Loading system and control method for the same

Abstract

To predict and inform a position of an obstacle when a manned conveyance vehicle travels in a facility, at low manufacturing cost, even if the number of work steps is small.SOLUTION: A loading system comprises: a collection unit 40 that collects teacher data 46 based on a relation between environment data relating to a manned conveyance vehicle 1 or the like which conveys a load L and obstacle data containing a position of an obstacle B resulting from falling of the load L; a learning model generation unit 41 that carries out machine learning from the teacher data 46 collected by the collection unit 40 and generates and stores a learning model through the machining learning; an acquisition unit 45 that acquires environment data at current time; a prediction unit 42 that acquires the obstacle data from the learning model by inputting the environment data of the current time acquired from the acquisition unit 45 to the learning model generated by the learning model generation unit 41; and a control unit 43 that performs control so as to inform the position of the obstacle B to an operator O operating the manned conveyance vehicle 1, on the basis of the obstacle data acquired by the prediction unit 42 as the manned conveyance vehicle 1 approaches the position of the obstacle B.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a cargo handling system in which a manned transport vehicle travels inside a facility and a control method thereof.

Background Art

[0002] Conventionally, cargo handling operations have been performed to load and unload goods from storage sections of shelves installed inside facilities such as factories and warehouses using transport vehicles such as forklifts. Among the transport vehicles, there is a manned transport vehicle on which an operator rides and drives.

[0003] By the way, as in Patent Document 1, an obstacle detection system for detecting obstacles inside a facility where a manned transport vehicle travels is disclosed. In this obstacle detection system, a distance measurement sensor is attached to the manned transport vehicle, and the distance measurement sensor detects obstacles to notify an operator who drives the manned transport vehicle.

[0004] In this obstacle detection system, the distance measurement sensor must be attached so as to accurately detect obstacles. Therefore, since it is necessary to manufacture the distance measurement sensor and accurately attach the distance measurement sensor to the manned transport vehicle, there are problems such as high manufacturing costs and many work processes.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Therefore, the problem to be solved by the present invention is to provide a cargo handling system and its control method that can predict and notify the position of an obstacle when a manned transport vehicle travels within a facility, despite having low manufacturing costs and few working processes.

Means for Solving the Problem

[0007] To solve the above problems, the cargo handling system according to the present invention is a cargo handling system including a manned transport vehicle that travels within a facility, a collection unit that collects teacher data based on the relationship between environmental data related to a manned transport vehicle or the like that transports goods and obstacle data including the position of an obstacle generated by the fall of a good; a learning model generation unit that performs machine learning from the teacher data collected by the collection unit and generates and stores a learning model by machine learning; an acquisition unit that acquires current environmental data; a prediction unit that inputs the current environmental data acquired from the acquisition unit into the learning model generated by the learning model generation unit to acquire obstacle data from the learning model; and a control unit that controls to notify an operator who drives the manned transport vehicle when the manned transport vehicle approaches the position of the obstacle based on the obstacle data acquired by the prediction unit.

[0008] Preferably, the environmental data includes the vehicle type of the manned transport vehicle.

[0009] Preferably, the environmental data includes any one or all of the age, years of experience, and gender of the operator who drives the manned transport vehicle.

[0010] Preferably, the environmental data includes the time zone for transporting goods.

[0011] Preferably, the environmental data includes the type of goods.

[0012] Preferably, the obstacle data includes the type of obstacle.

[0013] Moreover, the control method of the cargo handling system according to the present invention is a control method of a cargo handling system including a manned transport vehicle that travels inside a facility, a collection step of collecting teacher data based on the relationship between environmental data regarding a manned transport vehicle or the like that transports cargo and obstacle data including the position of an obstacle generated by the falling of the cargo; a learning model generation step of performing machine learning from the teacher data collected in the collection step and generating and storing a learning model by the machine learning; an acquisition step of acquiring current environmental data; a prediction step of inputting the current environmental data acquired from the acquisition step into the learning model generated in the learning model generation step to acquire obstacle data from the learning model; and a control step of controlling to notify an operator who drives the manned transport vehicle when the manned transport vehicle approaches the position of the obstacle based on the obstacle data acquired in the prediction step.

