Safety system

Abstract

This helps prevent delays in cargo handling operations while also preventing accidents caused by distracted driving. [Solution] The distraction monitoring system S comprises a shooting unit 12, an insertion / removal detection unit 22, a gaze estimation unit 24, a distraction determination unit 28, and a speed change unit 30. The shooting unit 12 photographs the forks of the forklift to generate a fork image and also photographs the driver to generate a head image. The insertion / removal detection unit 22 detects the insertion and removal of the forks from the pallet based on the fork image, and the gaze estimation unit 24 estimates the direction of the driver's gaze based on the head image. The distraction determination unit 28 determines whether the driver is distracted while inserting or removing the forks, based on the estimated direction of the driver's gaze during the insertion and removal of the forks. If the distraction determination unit 28 determines that the driver is distracted, the speed change unit 30 changes the operating speed of the forklift.

Description

Technical Field

[0001] The present invention relates to a side vision monitoring system for a forklift, and particularly to a side vision monitoring system related to side vision during fork insertion and removal.

Background Art

[0002] In the operation by a forklift, due to various reasons, the load may fall. The causes of the load falling include, for example, the driver's side vision.

[0003] As a system for preventing the side vision of a vehicle driver, for example, Patent Document 1 discloses a side vision detection device. This device includes a face orientation detection means, a first determination means, a line-of-sight direction detection means, and a second determination means. This device detects the orientation of a person (driver)'s face by the face orientation detection means, and detects the orientation of the person's line of sight by the line-of-sight direction detection means. And first, this device determines whether a person is having side vision based on the orientation of the person's face by the first determination means. When it is determined that the person is not having side vision, it determines whether the person is having side vision based on the orientation of the person's line of sight by the second determination means. Thereby, this device improves the accuracy of side vision detection. Also, this device determines whether a person is having side vision according to the speed of the vehicle being driven. Further, this device can also stop the engine of its own vehicle when side vision is detected.

[0004] By the way, in the operation of a forklift, there is an operation of inserting and removing the fork into and from a pallet. And side vision during fork insertion and removal is directly related to the load falling during fork insertion and removal, subsequent traveling operations, and load falling during load lifting and lowering, so it is very dangerous. However, the side vision detection device disclosed in Patent Document 1 does not particularly consider side vision during fork insertion and removal. Also, if the operation of the forklift is immediately stopped upon detecting side vision, there is a possibility that the cargo handling work will be delayed.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2009-15549 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Therefore, the problem that the present invention aims to solve is to provide a distraction monitoring system that can prevent accidents caused by distractions while avoiding delays in cargo handling operations. [Means for solving the problem]

[0007] To solve the above problems, the distraction monitoring system according to the present invention A forklift fork photography unit that photographs the forks and generates fork images, A driver imaging unit that photographs the driver and generates a head image, An insertion / removal detection unit that detects the insertion and removal of forks from a pallet based on a fork image, A gaze estimation unit that estimates the direction of the driver's gaze based on a head image, A distraction detection unit determines whether the driver is distracted while inserting or removing the forks, based on the estimated direction of the driver's gaze during the insertion or removal of the forks. The system includes a speed change unit that changes the operating speed of the forklift when the distraction detection unit determines that the driver is distracted.

[0008] The above-mentioned distraction monitoring system is preferably, The distraction detection unit determines that the driver is not distracted if, while inserting or removing the forks, the driver's line of sight is directed towards the forks or the load they are carrying.

[0009] The above-mentioned distraction monitoring system is preferably, The distraction detection unit determines that the driver is not distracted if the direction of the driver's gaze is directed forward and that the direction of the gaze is within a predetermined angle from the center of the driver's seat in the horizontal direction, and that the driver is looking at both or either the forklift and the load loaded on it.

[0010] The above-mentioned distraction monitoring system is preferably, The operating speed of the forklift changed by the speed change unit includes either or both of the forklift's travel speed and the fork lifting speed.

[0011] The above-mentioned distraction monitoring system is preferably, A rotation detection unit that detects the rotation operation of a forklift, The system further includes a warning unit that alerts the driver if a turning operation is detected while the forks are being inserted or removed.

[0012] The above-mentioned distraction monitoring system is preferably, It further includes a rotation detection unit that detects the rotation operation of the forklift, The speed control unit, when it detects a turning operation while the forks are being inserted or removed, will reduce the forklift's maximum travel speed, even if it is determined that the driver is not distracted.

