Conveyance system

Abstract

To prevent an operator who operates a manned conveyance vehicle from falling asleep by using an unmanned flying body, in a conveyance system including the manned conveyance vehicle and the unmanned flying body.SOLUTION: An unmanned flying body 2 includes: an imaging part 25 for imaging a face of an operator O who is operating a manned conveyance vehicle 1; and a warning part 26 for giving a warning to the operator O who is operating the manned conveyance vehicle 1 by using a drowsiness determination part 35 for determining the drowsiness degree of the operator O based on the face image of the operator O by the imaging part 25, and an age determination part 36 for determining the age of the operator O based on the face image of the operator O by the imaging part 25. The age determination part 36 includes: a learning model generation part 51 for performing machine learning from teacher data 56 collected in a collection part 50, and for generating and storing a learning model by the machine learning; an acquisition part 55 for acquiring feature data of the face of the operator O at present time by the predetermined time; and a prediction part 52 for acquiring the age from the learning model.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a transport system equipped with a manned transport vehicle and an unmanned aerial vehicle. [Background technology]

[0002] A manned guided vehicle (e.g., a forklift) used in a facility such as a factory or a warehouse is configured to run and operate when an operator boards and operates the vehicle. The forklift is also configured to perform loading and unloading operations for picking up and placing cargo using forks.

[0003] A transport system is known that is configured to guide a manned transport vehicle operated by an operator using an unmanned aerial vehicle that can stop in the air (see Patent Document 1, etc.).

[0004] In the transportation system, the unmanned aerial vehicle is equipped with a projector that projects a guidance image onto a road surface. The guidance image, for example, displays an arrow pointing to a specific direction and is projected onto the road surface in front of the manned transport vehicle. This allows an operator operating the manned transport vehicle to be guided to a loading position by checking the guidance image.

[0005] However, in conventional transport systems, there is a problem in that it is not possible to prevent the operator operating the manned transport vehicle from falling asleep. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2020-52629 A Summary of the Invention [Problem to be solved by the invention]

[0007] Therefore, the problem that the present invention aims to solve is to provide a transport system that includes a manned guided vehicle and an unmanned aerial vehicle, which can prevent an operator who operates the manned guided vehicle using the unmanned aerial vehicle from falling asleep. [Means for solving the problem]

[0008] In order to solve the above problems, the transportation system of the present invention is a transportation system equipped with a manned transport vehicle and an unmanned aerial vehicle, wherein the unmanned aerial vehicle is equipped with a photographing unit that photographs the face of an operator operating the manned transport vehicle, a drowsiness determination unit that determines the operator's level of drowsiness based on the facial image of the operator captured by the photographing unit, and an age determination unit that determines the operator's age based on the facial image of the operator captured by the photographing unit, and an alarm unit that issues an alarm to the operator operating the manned transport vehicle based on the level of drowsiness and age. The age determination unit includes a collection unit that collects feature data related to human facial features and teacher data based on the relationship between human facial features and age; a learning model generation unit that performs machine learning from the teacher data collected in the collection unit and generates and stores a learning model by machine learning; an acquisition unit that acquires the current facial feature data of the operator at predetermined time intervals using the shooting unit; a prediction unit that acquires the age from the learning model by inputting the current facial feature data of the operator acquired from the acquisition unit into the learning model generated by the learning model generation unit; and a decision unit that decides whether to issue an alert to the operator based on the age acquired by the prediction unit.

[0009] Preferably, the drowsiness determination unit includes a collection unit that collects teacher data based on the relationship between feature data related to human facial features and the degree of drowsiness related to the human face, a learning model generation unit that performs machine learning from the teacher data collected in the collection unit and generates and stores a learning model by machine learning, an acquisition unit that acquires current facial feature data of the operator at predetermined time intervals using the shooting unit, a prediction unit that acquires the degree of drowsiness from the learning model by inputting the current facial feature data of the operator acquired from the acquisition unit into the learning model generated by the learning model generation unit, and a decision unit that decides whether to alert the operator based on the degree of drowsiness acquired by the prediction unit.

[0010] Preferably, when the degree of drowsiness is equal to or higher than a predetermined level and the age is equal to or lower than a predetermined age, the warning unit issues a warning to those around the manned guided vehicle in addition to the operator.

[0011] Moreover, it is preferable that the alarm unit issues an alarm using any one or a combination of sound, light, and water.

[0012] The unmanned aerial vehicle may also be equipped with a projection unit that projects an image onto the passageway, and the alarm unit may project an alarm image from the projection unit to warn those around the manned guided vehicle.

