Overturn prevention system
The forklift tip-over prevention system addresses the issue of earthquakes by using an earthquake detection unit and lifting control to lower loads safely, ensuring quick evacuation and preventing tipping and load fall.
Patent Information
- Application Number
- JP2024133752
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2026-02-20
AI Technical Summary
Existing forklift tip-over prevention systems do not account for the risk of tipping and load dropping during earthquakes, causing delays in driver evacuation and potential safety hazards.
A forklift tip-over prevention system that includes an earthquake detection unit, a lifting control unit to lower the load to a safe height upon detection of an earthquake, and interference determination units to avoid obstacles, ensuring quick evacuation and preventing tipping and load fall.
Enables rapid driver evacuation during earthquakes by safely lowering the load to prevent forklift tipping and load fall, enhancing safety and efficiency.
Smart Images

Figure 2026030729000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to an anti-tip system for a forklift. [Background technology]
[0002] A forklift is equipped with forks and a lifting device, which raises the forks to lift a pallet placed on the forks and the load placed on the pallet. Loads range from light to heavy, and the center of gravity of the load is not always in the center. Furthermore, the floor on which the forklift runs is not always flat. Therefore, there is a risk that the forklift will tip over when lifting or lowering a load.
[0003] Therefore, a tip-over prevention system has been developed, as disclosed in Patent Document 1, for example. The tip-over prevention system disclosed in Patent Document 1 acquires the load state of a load and changes a threshold value related to the wheel load of each wheel and a threshold value related to the difference in wheel load between the left and right wheels according to the acquired load state. This tip-over prevention system then performs tip-over prevention control to suppress a decrease in the wheel load of each wheel of the vehicle based on the results of comparison of the wheel load of each wheel with the threshold value, the results of comparison of the difference in wheel load between the left and right wheels with the threshold value, and the vehicle's motion state.
[0004] When an earthquake occurs in a facility where forklifts are operated, the shaking can cause objects to fall or shelves to tip over. Therefore, people in the facility need to evacuate quickly to avoid injury from falling objects or toppled shelves. While forklift drivers also need to evacuate, the following problems arise if an earthquake occurs during loading and unloading. For example, if the driver leaves the vehicle with a loaded load and dismounts to evacuate, the earthquake could cause the vehicle to tip over or the loaded load to fall. The tip-over prevention system disclosed in Patent Document 1 does not take into account such tipping and load dropping during an earthquake. Therefore, it is preferable for the driver to lower the load before evacuating. However, lowering the load causes a corresponding delay in the driver's evacuation. [Prior art documents] [Patent documents]
[0005] [Patent Document 1] Japanese Patent Publication No. 2022-070707 Summary of the Invention [Problem to be solved by the invention]
[0006] Therefore, the problem that the present invention aims to solve is to provide a forklift tip-over prevention system that allows the driver to quickly evacuate in the event of an earthquake and also prevents the forklift from tipping over and the load from falling. [Means for solving the problem]
[0007] In order to solve the above problems, the fall prevention system according to the present invention comprises: A tip-over prevention system for use in a forklift truck having a load carrying section and a lifting section that raises and lowers the load carrying section, a vehicle alighting detection unit that detects when the driver alights; and a lifting control unit that lowers the loading unit to a first predetermined height when disembarking is detected.
[0008] The fall prevention system preferably comprises: further comprising an earthquake detection unit that detects the occurrence of an earthquake or a prediction of the occurrence of an earthquake; The lifting control unit lowers the loading unit to a first predetermined height when an earthquake occurrence or a prediction of an earthquake occurrence is detected and disembarking is detected.
[0009] The fall prevention system preferably comprises: a lower camera that captures an image of the underside of the loading section and generates an image of the loading section; an interference determination unit that determines whether or not there is an interfering object that will interfere with the load carrying unit when the load carrying unit is lowered, based on the load carrying unit image; The lifting control section does not lower the load carrying section when the interference determination section determines that an obstruction exists.