[0013] The above-mentioned distraction monitoring system is preferably, The speed control unit stops the raising and lowering of the forks if it determines that the driver is distracted while the forks are being inserted or removed, and the height of the forks is above a predetermined height.

[0014] The above-mentioned distraction monitoring system is preferably, After the forks have been inserted or removed, the speed change unit will revert the forklift's operating speed back to its original setting once it confirms that the forks have been inserted or removed properly.

[0015] The above-mentioned distraction monitoring system is preferably, If it is determined that the driver is looking away during the insertion and removal of the fork and the speed of the forklift is changed, and then it is confirmed that the fork has been properly inserted and removed, the operating speed of the forklift after the change is restored even during the insertion and removal of the fork.

[0016] The above-mentioned looking-away monitoring system preferably further includes a load presence / absence detection unit that detects whether or not the fork is loaded with a load based on the fork image. If it is detected that the fork is not loaded with a load after the insertion and removal of the fork, the speed change unit restores the operating speed of the forklift after the change.

[0017] <00000s84>The above-mentioned looking-away monitoring system preferably The looking-away determination unit determines that the driver is looking away only when the driver is looking away for a time longer than the looking-away allowable time during the insertion and removal of the fork.

[0018] The above-mentioned looking-away monitoring system preferably further includes a notification unit that notifies of looking away at a predetermined location when it is determined that the driver is looking away.

[0019] To solve the above problems, the forklift according to the present invention includes the looking-away monitoring system described in any of the above.

[0020] To solve the above problems, the looking-away monitoring program according to the present invention is a looking-away monitoring program used in a looking-away monitoring system having a driver imaging unit that images a driver to generate a head image, a fork imaging unit that images a fork of a forklift to generate a fork image, and a computer, and causes the computer to a insertion / removal detection unit that determines whether or not the fork is being inserted or removed based on the fork image, a line-of-sight estimation unit that estimates the direction of the driver's line of sight based on the head image, A distraction detection unit determines whether the driver is distracted while inserting or removing the forks, based on the estimated direction of the driver's gaze during the insertion or removal of the forks. If it is determined that the driver is distracted while inserting or removing the forks, the speed change unit will be activated to change the operating speed of the forklift. [Effects of the Invention]

[0021] The distraction monitoring system according to the present invention can prevent accidents caused by distractions while avoiding delays in cargo handling operations. [Brief explanation of the drawing]

[0022] [Figure 1] This is a schematic side view of the forklift according to the present invention. [Figure 2] Figure 1 is a schematic top view of the forklift shown. [Figure 3] This is a block diagram of the distraction monitoring system according to the present invention. [Figure 4] This is a top view showing where the forks are inserted into the pallet. A shows the forks positioned in front of the pallet, B shows the forks being inserted into the pallet, and C shows the forks fully inserted into the pallet. [Figure 5] This diagram shows the flow of the distraction monitoring system. [Figure 6] This is a top view showing the driver's seat of a forklift to illustrate a modified version of the distraction monitoring system. [Modes for carrying out the invention]

[0023] Hereinafter, with reference to the attached figures, a distraction monitoring system according to one embodiment of the present invention and a forklift equipped with said distraction monitoring system will be described. In the figures, the X axis indicates the front-to-back direction, the Y axis indicates the left-to-right direction, and the Z axis indicates the up-and-down direction. Forklift 1 in this embodiment is a battery-powered counterbalanced forklift, but this is merely an example, and the forklift in the present invention is not limited to this.

[0024] <Forklift Configuration> Figure 1 is a side view showing a forklift 1 equipped with the distraction monitoring system S of this embodiment, and Figure 2 is a top view showing the forklift 1 of this embodiment. Note that in Figure 2, the head guard 8, which will be explained later, is shown as transparent. Figure 3 is a block diagram of the distraction monitoring system S. As shown in Figures 1 and 2, the forklift 1 comprises a plurality of wheels 2, a body 3, left and right masts 4, left and right forks 5, a backrest 6, a plurality of pillars 7, a head guard 8, a driver's seat 9, a drive unit 11, a camera unit 12, and a speaker 14.

[0025] Multiple wheels (four in this embodiment) 2 are arranged front, rear, left, and right, and the vehicle body 3 is positioned on top of the wheels 2.