[0013] The unmanned aerial vehicle may also be configured to travel along a guideway that guides the travel of a manned guided vehicle. Effect of the Invention

[0014] The transport system according to the present invention is a transport system equipped with a manned transport vehicle and an unmanned air vehicle, and the unmanned air vehicle can effectively prevent the operator operating the manned transport vehicle from falling asleep. [Brief description of the drawings]

[0015] [Figure 1] FIG. [Diagram 2] FIG. [Diagram 3]FIG. [Figure 4] FIG. 1 is a block diagram showing a transport system. [Diagram 5] FIG. 4 is a block diagram showing the configuration of a drowsiness determination unit. [Figure 6] FIG. 4 is a block diagram showing a configuration of an age determination unit. [Figure 7] FIG. 4 is a flowchart showing a control procedure for preventing drowsiness at the wheel. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, an embodiment of a transport system according to the present invention will be described with reference to the drawings.

[0017] The configuration of the transport system will be described with reference to FIGS.

[0018] 1 to 4, the transport system S includes a manned transport vehicle 1 that an operator O boards and operates. The manned transport vehicle 1 is configured to travel and operate when the operator O boards and operates it. In this embodiment, the manned transport vehicle 1 is a counterbalance type forklift, and is configured to be able to travel the vehicle body and raise and lower the forks when the operator O boards and operates it.

[0019] The transport system S comprises a plurality of shelves R installed in a facility such as a factory or warehouse. The shelves R have a plurality of steps in the height direction and are configured so that luggage L can be stored at a predetermined position on the steps. The manned transport vehicle 1 performs loading and unloading by placing and picking up luggage L at a predetermined position on the shelves R. The shelves R are arranged at intervals of a predetermined width so that the manned transport vehicle 1 can travel and perform loading and unloading, and an aisle P is formed between each shelf R (FIGS. 1 and 3).

[0020] The transport system S includes an unmanned aerial vehicle 2 capable of stopping in the air. The unmanned aerial vehicle 2 is called a drone, and is configured to fly to a predetermined stopping position in the air by rotating rotors provided at the tip of each of a number of arms, and to hover at the predetermined stopping position in the air.

[0021] The transport system S includes a management device 3 for controlling the unmanned aerial vehicle 2 (FIG. 4). The management device 3 includes a memory unit 30. The memory unit 30 stores a map M including shelves R and aisles P installed within a facility, luggage L placed within the facility, and the like.

[0022] Furthermore, the storage unit 30 stores the loading and unloading tasks T to be performed by the manned guided vehicle 1 as a loading and unloading schedule J. That is, the loading and unloading schedule J has a plurality of loading and unloading tasks T set in a predetermined order, such as task T1 of picking up the luggage L from a predetermined location on a predetermined shelf R, task T2 of placing the luggage L at a predetermined location on a predetermined shelf R, task T3 of placing the luggage L at a shipping location, task T4 of picking up the luggage L from a receiving location, etc. Also, the loading and unloading tasks T include position information of the luggage L and loading and unloading (picking up or placing) information for the luggage L.

[0023] The management device 3 is equipped with a loading / unloading instruction unit 34, which is configured to display the loading / unloading tasks T of the loading / unloading schedule J transmitted from the memory unit 30 on a display unit 11 provided in the driver's seat of the manned guided vehicle 1.

[0024] The display unit 11 is configured, for example, with a touch panel display. The loading / unloading instruction unit 34 displays the loading / unloading task T to be performed by the manned guided vehicle 1 on the display unit 11. The operator O operates the manned guided vehicle 1 to perform loading / unloading in accordance with the loading / unloading task T displayed on the display unit 11. When the loading / unloading task T is completed, the operator O presses an end button displayed on the display unit 11, and an end signal is sent to the loading / unloading instruction unit 34. When the loading / unloading instruction unit 34 receives the end signal, it is configured to display the loading / unloading task T to be performed next by the manned guided vehicle 1 on the display unit 11.

[0025] The manned guided vehicle 1 is equipped with a position detection unit 10. The position detection unit 10 is configured with a laser sensor, a GPS sensor, an electromagnetic induction sensor, etc. The position detection unit 10 is configured to detect a vehicle position D1 of the manned guided vehicle 1.