[0010] The fall prevention system preferably comprises: a lower camera that captures an image of the underside of the loading section and generates an image of the loading section; an interference determination unit that determines, based on the loading unit image, whether or not there is an interfering object that will interfere when the loading unit is lowered; a retreating unit that retreats the load carrying unit when dismounting is detected and the interference determining unit determines that there is an obstruction, When it is determined that there is no obstruction due to the backward movement of the loading unit, the lifting control unit lowers the loading unit.
[0011] The fall prevention system preferably comprises: The interference determination unit has a trained model that has been trained in advance using training data in which input data are a plurality of images of the loading section when an interfering object is present and a plurality of images of the loading section when no interfering object is present, and output data is the presence or absence of an interfering object.
[0012] The fall prevention system preferably comprises: The loading section image includes at least one of an image of a fork and an image of a pallet with the fork inserted therein.
[0013] The fall prevention system preferably comprises: a pallet presence / absence determination unit that determines whether the state of the load loading unit is a pallet presence state with forks inserted into the pallet, or a pallet absence state with only forks, The interference detection unit a first trained model that has been trained in advance using first training data in which a plurality of loading section images when there is no obstruction and a plurality of loading section images when there is an obstruction in a state without a pallet are used as input data, and the presence or absence of an obstruction is used as output data; a second trained model that has been trained in advance using second teacher data, which uses as input data a plurality of loading section images when there is no interfering object and a plurality of loading section images when there is an interfering object in a state where a pallet is present, and which uses as output data the presence or absence of an interfering object; When the pallet presence / absence determination unit determines that a pallet is not present, the first trained model is used to determine whether an interfering object is present, and when the pallet presence / absence determination unit determines that a pallet is present, the second trained model is used to determine whether an interfering object is present.
[0014] The fall prevention system preferably comprises: A rear camera that takes pictures of the area behind the forklift and generates rear images; a contact determination unit that determines, based on the rear image, whether or not there is an object that the forklift will come into contact with when it is backing up; The reversing unit is configured to reverse the forklift to reversing the load carrying unit, but will not reverse the forklift if the contact determination unit determines that there is an object to be contacted.
[0015] The fall prevention system preferably comprises: Further provided is a height detection unit that detects the height of the load carrying unit, The lifting control section does not lower the load carrying section when the detected height of the load carrying section is less than a second predetermined height.
[0016] The fall prevention system preferably comprises: Further provided is a getting-off notification unit that notifies the driver that the vehicle will be getting off, The getting-off detection unit is configured to detect that the driver has gotten off the vehicle when notified that the driver is getting off the vehicle.
[0017] The fall prevention system preferably comprises: The earthquake detection unit is further capable of detecting or predicting the magnitude of an earthquake, The lifting control unit lowers the loading unit to a first predetermined height when an earthquake of a predetermined magnitude or greater is detected or a prediction of an earthquake of a predetermined magnitude or greater is detected and disembarking is detected.
[0018] In order to solve the above problem, the fall prevention program according to the present invention includes: 1. A tip-over prevention program for use in a forklift having a loading unit, a lifting unit that lifts and lowers the loading unit, a dismounting detection unit that detects when a driver dismounts from the forklift, and a computer, On the computer, When disembarking is detected, the lifting unit is controlled to lower the loading unit to a first predetermined height. [Effects of the Invention]
[0019] The tip-over prevention system according to the present invention allows the operator to quickly evacuate in the event of an earthquake, and also prevents the forklift from tipping over and the load from falling. [Brief explanation of the drawings]
[0020] [Figure 1] 1 is a schematic side view showing an anti-tip system for a counter type forklift according to an embodiment of the present invention. [Figure 2] 1 is a schematic side view showing an anti-tip system for a reach forklift according to an embodiment of the present invention. [Figure 3] FIG. 1 is a block diagram of a fall prevention system. [Figure 4] FIG. 10 is a diagram illustrating the operation of an interference determination unit. [Figure 5] 1A and 1B are side views showing the fork, where A shows a state where there is no obstruction and B shows a state where there is an obstruction. [Figure 6] 1A and 1B are bottom views showing the fork, where A shows a state where there is no obstruction and B shows a state where there is an obstruction. [Figure 7] 1A and 1B are side views showing a pallet with a fork inserted, where A shows a state where there is no obstruction and B shows a state where there is an obstruction. [Figure 8] 1A and 1B are bottom views showing a pallet with forks inserted, where A shows a state where there is no obstruction and B shows a state where there is an obstruction. [Figure 9] FIG. 1 is a flow diagram of a fall prevention system. DETAILED DESCRIPTION OF THE INVENTION
[0021] An embodiment of the fall prevention system of the present invention will be described below with reference to the accompanying drawings, in which the X axis indicates the front-to-rear direction, the Y axis indicates the left-to-right direction, and the Z axis indicates the up-down direction.