[0026] The left and right masts 4 are positioned at the front of the vehicle body 3 and are configured to extend and retract vertically. The left and right forks 5 and backrest 6 are connected to the masts 4 and are configured to move up and down along the masts 4. The forks 5 are also configured to move back and forth in conjunction with the forward and reverse movement of the vehicle body 3.

[0027] Multiple (four) pillars 7 extend upward from the vehicle body 3, and the head guard 8 is positioned above the vehicle body 3 and supported by the pillars 7.

[0028] The driver's seat 9 is located above the vehicle body 3 and below the head guard 8, with the control panel located at its front. The control panel includes a steering wheel, several levers, several pedals, and so on. The several levers consist of, for example, a parking brake lever, a lift lever, a tilt lever, and a forward / reverse lever. The several pedals consist of, for example, an accelerator pedal, a clutch pedal, and a brake pedal.

[0029] The drive unit 11 is composed of drive components such as a motor and is connected to either the front or rear wheel 2. The drive unit 11 drives the connected wheel 2 to move the forklift 1 forward or backward. When the present invention is applied to an engine-powered forklift, the drive unit 11 is composed of drive components including an engine.

[0030] In this embodiment, the imaging unit 12 is composed of a single camera and serves as both the forklift imaging unit and the driver imaging unit in this invention. However, this is merely an example, and the forklift imaging unit and the driver imaging unit in this invention may be configured separately. Furthermore, the forklift imaging unit and the driver imaging unit in this invention may each be composed of multiple cameras.

[0031] The imaging unit 12 is positioned on the right front pillar 7 and is configured to capture the driver H's head. In this embodiment, the imaging unit 12 is configured to capture the eyes of the seated driver H. However, the location of the imaging unit 12 is merely an example, and the location of the imaging unit 12 is not limited to this. The imaging unit 12 captures the driver H's head and generates a head image.

[0032] Furthermore, the camera unit 12 photographs the fork 5 and generates a fork image. In this embodiment, the camera unit 12 has an ultra-wide-angle lens or a fisheye lens, and is configured to photograph the driver H's head and the fork 5. In this embodiment, the camera unit 12 does not move mechanically, so it cannot photograph the fork 5 if it has moved higher than the driver H's head. However, as described above, the camera unit 12 can be composed of two or more cameras, or a camera that moves up and down mechanically, so that the fork 5 can also be photographed if it has moved high.

[0033] Speaker 14 is located on the head guard 8 and notifies the driver H with a notification sound when distraction is detected. Also, as will be explained later, if a turning operation is detected while the fork 5 is in operation, speaker 14 warns the driver H with a warning sound that a turning operation has been detected. The location of speaker 14 is merely an example and is not limited thereto. The notification sound and warning sound may be a specific voice or a beep.

[0034] As shown in Figure 3, the forklift 1 further comprises a lifting unit 16, a rotation detection unit 18, and a control unit 20.

[0035] The lifting section 16 has a lift cylinder that extends and retracts by hydraulic pressure from the hydraulic fluid. The lifting section 16 is connected to the mast 4, and the extension and retraction of the lift cylinder causes the mast 4 to extend and retract. The fork 5 moves up and down along the mast 4 as the mast 4 extends and retracts.

[0036] The turning detection unit 18 detects the operation of the driver H by the control unit and detects that a turning operation has been performed. The turning operation detection method of the turning detection unit 18 is not particularly limited. For example, the turning detection unit 18 may detect that a turning operation has been performed based on the image generated by the imaging unit 12.

[0037] The control unit 20 is comprised of a computer located inside the vehicle body 3. The control unit 20 includes an arithmetic unit, a storage device, and memory. The storage device stores a distraction monitoring program that causes the computer to execute as an insertion / removal detection unit 22, a gaze prediction unit 24, a distraction detection unit 28, a speed change unit 30, a warning unit 32, and a notification unit 34, which will be described later.

[0038] <Functional Configuration of the Distraction Monitoring System> Next, the functional configuration of the distraction monitoring system S will be described. As shown in Figure 3, the control unit 20 includes an insertion / removal detection unit 22, a gaze estimation unit 24, a distraction determination unit 28, a speed change unit 30, a warning unit 32, and a notification unit 34.