[0026] The management device 3 includes a taxiway generating unit 31. The taxiway generating unit 31 generates a virtual taxiway 4 between the vehicle position D1 of the manned guided vehicle 1 and a loading position D2 based on information on the vehicle position D1 of the manned guided vehicle 1 transmitted from the position detecting unit 10, the facility map M transmitted from the memory unit 30, and the loading task T of the loading schedule J transmitted from the memory unit 30. The loading position D2 is a position on the passage P where the manned guided vehicle 1 picks up and places a load in the loading task T (FIG. 3).

[0027] 3, the taxiway generating unit 31 is configured to generate a virtual taxiway 4 that connects, for example, a vehicle position D1 and a loading position D2 on a passage P. The taxiway 4 is set, for example, so as to minimize the travel distance of the manned guided vehicle 1. In this embodiment, as shown in FIG. 3, the taxiway 4 is configured with a first straight section 41, a bent section 40, and a second straight section 42.

[0028] The unmanned aerial vehicle 2 is equipped with a position detection unit 20. The position detection unit 20 is composed of a GPS sensor, a gyro sensor, an ultrasonic sensor, a laser sensor, an air pressure sensor, a compass, an acceleration sensor, etc., and can detect the position of the unmanned aerial vehicle 2.

[0029] The unmanned aerial vehicle 2 is equipped with a flight control unit 21. The flight control unit 21 is configured to control the rotation of the rotor blades. Based on the detection result of the position detection unit 20 and the control of the flight control unit 21, the unmanned aerial vehicle 2 can fly to a predetermined air stop position on the taxiway 4 and hover so as to stop in the air at the air stop position.

[0030] The location determination unit 32 is further configured to determine a hovering position on the taxiway 4 where the unmanned aerial vehicle 2 will hover. The location determination unit 32 is configured to stop in the air at a position on the taxiway 4 that is a fixed distance away from the vehicle position D1 of the manned guided vehicle 1 in the forward direction of the manned guided vehicle 1. Therefore, when the manned guided vehicle 1 approaches the unmanned aerial vehicle 2, the unmanned aerial vehicle 2 moves away from the manned guided vehicle 1, so that the distance between the manned guided vehicle 1 and the unmanned aerial vehicle 2 becomes constant.

[0031] The unmanned aerial vehicle 2 includes a memory unit 22. The memory unit 22 stores a guidance image 200. The guidance image 200 is configured, for example, with an arrow for guiding the manned guided vehicle 1 to the loading position D2, and is configured so that the direction of the arrow differs depending on the loading position D2 (FIGS. 1 and 3).

[0032] The unmanned aerial vehicle 2 is equipped with a projection unit 23. The projection unit 23 is composed of, for example, a projector, and can project a guidance image 200 stored in a memory unit 22 onto an aisle P of a facility (FIGS. 1 to 3).

[0033] The management device 3 includes a projection instruction unit 33. The projection instruction unit 33 is configured to determine a guidance image 200 to be projected on the road surface P according to the taxiway 4 from the taxiway generation unit 31 and the position of the unmanned aerial vehicle 2 detected by the position detection unit 20, and to send a projection instruction to the projection unit 23 of the unmanned aerial vehicle 2.

[0034] The operator O can visually view the guidance image 200 projected onto the aisle P, drive the manned guided vehicle 1 along the guidance image 200 to the loading position D2, and operate the manned guided vehicle 1 in accordance with the loading task T displayed on the display unit 11 to perform loading work on the luggage L.

[0035] The unmanned aerial vehicle 2 is equipped with an imaging unit 25 having a CCD image sensor, a CMOS image sensor, etc. The imaging unit 25 is configured to continuously capture images of the face of the operator O who drives the manned guided vehicle 1 by driving the imaging unit 25 so that the imaging direction is always facing the operator O based on the position of the manned guided vehicle 1 from the position detection unit 10 and the position of the unmanned aerial vehicle 2 from the position detection unit 20.

[0036] The management device 3 also includes a drowsiness determination unit 35. The drowsiness determination unit 35 is configured to determine the drowsiness level (for example, levels 1 to 5) of the operator O based on the face image data of the operator O acquired by the imaging unit 25.

[0037] Using NEDO's evaluation method, drowsiness level 1 would be "not sleepy at all," drowsiness level 2 would be "slightly sleepy," drowsiness level 3 would be "sleepy," drowsiness level 4 would be "quite sleepy," and drowsiness level 5 would be "very sleepy."