Advantages of the Invention

[0014] The cargo handling system and its control method according to the present invention do not require a distance measurement sensor or the like to be attached to the manned transport vehicle, and can predict and notify the position of an obstacle when the manned transport vehicle travels inside the facility. Therefore, the manufacturing cost can be reduced and the number of work processes can be reduced.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0016] Hereinafter, based on the drawings, an embodiment of the cargo handling system and its control method according to the present invention will be described.

[0017] As shown in FIGS. 1 and 2, the conveying system includes a manned carrier vehicle 1 on which an operator O rides and operates. The manned carrier vehicle 1 is configured to run and operate when the operator O rides and operates it. In the present embodiment, the manned carrier vehicle 1 is a reach-type forklift, and is configured to be able to perform vehicle running, fork lifting and lowering, etc. when the operator O rides and operates it.

[0018] The conveying system includes a plurality of shelves R installed in facilities such as factories and warehouses. The shelf R has a plurality of stepped portions in the height direction and the horizontal direction, and is configured to be able to store the load L at a predetermined position of the stepped portion. The manned carrier vehicle 1 performs cargo handling by loading and unloading the load L at a predetermined position of the shelf R.

[0019] The conveying system includes a management device 3. As shown in FIG. 3, the management device 3 includes a storage unit 30. The storage unit 30 stores a map M composed of shelves R, passages installed in the facility, loads L arranged in the facility, etc.

[0020] The storage unit 30 stores the cargo handling task T performed by the manned carrier vehicle 1 as a cargo handling schedule J. That is, the cargo handling schedule J is set with a plurality of cargo handling tasks T such as a task of unloading the load L from a predetermined place of a predetermined shelf R, a task of loading the load L at a predetermined place of a predetermined shelf R, a task of loading the load L at the shipping place, a task of unloading the load L from the receiving place, etc. in a predetermined order. Further, the cargo handling task T includes information on the cargo handling position and cargo handling (unloading or loading) information for the load L.

[0021] As shown in FIG. 3, the management device 3 includes a handling instruction unit 34, and the handling instruction unit 34 is configured to display the handling task T of the handling schedule J transmitted from the storage unit 30 on the display unit 11 provided in the driver's seat of the manned carrier vehicle 1.

[0022] The display unit 11 is composed of, for example, a touch panel display. The handling instruction unit 34 displays the handling task T that the manned carrier vehicle 1 should perform on the display unit 11. The operator O drives and operates the manned carrier vehicle 1 according to the handling task T displayed on the display unit 11 to perform the handling work. When the handling task T is completed, the operator O presses the end button displayed on the display unit 11, and an end signal is transmitted to the handling instruction unit 34. The handling instruction unit 34 is configured to display the next handling task T that the manned carrier vehicle 1 should perform on the display unit 11 when it receives the end signal.

[0023] The operator O can drive the manned carrier vehicle 1 to the handling position and operate the manned carrier vehicle 1 according to the handling task T displayed on the display unit 11 to perform the handling work on the load L.

[0024] The manned carrier vehicle 1 includes a position detection unit 10. The position detection unit 10 is composed of a sensor for detecting the surrounding environment such as a laser sensor, a GPS sensor, an electromagnetic induction sensor, or a receiver for receiving signals from positioning satellites. The position detection unit 10 is configured to detect the vehicle position of the manned carrier vehicle 1.

[0025] As shown in FIG. 3, the management device 3 includes an obstacle prediction unit 31. The obstacle prediction unit 31 is configured to predict and output obstacle data including the position of the obstacle B based on the environmental data related to the manned carrier vehicle 1, the operator O, the load L, etc.

[0026] As shown in FIG. 4, the obstacle prediction unit 31 includes a collection unit 40 that collects teacher data 46. The teacher data 46 includes environmental data related to the manned carrier vehicle 1, the operator O, the load L, etc.

[0027] The environmental data includes (1) the type of the load L to be transported, (2) the vehicle type of the manned transporter 1, (3) the age of the operator O, (4) the number of years of experience of the operator O, (5) the gender of the operator O, (6) the time zone when the load L drops, and (7) the travel route of the manned transporter 1. The obstacle data includes (8) the position of the obstacle B generated by the dropping of the load L and (9) the type of the obstacle B.