[0022] <Forklift> Fig. 1 is a schematic side view showing an anti-toppling system S for a counter-load forklift F according to this embodiment, and Fig. 2 is a schematic side view showing an anti-toppling system S for a reach forklift F according to this embodiment. The forklifts F in Figs. 1 and 2 are merely examples of this embodiment, and the forklift according to the present invention is not limited to these.
[0023] 1 and 2, a forklift F includes front and rear wheels 2, a vehicle body 3, a driver's seat 4, a head guard 5, left and right masts 6, left and right forks 7, a lower camera 8, a rear camera 9, a drive unit 10, a lifting unit 11, a driver's seat camera 13, and a speaker 14. The forks 7 and the pallet P into which the forks 7 are inserted correspond to the "load carrying portion" of the present invention.
[0024] A vehicle body 3 is disposed on front and rear wheels 2, and a driver's seat 4 is disposed on the vehicle body 3. A head guard 5 extends horizontally and is disposed above the driver's seat 4. Left and right masts 6 are configured to be vertically extendable and retractable and are disposed in front of the vehicle body 3. Left and right forks 7 are connected to the masts 6 and configured to be vertically movable along the masts 6.
[0025] The lower camera 8 is composed of two cameras, each of which is disposed on the outer side of the left and right masts 6. The number of cameras in the lower camera 8 is merely an example, and the lower camera 8 may be composed of one camera or three or more cameras. The location of the lower camera 8 is also merely an example, and is not particularly limited as long as it is possible to capture an image of the underside of the forks 7. For example, the lower camera 8 may be disposed on the vehicle body 3 or on the reach leg of a reach forklift F. The lower camera 8 captures an image of the underside of the load carrying section (forks 7, pallet P) and generates a loading section image LG.
[0026] Note that there are times when the height of the forks 7 is low and the underside of the load carrying portion is not photographed by the lower camera 8, but as will be described later, when the height of the load carrying portion is low, photographing by the lower camera 8 is not necessary, so there is no problem even if the underside of the load carrying portion is not photographed by the lower camera 8. However, to avoid a situation where the underside of the load carrying portion is not photographed because the height of the forks 7 is high, the lower camera 8 may be configured, for example, to be able to change the photographing angle according to the height of the forks 7 or to be able to photograph at a wide angle. Furthermore, the lower camera 8 may be provided, for example, on an extension portion (e.g., the inner mast) rather than on a fixed portion of the mast 6, so that the position of the forks 7 can be tracked according to their height.
[0027] The rear camera 9 is disposed on the head guard 5. The rear camera 9 photographs the area behind the vehicle body 3 and generates a rear image. However, the location of the rear camera 9 is merely an example, and is not particularly limited as long as it can photograph an object in contact with the rear, which will be described later.
[0028] The drive unit 10 is made up of drive parts including a motor, and is connected to either the front or rear wheels 2. The drive unit 10 drives the connected wheels 2 to move the forklift F forward and backward. When the present invention is applied to an engine-powered forklift, the drive unit 10 is made up of drive parts including an engine.
[0029] The lifting section 11 has a tank that stores hydraulic oil, a lift cylinder that expands and contracts with the hydraulic pressure of the hydraulic oil, and a control valve, and extends and contracts the mast 6, thereby raising and lowering the fork 7.
[0030] The driver's seat camera 13 is disposed on the head guard 5. The driver's seat camera 13 photographs the driver's seat 4 and generates a driver's seat image. However, the location of the driver's seat camera 13 is merely an example and is not limited to this.