[0039] The insertion / removal detection unit 22 detects the insertion and removal of the fork 5 from the pallet P based on the fork image. Figure 4 is a top view showing the fork 5 being inserted into the pallet P; Figure 4A shows the fork 5 positioned in front of the pallet P, Figure 4B shows the fork 5 being inserted into the pallet P, and Figure 4C shows the fork 5 fully inserted into the pallet P. As shown in Figure 4, when the fork 5 is inserted, the top surface of the fork 5 is hidden by the pallet P, reducing the exposed portion of the fork 5. The insertion / removal detection unit 22 then detects the percentage of this exposed portion from the fork image and, according to the percentage of the exposed portion, detects whether the fork 5 is before insertion, inserted, or fully inserted. However, this is merely an example, and the method by which the insertion / removal detection unit 22 detects the insertion and removal of the fork 5 is not particularly limited.

[0040] The insertion / removal detection unit 22 may have a first trained model that, in order to detect the insertion or removal of the fork 5, is pre-trained using training data that takes multiple fork images as input data and the proportion of the exposed portion of the fork 5 as output data, so that when a fork image is input, the model can output the proportion of the exposed portion of the fork 5. In this case, the first trained model may be trained using fork images for each height of the fork 5. As a result, even if one of the forks 5 is hidden by the mast 4 due to a change in the height of the fork 5, the first trained model can detect the proportion of the exposed portion of the fork 5 based on the balance of images of the mast 4 and the forks 5 (either the left or right fork 5, or both forks 5).

[0041] The gaze estimation unit 24 estimates the direction of the driver H's gaze (hereinafter referred to as "gaze direction L") based on the head image. The gaze estimation unit 24 may have a second trained model that can output the gaze direction L when a head image is input, for example, by pre-training with training data that takes multiple head images as input data and the direction of gaze as output data. In this training, it is preferable to use head images of driver H wearing a helmet. It is also preferable that the head images in this training include images of driver H wearing glasses or a mask.

[0042] Furthermore, in this embodiment, the head image is configured to include the eyes of the driver H, but this is merely an example, and for example, the head image generated by the imaging unit 12 may not include the eyes. In this case, it is preferable to use a head image that does not include the eyes as training data in the training learning process so that the gaze direction L can be estimated based on a head image that does not include the eyes.

[0043] Furthermore, to enable appropriate estimation of the gaze direction L based on a head image that does not include the eyes, the imaging unit 12 may, for example, generate a head image that includes the upper body of the driver H. In this case, the second trained model is trained using the head image including the upper body of the driver H as training data to estimate the gaze direction of the driver H. Note that the training may be performed while generating multiple trained models. These multiple trained models may include, for example, a trained model that has been machine-learned to output skeletal information (information on the orientation and shape of the skeleton) from a head image that includes the upper body, and a trained model that has been machine-learned to output the gaze direction L from the skeletal information of multiple drivers H.

[0044] The second pre-trained model may be pre-trained using, for example, training data that takes the height of the fork 5 and a head image as input data and the gaze direction L as output data. This allows the second pre-trained model to predict an appropriate gaze direction L in response to changes in the height of the fork 5. In this case, the distraction monitoring system S further includes a height detection unit capable of detecting the height of the fork 5.

[0045] The distraction detection unit 28 determines whether or not the driver H is distracted while inserting or removing the forks 5, based on the estimated direction of the driver H's gaze during the insertion or removal of the forks 5. The distraction detection unit 28 determines that the driver H is not distracted when the direction of the driver H's gaze is towards the forks 5 or the load W loaded on the pallet P during the insertion or removal of the forks 5.

[0046] Furthermore, the distraction detection unit 28 determines that the driver is distracted only if they are distracted for a period longer than the permitted distraction time while inserting or removing the fork 5. In other words, the distraction detection unit 28 considers the driver not distracted if the distraction time is less than the permitted distraction time. This is because there is a moment when the driver intentionally avoids looking in the direction of the fork 5 in order to check for surrounding obstacles while inserting or removing the fork 5. The permitted distraction time may be set, for example, between 0 and 2 seconds. Also, this permitted distraction time may be set differently when removing and inserting the fork 5.