[0038] 5, the drowsiness determination unit 35 includes a collection unit 40 that collects teacher data 46. The teacher data 46 includes feature data D related to human facial features. In this embodiment, the feature data D related to various human facial features includes eye opening degree C1, eyelid opening and closing frequency C2, gaze movement amount C3, mouth opening degree C4, head up and down movement C5, and yawning frequency C6.

[0039] The drowsiness determination unit 35 includes a learning model generation unit 41 that performs machine learning from the teacher data 46 collected by the collection unit 40 and generates and stores a learning model by machine learning. The learning model generation unit 41 of this embodiment performs supervised learning. In supervised learning, a large amount of teacher data 46, i.e., pairs of input data ID and output data OD, are input to the learning model generation unit 41.

[0040] The input data ID includes the eye opening degree C1, the number of times the eyelids are opened and closed C2, the amount of gaze movement C3, the mouth opening degree C4, the up and down movement of the head C5, and the number of times the operator yawns C6. The output data OD is the degree of drowsiness. The input data ID is evaluated as the degree of drowsiness indicating the drowsiness of the operator O, and a numerical parameter (level) from 1 to 5 is set.

[0041] Examples of cases where the numerical parameter (level) of the drowsiness level is high, that is, the drowsiness level of the operator O is determined to be high include (1) when the eye opening degree C1 is lower than the pre-stored eye opening degree, (2) when the number of eyelid openings C2 is less than the pre-stored number of eyelid openings, (3) when the amount of gaze movement C3 is less than the pre-stored amount of gaze movement, (4) when the mouth opening degree C4 is higher than the pre-stored mouth opening degree, (5) when the head up and down movement C5 is more than the pre-stored head up and down movement, and (6) when the number of yawns C6 is more than the pre-stored number of yawns. The drowsiness determination unit 35 determines the drowsiness level of the operator O based on the numerical parameters (1) to (6) above.

[0042] The degree of drowsiness may be set by any one of the numerical parameters of (1) eye opening degree C1, (2) number of times the eyelids open and close C2, (3) amount of gaze movement C3, (4) mouth opening degree C4, (5) up and down movement of the head C5, and (6) number of times the user yawns C6, or by a numerical parameter weighted by a weighting coefficient.

[0043] In fact, the drowsiness level of the operator O can often be easily recognized by the state and behavior of the operator O's eyes, eyelids, mouth, etc. Therefore, it can be inferred that there is a certain relationship, such as a correlation, between (1) eye opening degree C1, (2) number of times the eyelids are opened and closed C2, (3) amount of line of sight movement C3, (4) mouth opening degree C4, (5) head up and down movement C5, (6) number of times the operator yawns C6, and the drowsiness level of the operator O.

[0044] The learning model generation unit 41 uses a machine learning algorithm such as a general neural network. The learning model generation unit 41 performs machine learning using correlated input data ID and output data OD as teacher data 46 to generate a model (learning model) that estimates an output from an input, that is, a model that outputs a drowsiness level when an input data ID is input.

[0045] The drowsiness determination unit 35 includes an acquisition unit 45 that acquires the current input data ID at every predetermined time. As described above, the input data ID includes (1) eye opening degree C1, (2) eyelid opening and closing frequency C2, (3) gaze movement amount C3, (4) mouth opening degree C4, (5) head up and down movement C5, and (6) yawning frequency C6. The acquisition unit 45 is configured to acquire the input data ID by a known image analysis technique based on the face image data of the operator O captured by the imaging unit 25. The input data ID is acquired at every predetermined time (for example, every second).

[0046] The drowsiness determination unit 35 is equipped with a prediction unit 42 that predicts a numerical parameter indicating the degree of drowsiness of the operator O by applying the learning model generated by the learning model generation unit 41 to the current input data ID acquired from the acquisition unit 45.

[0047] The drowsiness determination unit 35 includes a decision unit 43, which decides whether to issue a warning to the operator O based on the output data OD predicted by the prediction unit 42, and outputs a decision signal to the warning unit .

[0048] The drowsiness determination unit 35 determines that the driver is not drowsy when the drowsiness level is level 1, and determines that the driver is drowsy when the drowsiness level is levels 2 to 5.

[0049] The management device 3 also includes an age determination unit 36. The age determination unit 36 ​​is configured to determine the age of the operator O based on the face image data of the operator O acquired by the imaging unit 25.

[0050] 6, the age determination unit 36 ​​includes a collection unit 50 that collects teacher data 56. The teacher data 56 includes feature data D related to human facial features. In this embodiment, the feature data D related to various human facial features includes facial skin E1, the relationship between the position and size of the eyes E2, the relationship between the position and size of the nose E3, the relationship between the position and size of the mouth E4, and skin color E5.