[0031] The speaker 14 is disposed on the head guard 5. The speaker 14 outputs an alarm sound for notifying the user of an earthquake or a descent, which will be described later. The alarm sound may be a voice that notifies the user of an earthquake or a descent, or may be a specific sound such as a beep.
[0032] 3 is a block diagram of the tip-over prevention system S. As shown in FIG. 3, the forklift F further includes a communication unit 15, a hydraulic sensor 18, and a control unit 20.
[0033] The communication unit 15 is disposed on the vehicle body 3 (not shown) and is configured to be able to receive emergency earthquake alerts transmitted from the Japan Meteorological Agency. More specifically, the communication unit 15 receives detailed earthquake alerts that include ground information for each location transmitted by a licensed forecasting business operator. The received earthquake alerts are transmitted to the earthquake detection unit 21.
[0034] The hydraulic pressure sensor 18 detects the hydraulic pressure between the control valve and the lift cylinder of the lifting unit 11. The hydraulic pressure sensor 18 is configured to be able to detect the difference in hydraulic pressure between when the fork 7 has a pallet P attached and when the fork 7 does not have a pallet P attached.
[0035] The control unit 20 is configured by a computer arranged in the vehicle body 3, and has an arithmetic unit, a storage device, and a memory. The storage device stores a tip-over prevention program that causes the computer to function as a lift control unit 35, which will be described later.
[0036] <Functional configuration> As shown in Figure 3, the control unit 20 has an earthquake detection unit 21, a disembarkation detection unit 22, a height detection unit 24, a pallet presence / absence determination unit 26, an interference determination unit 28, a contact determination unit 32, a retreat unit 34, a lift control unit 35, and a notification unit 36.
[0037] The earthquake detection unit 21 receives an earthquake early warning and thereby detects an "earthquake prediction." The earthquake prediction includes an estimated value of the size of the earthquake (either or both of the seismic intensity and magnitude).
[0038] The dismounting detection unit 22 refers to the driver's seat image, and determines that the driver H has dismounted when the driver H is not included in the driver's seat image, and determines that the driver H has not dismounted when the driver H is included in the driver's seat image. In this way, the dismounting detection unit 22 detects that the driver H has dismounted. Note that the method of detecting that the driver H has dismounted by the dismounting detection unit 22 is not limited to this. For example, the forklift F may further include a sensor provided in the driver's seat 4, and the dismounting detection unit 22 may detect that the driver H has dismounted by using this sensor. The type of this sensor is not particularly limited.
[0039] The height detection unit 24 detects the height of the forks 7. The height detection method of the height detection unit 24 is not particularly limited. The forklift F further includes a sensor (not shown) for detecting the height of the forks 7, and the height detection unit 24 uses this sensor to detect the height of the forks 7. The sensor for detecting the height of the forks 7 may be, for example, a distance sensor, a position sensor, or an optical encoder. In the case of a distance sensor, it may be an ultrasonic or laser type, and detects the height of the forks 7 by emitting ultrasonic waves or emitting a laser to detect the distance from the underside of the forks 7 to the ground. In the case of a position sensor, it is disposed in a lift cylinder of the lifting unit 11 and detects the extension / retraction state of the lift cylinder to detect the height of the forks 7. In the case of an optical encoder, it is attached to a shaft (not shown) that rotates as the forks 7 are raised and lowered, and detects the height of the forks 7 by counting the number of rotations of the shaft.
[0040] The pallet presence / absence determining unit 26 determines whether the load carrying unit is in a pallet-present state (attached state) with the forks 7 inserted into the pallet P, or in a pallet-less state (unattached state) with only the forks 7. In this embodiment, the pallet presence / absence determining unit 26 determines whether the load carrying unit is in a pallet-present state based on the value detected by the hydraulic sensor 18.
[0041] 4, the interference determination unit 28 has a first trained model 280 and a second trained model 281. The interference determination unit 28 determines whether or not there is an interfering object I in order to prevent the fork 7 from interfering with the interfering object I when the fork 7 is lowered. In other words, the interference determination unit 28 determines whether or not the fork 7 can be lowered.