[0047] The speed change unit 30 changes the operating speed of the forklift 1 when the distraction detection unit 28 determines that the driver H is distracted. The operating speed of the forklift 1 that the speed change unit 30 changes includes both or either the travel speed of the forklift 1 and the lifting speed of the forks 5. In this embodiment, the speed change unit 30 changes both the maximum travel speed of the forklift 1 and the maximum lifting speed of the forks 5. For example, the speed change unit 30 reduces the maximum travel speed to between 0 and 3 km / h and the maximum lifting speed to between 0 and 150 mm / s. The specific speed at which the speed change unit 30 reduces the speed is not particularly limited. This allows the speed change unit 30 to prevent accidents caused by distraction while preventing delays in cargo handling operations. Furthermore, under predetermined conditions of the forklift 1, the speed change unit 30 may set the maximum travel speed to 0 km / h to effectively stop travel, or set the maximum lifting speed to 0 mm / s to effectively stop both upward and downward movement, or either of them.

[0048] Furthermore, if the speed change unit 30 detects a turning operation while the forks 5 are being inserted or removed, it will at least reduce the travel speed of the forklift 1, even if it is determined that the driver H is not distracted. This is because, even if the driver is not distracted, a turning operation while the forks 5 are being inserted or removed can cause the orientation of the forks 5 to be at an angle to the pallet P (the insertion hole in the pallet P), which may lead to an accident where the load W falls. On the other hand, a turning operation may be intentionally performed while the forks 5 are being inserted or removed in order to fine-tune the position of the pallet P (load W). Therefore, even if a turning operation is detected while the forks 5 are being inserted or removed, the entire operation of the forks 5 should not be stopped. In this embodiment, if a turning operation is detected while the forks 5 are being inserted or removed, the speed change unit 30 will only reduce the travel speed of the forklift 1, but the lifting speed may also be reduced.

[0049] If the warning unit 32 detects a turning operation while the forks 5 are being inserted or removed, it warns the driver H via the speaker 14. This prevents the driver H from turning while the forks 5 are being inserted or removed, and also informs the driver H that the operating speed of the forklift 1 is being reduced.

[0050] If the notification unit 34 determines that the driver is distracted, it notifies the driver H of the distraction via the speaker 14. The content of the notification may differ depending on whether the forks 5 are being removed or inserted. In this way, the notification unit 34 suppresses driver H's distraction and also informs driver H that the operating speed of the forklift 1 should be reduced.

[0051] <Flowchart of the distraction monitoring system> Next, referring to Figure 5, we will explain the operation of the distraction monitoring system S again.

[0052] (1) The distraction monitoring system S captures images of the driver H using the camera unit 12 and generates a head image (see S(step)1 in Figure 5).

[0053] (2) Next, the distraction monitoring system S estimates the direction of the driver H's gaze L based on the head image (see S2 in Figure 5).

[0054] (3) Next, the distraction monitoring system S detects the insertion or removal of the fork 5 using the insertion / removal detection unit 22 (see S3 in Figure 5).

[0055] (4) Next, the distraction monitoring system S determines whether the driver H is distracted while inserting or removing the fork 5 using the distraction determination unit 28 (see S4 in Figure 5).

[0056] (5) Next, if the distraction monitoring system S determines that the driver H is distracted (Yes in S5 of Figure 5), the notification unit 34 notifies the driver H that he is distracted (see S6 in Figure 5) and also slows down or stops the lifting speed and travel speed (see S7 in Figure 5).

[0057] (6) If the distraction monitoring system S determines that driver H is not distracted (No. S5 in Figure 5), but detects a turning operation while the fork 5 is being inserted or removed (Yes in S8 in Figure 5), it will warn driver H and reduce the driving speed (see S9 in Figure 5). On the other hand, if the distraction monitoring system S determines that driver H is not distracted (No. S5 in Figure 5) and does not detect a turning operation while the fork 5 is being inserted or removed (No. S8 in Figure 5), it will do nothing (return to S1 in Figure 5).

[0058] The distraction monitoring system S, through the above configuration and operation, can appropriately detect distraction, avoid delays in cargo handling operations, and prevent accidents caused by distraction. Furthermore, if a turning operation is performed while the forks 5 are being inserted or removed, the distraction monitoring system S will issue a warning and reduce the travel speed, even if the driver is not distracted, thereby suppressing the turning operation while the forks 5 are being inserted or removed, and also suppressing accidents that have already occurred.

[0059] Although one embodiment of the distraction monitoring system according to the present invention has been described above, the present invention is not limited to the above embodiment. For example, the distraction monitoring system and forklift according to the present invention may be implemented by the following modifications, or by appropriately combining the following modifications.