[0028] The obstacle prediction unit 31 includes a learning model generation unit 41 that performs machine learning from the teacher data 46 (the above (1) to (9)) collected by the collection unit 40 and generates and stores a learning model by the machine learning. The learning model generation unit 41 of the present embodiment performs supervised learning. In supervised learning, a large amount of teacher data 46, that is, a set of input data ID and output data OD, is input to the learning model generation unit 41.

[0029] The input data ID is environmental data and includes (1) the type of the load L, (2) the vehicle type of the manned transporter 1, (3) the age of the operator O, (4) the number of years of experience of the operator O, (5) the gender of the operator O, (6) the time zone when the load L drops, and (7) the travel route of the manned transporter 1. The output data OD is obstacle data and includes (8) the position of the obstacle B generated by the dropping of the load L and (9) the type of the obstacle B.

[0030] In fact, it can be inferred that there is a certain relationship such as a correlation relationship between the position and type of the obstacle B generated by the dropping of the load L and the vertically long and small load L that is likely to drop, the horizontally long and large load L that is not likely to drop, the reach-type forklift which is a vehicle type of the manned transporter 1 that runs and operates smoothly, the counterweight-type forklift which is a vehicle type of the manned transporter 1 with a large inertial force when running and stopping, the age of the operator O that affects the reflex nerve and motor nerve, the number of years of experience and gender of the operator O that affect the running and operation, the time zone that affects the running and operation depending on the brightness in the facility, and the travel route of the manned transporter 1 that affects the running and operation during turning, acceleration, and deceleration.

[0031] 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 the input data ID and the output data OD having a correlation as teacher data 46, thereby generating a model (learning model) for estimating the output from the input, that is, a model that outputs the output data OD when the input data ID is input.

[0032] The obstacle prediction unit 31 includes an acquisition unit 45 that acquires the current input data ID at regular intervals. As described above, the input data ID includes (1) the type of the load L, (2) the vehicle type of the manned transport vehicle 1, (3) the age of the operator O, (4) the number of years of experience of the operator O, (5) the gender of the operator O, (6) the time zone when the load L falls, and (7) the travel route of the manned transport vehicle 1.

[0033] The obstacle prediction unit 31 includes a prediction unit 42 that predicts the position and type of the obstacle B, which is the output data OD, 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.

[0034] The obstacle prediction unit 31 includes a control unit 43. The control unit 43 predicts the position of the obstacle B based on the output data OD predicted by the prediction unit 42, and when the manned transport vehicle 1 approaches the predicted position of the obstacle B based on the predicted position of the obstacle B and the position of the manned transport vehicle 1, the control unit 43 controls the display unit 11 to display the position and type of the obstacle B in order to notify the operator O driving the manned transport vehicle 1 to drive attentively.

[0035] As shown in FIG. 5, the above-described cargo handling system executes the following control method.

[0036] The collection unit 40 collects the teacher data 46 (collection step S1). Then, the learning model generation unit 41 performs machine learning from the teacher data 46 collected by the collection unit 40 in the collection step S1, and generates and stores a learning model by the machine learning (learning model generation step S2). The acquisition unit 45 acquires the current input data ID (acquisition step S3).

[0037] The prediction unit 42 applies the learning model generated in the learning model generation step S2 to the current input data ID acquired in the acquisition step S3, thereby predicting the position and type of the obstacle B (prediction step S4). The control unit 43 performs control so as to display a guidance screen on the display unit 11 based on the output data OD predicted in the prediction step S4 (control step S5).

[0038] As described above, the preferred embodiments of the present invention have been described, but the configuration of the present invention is not limited to these embodiments. For example, when the manned transport vehicle 1 approaches the predicted position of the obstacle B, in the above embodiment, the position and type of the obstacle B are displayed on the display unit 11, but only the position of the obstacle B may be displayed. Further, when notifying the operator O of the position of the obstacle B, instead of displaying it on the display unit 11, for example, a lamp may be provided in the driver's seat of the manned transport vehicle 1 and configured to turn on the lamp. Further, the cargo handling vehicle 1 is not limited to a counterweight type forklift, and may be a reach forklift, a side forklift, an order picking truck, a tractor, or the like.