[0042] The first trained model 280 is trained in advance using first training data in which input data are a plurality of loading section images LG when there is no interfering object I and a plurality of loading section images LG when there is an interfering object I in a pallet-free state, and output data is the presence or absence of the interfering object I. This training method may be a deep neural network, and is not particularly limited.
[0043] FIG. 5 is a side view of the fork 7, with FIG. 5A showing a state where there is no interfering object I and FIG. 5B showing a state where there is an interfering object I. FIG. 6 is a bottom view of the fork 7, with FIG. 6A showing a state where there is no interfering object I and FIG. 6B showing a state where there is an interfering object I. As shown in each figure, if there is an interfering object I below the fork 7, a phenomenon occurs in which part of the underside of the fork 7 is missing or the outline of the fork 7 is missing in the load part image LG. By having previously undergone the above-described learning, the first trained model 280 outputs the presence or absence of an interfering object I when the load part image LG is input. For example, the first trained model 280 may be trained to output a "no interfering object I" only when the underside of the fork 7 is completely visible in the load part image LG.
[0044] The second trained model 281 is trained in advance using second training data in which input data are a plurality of loading section images LG when there is no interfering object I and a plurality of loading section images LG when there is an interfering object I in a pallet-present state, and output data is the presence or absence of the interfering object I. The training method for the second trained model 281 may also be a deep neural network and is not particularly limited.
[0045] FIG. 7 is a side view showing a pallet P with a fork 7 inserted, with FIG. 7A showing a state where there is no interfering object I and FIG. 7B showing a state where there is an interfering object I. FIG. 8 is a bottom view showing a pallet P with a fork 7 inserted, with FIG. 8A showing a state where there is no interfering object I and FIG. 8B showing a state where there is an interfering object I. As shown in each figure, if there is an interfering object I below the pallet P, part of the underside of the pallet P will not be visible in the loading section image LG, or the outline of the pallet P will be missing. By having previously undergone the above-described learning, the second trained model 281 outputs the presence or absence of an interfering object I when the loading section image LG is input. For example, the second trained model 281 may be trained to output "no interfering object I" only when the underside of the pallet P is fully visible in the loading section image LG.
[0046] 4, the interference determination unit 28 determines whether to use the first trained model 280 or the second trained model 281 based on the result of the pallet P presence / absence determination by the pallet presence / absence determination unit 26. When a pallet is absent, the interference determination unit 28 inputs the loading section image LG to the first trained model 280 and determines the presence or absence of an interfering object I based on the output result of the first trained model 280. When a pallet is present, the interference determination unit 28 inputs the loading section image LG to the second trained model 281 and determines the presence or absence of an interfering object I based on the output result of the second trained model 281.
[0047] The contact determination unit 32 determines, based on the rear image, whether or not there is an object that the forklift F will come into contact with when it backs up. The object may be, for example, a person, a wall, or an obstacle. The contact determination unit 32 may have a third trained model. The third trained model may be configured, for example, to be trained in advance using training data in which a plurality of rear images are used as input data and a likelihood score of the presence of an object in contact is used as output data, and to output the likelihood score of the presence of an object in contact when a rear image is input. The contact determination unit 32 determines whether or not there is an object in contact based on the likelihood score. Note that the determination method of the contact determination unit 32 is not particularly limited. For example, the forklift F may be equipped with a sensor instead of the rear camera 9, and the contact determination unit 32 may determine whether or not there is an object in contact based on detection by the sensor.
[0048] When it is detected that the driver H has dismounted and the interference determination unit 28 determines that there is an obstructing object I, the reverse unit 34 controls the drive unit 10 to cause the forklift F to move backward. However, the reverse unit 34 is configured not to cause the forklift F to move backward if the contact determination unit 32 determines that there is an object to be contacted. In this way, the reverse unit 34 can prevent the forklift F from coming into contact with the object to be contacted when the forklift F is moved backward. Note that "moving the load carrying unit backward" in the present invention also includes the forks 7 (load carrying unit) being moved backward in response to the forklift F being moved backward.