[0051] The age determination unit 36 ​​includes a learning model generation unit 51 that performs machine learning from the teacher data 56 collected by the collection unit 50 and generates and stores a learning model by machine learning. The learning model generation unit 51 of this embodiment performs supervised learning. In supervised learning, a large amount of teacher data 56, i.e., pairs of input data ID and output data OD, are input to the learning model generation unit 51.

[0052] The input data ID includes facial skin E1, the relationship between the position and size of the eyes E2, the relationship between the position and size of the nose E3, the relationship between the position and size of the mouth E4, and the skin color E5. The output data OD is age. The input data ID is evaluated as the age of the operator O, and the numerical parameter of age is set.

[0053] Examples of cases where the numerical parameter of age is high, that is, the age of the operator O is determined to be high, are (1) when the facial skin E1 has more wrinkles than the number of pre-stored wrinkles, (2) when the relationship E2 between the position and size of the eyes is significantly different from the pre-stored position and size relationship, (3) when the relationship E3 between the position and size of the nose is significantly different from the pre-stored position and size relationship, (4) when the relationship E4 between the position and size of the mouth is significantly different from the pre-stored position and size relationship, and (5) when the skin color E5 is darker than the pre-stored skin color. The drowsiness determination unit 36 ​​determines the age of the operator O based on the numerical parameters (1) to (5) above.

[0054] The age may be set by any one of the numerical parameters of (1) facial skin E1, (2) the relationship between the position and size of the eyes E2, (3) the relationship between the position and size of the nose E3, (4) the relationship between the position and size of the mouth E4, and (5) skin color E5, or by a weighted average of the numerical parameters using a weighting coefficient.

[0055] In fact, the age of operator O can often be easily determined by the condition of the facial skin, eyes, nose, mouth, and skin of operator O. Therefore, it can be inferred that there exists a certain relationship, such as a correlation, between (1) facial skin E1, (2) relationship between the position and size of the eyes E2, (3) relationship between the position and size of the nose E3, (4) relationship between the position and size of the mouth E4, and (5) skin color E5 and the age of operator O.

[0056] The learning model generation unit 51 uses a machine learning algorithm such as a general neural network. The learning model generation unit 51 performs machine learning using correlated input data ID and output data OD as teacher data 56 to generate a model (learning model) that estimates an output from an input, that is, a model that outputs a drowsiness level when an input data ID is input.

[0057] The age determination unit 36 ​​includes an acquisition unit 55 that acquires the current input data ID at predetermined time intervals. As described above, the input data ID includes (1) facial skin E1, (2) relationship between eye position and size E2, (3) relationship between nose position and size E3, (4) relationship between mouth position and size E4, and (5) skin color E5. The acquisition unit 55 is configured to acquire the input data ID by a known image analysis technique based on the facial image data of the operator O captured by the imaging unit 25. The input data ID is acquired at predetermined time intervals (for example, every second).

[0058] The age determination unit 36 ​​is equipped with a prediction unit 52 that predicts a numerical parameter indicating the age of the operator O by applying the learning model generated by the learning model generation unit 51 to the current input data ID acquired from the acquisition unit 55.

[0059] The age determination unit 36 ​​includes a decision unit 53, which decides whether to issue a warning to the operator O based on the output data OD predicted by the prediction unit 52, and outputs a decision signal to the warning unit 26.

[0060] The unmanned aerial vehicle 2 is equipped with an alarm unit 26. The alarm unit 26 is configured to issue an alarm to the operator O and workers around the manned guided vehicle 1 based on the determinations from the drowsiness determination unit 35 and the age determination unit 36.

[0061] The alarm unit 26 may be an audio device configured to emit a sound as an alarm. The volume of the sound may be changed according to the level of drowsiness and age. The alarm unit 26 may be configured to emit light as an alarm, with the intensity of the light being variable depending on the level of drowsiness and age. The alarm unit 26 may be configured to emit water as an alarm, with the strength of the water being changed according to the level of drowsiness and age.

[0062] The control procedure for preventing drowsiness will be described with reference to Fig. 7. Note that to avoid duplication, parts that have already been explained will be omitted.

[0063] The drowsiness determination unit 35 and the age determination unit 36 ​​collect teacher data 46, 56 by the collection units 40, 50 (step S1). Then, the learning model generation units 41, 51 perform machine learning from the teacher data 46, 56 collected by the collection units 40, 50 in step S1, and generate and store a learning model by machine learning (step S2).