[0060] <Variation> The distraction detection unit 28 may, for example, as shown in Figure 6, determine that the driver H is looking at both or either the fork 5 and the load W, and that the driver is not distracted, if the estimated gaze direction L is directed forward and the gaze direction L is within a predetermined angle θ from the center of the driver's seat 9 in the horizontal direction. In other words, in this case, the distraction detection unit 28 does not include whether the gaze direction L is directed vertically or horizontally in its determination criteria. The predetermined angle θ may be, for example, an angle in the range of 5° to 15° to the left and right from the center of the driver's seat 9 (i.e., an angle in the range of 10° to 30° when the left and right angles are combined).

[0061] The speed change unit 30 may stop the raising and lowering of the forks 5 if it is determined that the driver H is distracted while the forks 5 are being inserted or removed, and the height of the forks 5 is above a predetermined height. This is because if the forks 5 are not properly inserted into or removed from the pallet P, and are at a high position, it could lead to a serious accident.

[0062] The speed change unit 30 determines that the driver H is distracted while the forks 5 are being inserted or removed, and changes the speed of the forklift 1. If it is confirmed that the forks 5 were inserted or removed properly, the unit returns the operating speed of the forklift 1 to its original speed, even while the forks 5 are being inserted or removed. This is because if the forks 5 were inserted or removed properly, it is not necessary to continue reducing the operating speed of the forklift 1.

[0063] Confirmation that the forks 5 were properly inserted and removed may, for example, be made by the driver H visually confirming that the forks 5 were properly inserted and removed, that is, that the orientation of the forks 5 was appropriate relative to the orientation of the pallet P. In this case, the driver H may indicate that the insertion and removal were done properly with a gesture such as pointing, and the distraction monitoring system S may be configured to recognize this gesture from the image of the driver H generated by the camera unit 12.

[0064] Alternatively, the distraction monitoring system S may further include a button to be pressed after confirming that the forks 5 have been properly inserted and removed, and when this button is pressed, it may be configured to restore the decelerated operating speed of the forklift 1 to its original speed.

[0065] Furthermore, as another verification method, the distraction monitoring system S may be configured to confirm that the insertion and removal of the forks 5 was performed appropriately based on the fork images generated by the imaging unit 12. In this case, the control unit 20 may further include a trained model for confirming the insertion and removal of the forks 5. This trained model may be pre-trained using training data that takes fork images as input data and outputs whether or not the insertion and removal was performed appropriately as output data. When a fork image is input, it may be configured to output a score indicating whether or not the insertion and removal was performed appropriately. The speed change unit 30 may then be configured to return the operating speed of the modified forklift 1 to its original speed if this score is equal to or greater than a predetermined score.

[0066] Furthermore, after inserting or removing the forks 5, the speed change unit 30 returns the modified operating speed of the forklift 1 to its original state once it confirms that the insertion or removal of the forks 5 has been performed properly. This is because, as mentioned above, if the insertion or removal of the forks 5 has been performed properly, it is not necessary to continue to reduce the operating speed of the forklift 1.

[0067] The speed change unit 30 may revert the modified operating speed of the forklift 1 to its original setting if it detects that the forks 5 are not loaded with load W after the forks 5 have been inserted or removed. This is because if the forks 5 are not loaded with load W after the forks 5 have been inserted or removed, the risk of subsequent travel and lifting of the forks 5 is low. In this case, the control unit 20 may further include a load presence / absence detection unit that detects whether or not the forks 5 are loaded with load W. The load presence / absence detection unit may, for example, detect whether or not the forks 5 are loaded with load W based on a fork image, or it may detect whether or not the forks 5 are loaded with load W based on a value detected by a load sensor. In this case, the load sensor may be, for example, a pressure sensor (a sensor that measures the pressure of the lift cylinder) that the forklift 1 has. The method by which the load presence / absence detection unit detects the presence or absence of load W is not particularly limited. The distraction monitoring system S may also further include a device that notifies the driver H that the forks 5 are not loaded with load W.