[0049] The lifting control unit 35 is configured to control the lifting unit 11 to lower the forks 7 to a first predetermined height when a predetermined condition is satisfied. In the present invention, the "first predetermined height" refers to a height of the forks 7 that can prevent the forklift F from tipping over even in the event of an earthquake. In this embodiment, the first predetermined height is set to a height at which the forks 7 contact the ground to more reliably prevent the forklift F from tipping over and the load W from falling. However, this is merely an example, and the first predetermined height of the present invention is not limited to this. For example, the first predetermined height may be 1 m above the ground, or at the eye level of the driver H.
[0050] The predetermined conditions for the lift control unit 35 in this embodiment mean that all of the following conditions are satisfied. -Predictions of earthquake occurrence have been detected Driver H was detected getting off the vehicle. The height of the fork 7 when the predicted occurrence of an earthquake is detected is equal to or greater than the second predetermined height. The interference determination unit 28 determines that there is no interfering object I, or determines that there is an interfering object I and then determines that there is no interfering object I due to the fork 7 moving backward.
[0051] The second predetermined height is the height of the forks 7 that can prevent the forklift F from tipping over even in the event of an earthquake. The second predetermined height may be, for example, 1 m from the ground, or the height of the driver H's eye level.
[0052] When the forklift F is a reach forklift F, the lifting control unit 35 performs a so-called reach-in operation, that is, an operation of retracting the forks 7 and the mast 6 along the reach leg, and then lowers the forks 7.
[0053] When the earthquake detection unit 21 detects a predicted earthquake occurrence, the notification unit 36 causes the speaker 14 to output an alarm sound indicating the occurrence of an earthquake. In addition, before the elevation control unit 35 starts lowering the forks 7, the notification unit 36 causes the speaker 14 to output an alarm sound indicating the descent of the forks 7.
[0054] <Operation of the fall prevention system> Next, a series of operations of the fall prevention system S will be described with reference to FIG.
[0055] (1) When the fall prevention system S detects a predicted earthquake occurrence using the earthquake detection unit 21 (Yes in S (step) 1 in FIG. 9), it notifies the speaker 14 of the predicted earthquake occurrence and detects whether the driver H has dismounted using the dismount detection unit 22. When the fall prevention system S does not detect that the driver H has dismounted (No in S2 in FIG. 9), it does not proceed to the next step.
[0056] (2) Next, when it is detected that the driver H has dismounted (Yes in S2 of FIG. 9), the fall prevention system S detects whether the height of the forks 7 is equal to or greater than a second predetermined height.
[0057] (3) Then, if the height of the fork 7 is less than the second predetermined height (No in S3 in Figure 9), the fork 7 will not tip over, and the anti-tip system S completes the operation without lowering the fork 7 (END in Figure 9).
[0058] (4) If the height of the fork 7 is equal to or greater than the second predetermined height (Yes in S3 of FIG. 9), the anti-tip stem determines whether or not there is an obstruction I below the fork 7 using the interference determination unit 28.
[0059] (5) Next, when the interference determination unit 28 determines that there is no interfering object I (Yes in S4 of FIG. 9), the anti-tip system S notifies the speaker 14 to lower the forks 7, and then causes the lift control unit 35 to lower the forks 7 to the first height (S5 of FIG. 9). This prevents the forklift F from tipping over.
[0060] (6) Furthermore, when the interference determination unit 28 determines that there is an interfering object I (No in S4 of FIG. 9), the tip-over prevention system S determines by the contact determination unit 32 whether there is an object to be contacted behind the forklift F.
[0061] (7) When the contact determination unit 32 determines that there is no object of contact behind the forklift F (Yes in S6 of FIG. 9), the anti-tip system S causes the forklift F to move backward by the reverse unit 34 (S7 of FIG. 9). Next, when the forklift F moves backward and there is no longer any obstruction I below the forks 7 (Yes in S4 of FIG. 9), the anti-tip system S lowers the forks 7 (S5 of FIG. 9).
[0062] (8) On the other hand, if the contact determination unit 32 determines that there is a contact object behind the forklift F (No in S6 of FIG. 9), the tip-over prevention system S waits until the contact object disappears. As a result, for example, if there is a person behind the forklift F, the tip-over prevention system S determines that there is no contact object after the person leaves the rear of the forklift F (Yes in S6 of FIG. 9), and then moves the forklift F backward (S7 of FIG. 9).