[0064] The drowsiness determination unit 35 and the age determination unit 36 ​​acquire image data of the face of the operator O captured by the imaging unit 25 as the current input data ID at predetermined time intervals using the acquisition units 45, 55 (step S3).

[0065] The drowsiness determination unit 35 predicts the drowsiness level of the operator O by applying the learning model generated in step S2 by the prediction unit 42 to the current input data ID acquired in step S3 (step S4).

[0066] Furthermore, the age determination unit 36 ​​predicts the age according to the operator O by applying the learning model generated in step S2 by the prediction unit 52 to the current input data ID acquired in step S3 (step S5).

[0067] The drowsiness determination unit 35 determines the drowsiness level of the operator O based on the output data OD predicted by the determination unit 43 in step S4, and determines whether or not the operator O is drowsy (step S6). If the drowsiness determination unit 35 determines the drowsiness level to be level 1, it is determined that the operator O is not drowsy, and if the drowsiness determination unit 35 determines the drowsiness level to be levels 2 to 5, it is determined that the operator O is drowsy.

[0068] If the determination unit 43 determines that the operator O is not drowsy (Level 1), the process returns to step S3, and the photographing unit 25 of the unmanned aerial vehicle 2 photographs the face of the operator O of the manned guided vehicle 1, and the acquisition unit 45 acquires the facial image data. If the determination unit 43 determines that the operator O is drowsy (Levels 2 to 5), it determines which of Levels 2 to 5 the drowsiness level is (Step S6).

[0069] If the drowsiness determination unit 35 determines that the drowsiness level is level 2 or 3 (step S7) and the age determination unit 36 ​​determines that the operator O's age is 40 years or older (step S8), the drowsiness level is light and the depth of sleep is shallow when the operator is older, so if an alarm is issued to the operator O, there is a high possibility that the operator O will wake up. Therefore, the decision units 43, 53 issue an alarm only to the operator O (step S9).

[0070] Then, when the drowsiness level is at levels 4 and 5 (step S7), the operator O is determined to be deeply dozing, and since there is a high possibility that the operator O will wake up even if an alarm is issued to the operator O, the decision unit 43 issues an alarm to not only the operator O but also the workers around the manned guided vehicle 1 (step S10).

[0071] Furthermore, even if the drowsiness determination unit 35 determines that the drowsiness level is level 2 or 3 (step S7) and the age determination unit 36 ​​determines that the operator O's age is under 40 years old (step S8), the drowsiness level is light, but since a young age means a deep sleep, the operator O is unlikely to wake up even if an alarm is issued to the operator O, so the decision unit 53 alerts not only the operator O but also the workers around the manned guided vehicle 1 (step S10).

[0072] When the alarm unit 26 consists of an acoustic device, the acoustic device is equipped with a directional sound generating unit, and when warning only the operator O, a sound is emitted locally toward the operator O, and when warning workers around the manned guided vehicle 1 in addition to the operator O, a sound is emitted over a wide area toward the operator O.

[0073] Furthermore, when the alarm unit 26 is composed of a lighting device, the lighting device is provided with a light generating unit having directionality, and when warning only the operator O, light is emitted locally toward the operator O, and when warning workers around the manned guided vehicle 1 in addition to the operator O, light is emitted over a wide range toward the operator O. Furthermore, the lighting device may issue a warning by flashing light.

[0074] In addition, when the alarm unit 26 consists of a water-discharging device, the water-discharging device is equipped with a directional water injection unit, and when warning only the operator O, water is sprayed locally toward the operator O, and when warning workers around the manned guided vehicle 1 in addition to the operator O, water is sprayed over a wide area toward the operator O.

[0075] In addition, the alarm unit 26 may use the projection unit 25 to display a large alarm image (for example, an alarm message such as "The driver of this forklift is asleep! Be careful!") in the aisle P around the manned transport vehicle 1, thereby alerting workers around the manned transport vehicle 1 that the operator O operating the manned transport vehicle 1 is asleep.

[0076] Then, the dozing prevention control of steps S3 to S10 is repeatedly performed so as to successively determine whether the operator O is dozing and to issue an alarm.

[0077] Although the preferred embodiments of the present invention have been described above, the configuration of the present invention is not limited to these embodiments. For example, the following modifications are possible.

