operating the overhead crane on the ground is provided with buttons for selecting whether the human detection system is enabled or disabled, for inspecting the human detection system, and for slinging work, and the human detection range is switched and the human detection system is inspected according to the presence or absence of a suspended load. A method for operating an overhead crane equipped with a human detection system according to claim 1 or 2, characterized in that the overhead crane is provided with a human detection status indicator light and a human detection number indicator, and the status of the human detection system is displayed.

Images