[0039] The effects of the present invention will be described. The cargo handling system according to the present invention is based on the relationship between the environmental data regarding the manned transport vehicle 1 that transports the cargo L and the obstacle data including the position of the obstacle B generated by the fall of the cargo L. When the manned transport vehicle 1 approaches the position of the obstacle B, it notifies the operator O who drives the manned transport vehicle 1. Therefore, when the manned transport vehicle 1 approaches the predicted position of the obstacle B, the operator O who drives the manned transport vehicle 1 can drive while being careful not to collide with the obstacle B.

[0040] In addition, by including in the environmental data the vehicle type of the manned transporter 1, for example, a reach forklift which is a vehicle type of the manned transporter 1 that runs and operates smoothly, a counterweight forklift which is a vehicle type of the manned transporter 1 with a large inertial force during running and stopping, etc., the correlation with the obstacle B generated by the fall of the load L can be considered, so the position of the obstacle B can be predicted more accurately.

[0041] In addition, by including in the environmental data the age of the operator O that affects the reflex nerves and motor nerves, the number of years of experience and gender of the operator O that affect running and operation, the correlation with the obstacle B generated by the fall of the load L can be considered, so the position of the obstacle B can be predicted more accurately.

[0042] In addition, since the time zone affects running and operation depending on the brightness inside the facility, by including in the environmental data the time zone for transporting the load, the correlation with the obstacle B generated by the fall of the load L can be considered, so the position of the obstacle B can be predicted more accurately.

[0043] In addition, since the type of the load L, for example, a vertically long-shaped or small load L is likely to fall, and a horizontally long-shaped or large load is less likely to fall, by including the type of the load L in the environmental data, the correlation with the obstacle B generated by the fall of the load L can be considered, so the position of the obstacle B can be predicted more accurately.

[0044] In addition, since the obstacle data includes the type of the obstacle B, the type of the obstacle B can be notified to the operator who drives the manned transporter 1, so the operator O can appropriately drive the manned transporter 1 so that the manned transporter 1 runs according to the type of the obstacle B.

Explanation of Signs

[0045] 1 Manned transporter 31 Obstacle prediction unit 40 Collection unit 41 Learning model generation unit 42 Prediction unit 43 Control unit 45 Acquisition unit 46 Teacher data O Operator B Obstacle L Luggage

Claims

1. A cargo handling system comprising a manned transport vehicle that travels within a facility, a collection unit that collects teacher data based on the relationship between environmental data regarding the manned transport vehicle that transports cargo and obstacle data including the positions of obstacles generated by the falling of the cargo, a learning model generation unit that performs machine learning from the teacher data collected by the collection unit and generates and stores a learning model by the machine learning, an acquisition unit that acquires the environmental data at the current time, a prediction unit that inputs the environmental data at the current time acquired from the acquisition unit into the learning model generated by the learning model generation unit, and thereby acquires the obstacle data from the learning model, and a control unit that controls to notify an operator who drives the manned transport vehicle when the manned transport vehicle approaches the position of the obstacle based on the obstacle data acquired by the prediction unit. A cargo handling system characterized by the above.

2. The environmental data includes the vehicle type of the manned transport vehicle. The cargo handling system according to claim 1, characterized by the above.

3. The environmental data includes any one or all of the age, years of experience, and gender of the operator who drives the manned transport vehicle. The cargo handling system according to claim 1, characterized by the above.

4. The environmental data includes the time zone for transporting the cargo. The cargo handling system according to claim 1, characterized by the above.

5. The environmental data includes the type of the cargo. The cargo handling system according to claim 1, characterized by the above.

6. The obstacle data includes the type of the obstacle. The cargo handling system according to claim 1, characterized by the above.

7. A control method for a cargo handling system comprising a manned transport vehicle that travels within a facility, a collection step of collecting teacher data based on the relationship between environmental data regarding the manned transport vehicle that transports cargo and obstacle data including the positions of obstacles generated by the falling of the cargo, a learning model generation step of performing machine learning from the teacher data collected in the collection step and generating and storing a learning model by the machine learning, an acquisition step of acquiring the environmental data at the current time, and a prediction step of inputting the environmental data at the current time acquired from the acquisition step into the learning model generated in the learning model generation step, and thereby acquiring the obstacle data from the learning model. Based on the obstacle data obtained by the prediction step, when the manned transport vehicle approaches the position of the obstacle, a control step of controlling to notify an operator who drives the manned transport vehicle; and, comprising A method for controlling a cargo handling system, characterized by the above.

Images