[0063] As a result, the tip-over prevention system S can automatically lower the forks 7 when an earthquake occurs or when the driver H gets off the vehicle without lowering the forks 7. Therefore, the tip-over prevention system S allows the driver H to evacuate quickly in the event of an earthquake, and can prevent the forklift F from tipping over or the load W from falling. Furthermore, when the tip-over prevention system S detects an earthquake early warning, it can request the driver H to evacuate (get off the vehicle) quickly by notifying the driver of the occurrence of an earthquake through the speaker 14.
[0064] Furthermore, when there is an obstruction I below the forks 7, the tip-over prevention system S moves the forklift F backward before lowering the forks 7, thereby appropriately preventing the forks 7 from interfering with the obstruction I. In addition, when there is an object to be contacted behind the forklift F, the tip-over prevention system S moves the forklift F backward after the object to be contacted has passed, thereby appropriately preventing the forklift F from contacting the object.
[0065] While one embodiment of the tip-over prevention system of the present invention has been described above, the present invention is not limited to the above embodiment. For example, the tip-over prevention system of the present invention may be implemented in the following modified examples.
[0066] The load carrying portion is not limited to the fork 7. For example, the load carrying portion may be a ram fork or the like, and the interference determination portion 28 may determine whether or not there is an obstruction I below the ram fork or the like.
[0067] The fall prevention system may further include, for example, an earthquake detection device that detects earthquakes, and the earthquake detection unit 21 may detect earthquakes using this earthquake detection device.
[0068] The lifting control unit 35 may be configured to lower the loading unit only when an earthquake or a prediction of an earthquake of a predetermined magnitude or greater is detected. In this case, the predetermined magnitude of the earthquake may be set to, for example, a seismic intensity of 4 or a magnitude of 5.
[0069] The interference determination unit 28 may have a trained model that has been trained in advance so that it can determine the presence or absence of an interfering object I from only the loading unit image LG, even if there is no information on whether a pallet P is attached to the forks 7. In this case, the tip-over prevention system S does not need to be equipped with the hydraulic sensor 18 and the pallet presence / absence determination unit 26.
[0070] The retreating unit 34 may be configured to control a reach-in operation. In this case, when the interference unit determines that an obstruction I is present, the retreating unit 34 first performs a reach-in operation to retreat the load carrying unit (forks 7) before reversing the forklift F.
[0071] The fall prevention system S may further include a dismounting notification unit that notifies the driver H that he / she is dismounting. In this case, the dismounting detection unit 22 detects that the driver H has dismounted by receiving the notification that he / she is dismounting. The dismounting notification unit may, for example, have a button electrically connected to the control unit 20, and the dismounting detection unit 22 may be configured to detect that the driver H has dismounted when the button is pressed by the driver H. Alternatively, the dismounting notification unit may be configured to include a microphone electrically connected to the control unit 20, and the dismounting detection unit 22 may be configured to detect that the driver H has dismounted by receiving a specific voice uttered by the driver H via the microphone. [Explanation of symbols]
[0072] S Anti-tip system F Forklift H Driver I Interference P Palette LG loading area image W load 2 wheels 3. Body 4 Driver's seat 5 Head Guard 6 Mast 7. Fork 8. Bottom camera 9 Rear camera 10 Drive unit 11 Lifting section 13 Driver's seat camera 14 Speaker 15 Communications Department 18 Oil pressure sensor 20 Control Unit 21 Earthquake detection unit 22 Exit detection unit 24 Height detection unit 26 Pallet presence detection unit 28 Interference detection section 280 First trained model 281 Second trained model 32 Contact determination section 34 Retreat 35 Lift control section 36 Notification Department
Claims
1. A tip-over prevention system for use in a forklift truck having a load carrying section and a lifting section that raises and lowers the load carrying section, a vehicle alighting detection unit that detects when the driver alights; A tip-over prevention system comprising: a lifting control unit that lowers the loading unit to a first predetermined height when disembarking is detected.