[0078] In the above embodiment, the unmanned aerial vehicle 2 is configured to travel along a taxiway 4 for guiding the travel of the manned guided vehicle 1, but it may also be configured to photograph the operator O of the manned guided vehicle 1 at any time and along any route without traveling along the taxiway 4.

[0079] In the above embodiment, the drowsiness determination unit 35 may be configured to determine the drowsiness level based on the head angle, facial expression, changes in eyelid opening and closing, etc. of the operator O. The drowsiness determination unit 35 may also be configured to store a facial image of the operator O when he is not drowsy, and to determine the drowsiness level by comparing the stored facial image of the operator O with the facial image of the operator O acquired by the imaging unit 25.

[0080] In addition, the age determination unit 36 ​​may be configured using known image determination techniques such as biological features, local binary patterns, scale-invariant feature transformation, and deep learning.

[0081] In the above embodiment, the alarm unit 26 is composed of either an audio device, a lighting device, or a water-discharging device, but these may be selectively combined, or the unmanned aerial vehicle 2 may be equipped with all of them.

[0082] In the above embodiment, the unmanned aerial vehicle 2 is configured to project a guidance image 200 onto the passage P so that the operator O can visually recognize the taxiway 4, but may also be configured to include a voice generating unit (not shown) that emits sounds such as voice, buzzer, and chime so that the operator O can hear the taxiway 4. The voice generating unit is configured to emit sounds such as "Turn left 15 m ahead", "Destination 30 m ahead", "There is an obstacle ahead. Please be careful", etc.

[0083] The effects of the present invention will now be described.

[0084] In a transport system S equipped with a manned transport vehicle 1 and an unmanned aerial vehicle 2, the unmanned aerial vehicle 2 is equipped with a photographing unit 25 that photographs the face of an operator O operating the manned transport vehicle 1, a drowsiness determination unit 35 that determines the degree of drowsiness of the operator O based on the facial image of the operator O captured by the photographing unit 25, and an age determination unit 36 ​​that determines the age of the operator O based on the facial image of the operator O captured by the photographing unit 25, and an alarm unit 26 that issues an alarm to the operator O operating the manned transport vehicle 1 based on the degree of drowsiness and age. The age determination unit 36 ​​also includes a collection unit 50 that collects teacher data 56 based on feature data regarding human facial features and the relationship between human face and age, a learning model generation unit 51 that performs machine learning from the teacher data 56 collected by the collection unit 50 and generates and stores a learning model by machine learning, an acquisition unit 55 that acquires facial feature data of the current operator O at predetermined time intervals using the shooting unit 25, a prediction unit 52 that acquires the age from the learning model by inputting the facial feature data of the current operator O acquired from the acquisition unit 55 into the learning model generated by the learning model generation unit 51, and a decision unit 53 that decides whether to issue an alert to the operator O based on the age acquired by the prediction unit 52.

[0085] Therefore, by providing the photographing unit 25 on the unmanned aerial vehicle 2 and photographing from outside the manned guided vehicle 1, it is not necessary to provide the photographing unit 25 on each manned guided vehicle 1, and the configuration for photographing the operator O can be minimized. In addition, since the operator O does not know when the unmanned aerial vehicle 2 will fly in, it is expected that the operator O will operate the manned guided vehicle 1 with a sense of tension. In addition, the drowsiness determination unit 35 determines the drowsiness level of the operator O, and the age determination unit 36 ​​determines the age of the operator O, so that when the operator O falls asleep, an alarm can be issued to the operator O to wake him up. In addition, by using a learning model generated by machine learning, the age of the operator O can be determined more accurately.

[0086] The drowsiness determination unit 35 also includes a collection unit 40 that collects teacher data 46 based on the relationship between feature data regarding human facial features and the degree of drowsiness regarding the human face, a learning model generation unit 41 that performs machine learning from the teacher data 46 collected by the collection unit 40 and generates and stores a learning model by machine learning, an acquisition unit 45 that acquires facial feature data of the current operator O at predetermined time intervals using the shooting unit 25, a prediction unit 42 that acquires the degree of drowsiness from the learning model by inputting the facial feature data of the current operator O acquired from the acquisition unit into the learning model generated by the learning model generation unit 41, and a decision unit 43 that decides whether to issue an alert to the operator O based on the degree of drowsiness acquired by the prediction unit 42.

[0087] Therefore, by using a learning model generated by machine learning, it is possible to more accurately determine the degree of drowsiness of the operator O.

[0088] Moreover, it is desirable that the warning unit 26 warns the surroundings of the manned guided vehicle 1 in addition to the operator O when the degree of drowsiness is equal to or higher than a predetermined level.