[0068] The notification unit 34 may be configured to notify a designated party, such as the manager of driver H or a system managing driver H, if a turning operation is performed while the fork 5 is being inserted or removed, or if it is determined that driver H is distracted. In this case, the distraction monitoring system S may further include a communication device for making these notifications. This helps the distraction monitoring system S to manage driver H to prevent him from being distracted. [Explanation of symbols]

[0069] S Side-watching monitoring system 1 Forklift H Driver L View direction P Palette W load 2 wheels 3. Vehicle Body 4 Mast 5 Forks 6 Backrest 7 Pillar 8 Headguard 9. Driver's seat 11 Drive unit 12 Photography Department 14 speakers 16 Lifting section 18 Swivel detection unit 20 Control Unit 22 Insertion / removal detection unit 24 Line of sight estimation part 28. Distraction Judgment Section 30 Speed ​​change section 32 Warning part 34 Notification Department

Claims

1. A forklift fork photography unit that photographs the forks and generates fork images, A driver imaging unit that photographs the driver and generates a head image, Based on the fork image, an insertion / removal detection unit detects the insertion and removal of the fork from the pallet, A gaze estimation unit that estimates the direction of the driver's gaze based on the head image, A distraction detection unit determines whether the driver is distracted while inserting or removing the forks, based on the estimated direction of the driver's gaze during the insertion or removal of the forks. A distraction monitoring system comprising: a speed change unit that changes the operating speed of the forklift when the distraction detection unit determines that the driver is distracted.

2. The distraction monitoring system according to claim 1, wherein the distraction detection unit determines that the driver is not distracted when the direction of the driver's gaze is toward the fork or the load loaded on the fork while the fork is being inserted or removed.

3. The distraction detection unit determines that the driver is not driving while distracted if the direction of the driver's gaze is directed forward and the direction of the gaze is within a predetermined angle from the center of the driver's seat in the horizontal direction, and determines that the driver is looking at both or either the forks and the load loaded on the forks.

4. The distraction monitoring system according to claim 1, wherein the operating speed of the forklift changed by the speed change unit includes both or either the travel speed of the forklift and the lifting speed of the forks.

5. The forklift includes a turning detection unit that detects the turning operation of the forklift, The distraction monitoring system according to claim 1, further comprising a warning unit that warns the driver when the turning operation is detected while the fork is being inserted or removed.

6. The forklift is further equipped with a turning detection unit that detects the turning operation of the forklift, The distraction monitoring system according to claim 1, wherein the speed change unit reduces the maximum travel speed of the forklift when the turning operation is detected while the forks are being inserted or removed, even if it is determined that the driver is not distracted.

7. The distraction monitoring system according to claim 1, wherein the speed change unit stops the raising and lowering of the forks when it is determined that the driver is distracted while the forks are being inserted or removed, and the height of the forks is above a predetermined height.

8. The distraction monitoring system according to claim 1, wherein the speed change unit returns the modified operating speed of the forklift to its original value after it has been confirmed that the forks have been properly inserted or removed.

9. The distraction monitoring system according to claim 1, wherein the speed change unit determines that the driver is distracted while inserting or removing the forks and changes the speed of the forklift, and then, if it is confirmed that the inserting or removing of the forks was performed properly, it returns the operating speed of the forklift to its original speed even while the forks are being inserted or removed.

10. The system further includes a load presence / absence detection unit that detects whether or not the fork is loaded based on the fork image, The distraction monitoring system according to claim 1, wherein the speed change unit, after inserting or removing the forks, detects that the forks are not loaded with the load, and restores the changed operating speed of the forklift to its original setting.

11. The distraction monitoring system according to claim 1, wherein the distraction detection unit determines that the person is distracted only when the person is distracted for a period of time longer than the permissible distraction time while inserting or removing the fork.

12. The distraction monitoring system according to claim 1, further comprising a notification unit that notifies a predetermined location of distraction when it is determined that the person is distracted.

13. A forklift equipped with a distraction monitoring system according to any one of claims 1 to 12.

14. A distraction monitoring program used in a distraction monitoring system having a driver imaging unit that photographs the driver and generates a head image, a fork imaging unit that photographs the forks of a forklift and generates a fork image, and a computer, To the aforementioned computer, Based on the fork image, an insertion / removal detection unit determines whether or not the fork is being inserted or removed, A gaze estimation unit that estimates the direction of the driver's gaze based on the head image, A distraction detection unit determines whether the driver is distracted while inserting or removing the forks, based on the estimated direction of the driver's gaze during the insertion or removal of the forks. A speed change unit and a distraction monitoring program are to be executed if it is determined that the driver is distracted while inserting or removing the forks.

Images