2. further comprising an earthquake detection unit that detects the occurrence of an earthquake or a prediction of the occurrence of an earthquake; The fall prevention system according to claim 1 , wherein the lifting / lowering control unit lowers the loading unit to the first predetermined height when an earthquake has occurred or a prediction of an earthquake has been detected and disembarking has been detected.
3. a lower camera that captures an image of the underside of the loading section and generates a loading section image; an interference determination unit that determines, based on the loading unit image, whether or not there is an interfering object that will interfere with the loading unit when the loading unit is lowered; The tip-over prevention system according to claim 1 , wherein the lifting / lowering control unit does not lower the loading unit when the interference determination unit determines that there is an obstruction.
4. a lower camera that captures an image of the underside of the loading section and generates a loading section image; an interference determination unit that determines, based on the loading unit image, whether or not there is an interfering object that will interfere with the loading unit when the loading unit is lowered; a retreating unit that retreats the load carrying unit when dismounting is detected and the interference determination unit determines that there is an obstruction, The tip-over prevention system according to claim 1 , wherein the lifting control unit lowers the loading unit when it is determined that the obstruction is no longer present due to the backward movement of the loading unit.
5. 4. The anti-tip system of claim 3, wherein the interference determination unit has a trained model that has been trained in advance using training data that uses multiple loading section images when the interfering object is present and multiple loading section images when the interfering object is not present as input data, and that outputs the presence or absence of the interfering object.
6. The tip-over prevention system according to claim 5 , wherein the loading section image includes at least one of an image of a fork and an image of a pallet into which the fork is inserted.
7. a pallet presence / absence determination unit that determines whether the state of the load loading unit is a pallet presence state with forks inserted into the pallet, or a pallet absence state with only the forks, The interference determination unit a first trained model that has been trained in advance using first teacher data, in which input data are a plurality of loading section images when the interfering object is not present and a plurality of loading section images when the interfering object is present in the pallet-absent state, and output data is the presence or absence of the interfering object; a second trained model that has been trained in advance using second teacher data, the second teacher data using as input data a plurality of loading section images when the interfering object is not present and a plurality of loading section images when the interfering object is present in the pallet-present state, and the second trained model using as output data the presence or absence of the interfering object, A tip-over prevention system as described in claim 3, wherein when the pallet presence / absence determination unit determines that the pallet is absent, the first trained model is used to determine whether or not the interfering object is present, and when the pallet presence / absence determination unit determines that the pallet is present, the second trained model is used to determine whether or not the interfering object is present.
8. a rear camera that captures an image of the rear of the forklift and generates a rear image; a contact determination unit that determines, based on the rear image, whether or not there is an object that the forklift will come into contact with when it moves backward; The tip-over prevention system of claim 4, wherein the reversing unit is configured to reversing the forklift to reversing the loading unit, but does not reversing the forklift if the contact determination unit determines that there is a contact object.
9. Further provided is a height detection unit that detects the height of the load carrying unit, The tip-over prevention system according to claim 1 , wherein the lifting control unit does not lower the loading unit when the detected height of the loading unit is less than a second predetermined height.
10. Further provided is a getting-off notification unit that notifies the driver that the vehicle will be getting off, The fall prevention system according to claim 1 , wherein the dismounting detection unit is configured to detect that the driver has dismounted from the vehicle when a notification that the driver is dismounting is received.
11. The earthquake detection unit is further capable of detecting or predicting the magnitude of an earthquake, The anti-tip system of claim 2, wherein the lifting control unit lowers the loading unit to the first predetermined height when a seismic intensity of a predetermined level or higher is detected or a prediction of an earthquake magnitude of a predetermined level or higher is detected and disembarking is detected.
12. 1. A tip-over prevention program for use in a forklift truck having a loading unit, a lifting unit that lifts and lowers the loading unit, a dismounting detection unit that detects when a driver dismounts from the forklift, and a computer, The computer, When dismounting is detected, the tip-over prevention program controls the lifting unit to lower the loading unit to a first predetermined height.
Citation Information
Patent Citations
Vehicle control device and vehicle control program
JP2022070707A