[0089] If the operator O's level of drowsiness is above a predetermined level, the operator O may not wake up even if an alarm is issued to the operator O. Therefore, by issuing an alarm to workers around the manned transport vehicle 1, it is possible to prevent the workers from colliding with the manned transport vehicle 1 operated by the dozing operator O.

[0090] The alarm unit 26 is adapted to emit any one or a combination of sound, light and water as an alarm. In this way, the alarm unit 26 emits sound, light, and water toward the dozing operator O, thereby waking up the operator O reliably and quickly.

[0091] In addition, the unmanned aerial vehicle 2 may be equipped with a projection unit 23 that projects an image onto the passage P, and the alarm unit 26 may project an alarm image from the projection unit 23 to issue an alarm to those around the manned guided vehicle 1. In this way, by the projection unit 23 projecting the warning image, it is possible to directly and effectively warn workers around the manned guided vehicle 1 that the operator O of the manned guided vehicle 1 is dozing off.

[0092] In addition, the unmanned aerial vehicle 2 is configured to travel along a taxiway 4 that guides the manned guided vehicle 1.

[0093] Therefore, the unmanned aerial vehicle 2 guiding the manned transport vehicle 1 can sequentially and accurately capture facial images of the operator O, which allows the drowsiness determination unit 35 to determine the degree of drowsiness of the operator O and the age determination unit 36 ​​to determine the age of the operator O. As a result, when the operator O falls asleep, a prompt and appropriate warning can be issued to the operator O to wake him up. [Explanation of symbols]

[0094] S Conveyor System O Operator 1 Manned guided vehicle 2 Unmanned Aerial Vehicles 4 Taxiway 23 Projection section 25 Photography Department 26 Alarm section 35 Drowsiness Judgment Unit 36 Age Determination Department 40,50 Collection Department 41,51 Learning model generation unit 42,52 Prediction Section 43,53 Decision section 45,55 Acquisition Department 46,56 Teacher data ID Input Data OD Output Data

Claims

1. In a transportation system including a manned guided vehicle and an unmanned aerial vehicle, The unmanned aerial vehicle is An imaging unit that images a face of an operator operating the manned guided vehicle; a drowsiness determination unit that determines a drowsiness level of the operator based on a facial image of the operator captured by the imaging unit, and an alarm unit that issues an alarm to the operator operating the manned guided vehicle based on the drowsiness level and the age using an age determination unit that determines an age of the operator based on the facial image of the operator captured by the imaging unit, The age determination unit is A collection unit that collects feature data related to human facial features and teacher data based on a relationship between the human face and age; a learning model generation unit that performs machine learning on the teacher data collected by the collection unit, and generates and stores a learning model by the machine learning; an acquisition unit that acquires current facial feature data of the operator at predetermined time intervals by the photographing unit; a prediction unit that acquires the age from the learning model by inputting the facial feature data of the operator at the current time acquired from the acquisition unit into the learning model generated by the learning model generation unit; a decision unit that decides whether to issue a warning to the operator based on the age acquired by the prediction unit. A transport system comprising:

2. The drowsiness determination unit is A collection unit that collects teacher data based on a relationship between feature data related to features of a human face and a sleepiness level related to the human face; a learning model generation unit that performs machine learning on the teacher data collected by the collection unit, and generates and stores a learning model by the machine learning; an acquisition unit that acquires current facial feature data of the operator at predetermined time intervals by the photographing unit; a prediction unit that acquires the drowsiness level from the learning model generated by the learning model generation unit by inputting the facial feature data of the operator at the current time acquired from the acquisition unit into the learning model; a determination unit that determines whether to issue a warning to the operator based on the drowsiness level acquired by the prediction unit.

2. The transport system according to claim 1.

3. The warning unit issues a warning to the operator and to those around the manned transport vehicle when the drowsiness level is equal to or higher than a predetermined level and the age is equal to or lower than a predetermined age.

2. The transport system according to claim 1.

4. The alarm unit issues any one or a combination of sound, light, and water as the alarm.

2. The transport system according to claim 1.

5. The unmanned aerial vehicle includes a projection unit that projects an image onto the passageway, The warning unit is configured to project a warning image from the projection unit to warn people around the manned guided vehicle.

4. The transport system according to claim 3.

6. The unmanned aerial vehicle is configured to travel along a guideway for guiding the travel of the manned guided vehicle.

2. The transport system according to claim 1.

Images