JP2020028196A - Determination device and determination method - Google Patents

Abstract

To avoid running out of a battery while working.SOLUTION: A determination device (1) includes a decrease degree identification unit (104) that identifies the degree of decrease in the remaining battery level in scheduled work by using a learned model in which the correlation between the degree of decrease in the remaining battery power during work using a work vehicle, and the battery status and the work content is machine-learned, and an execution determination unit (105) that determines whether the scheduled work is executable on the basis of the degree of decrease.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a determination device and the like for determining whether or not to perform a task using a work vehicle based on the remaining battery level of the work vehicle.

  2. Description of the Related Art There is known a technique for replacing a battery of a vehicle driven by a battery based on detection that a charged capacity has become equal to or less than a predetermined amount (for example, Patent Document 1 below).

JP-A-5-294147

  However, when the performance of a battery is deteriorated due to an increase in the number of times of use or the like, the rate of decrease in the remaining battery power becomes faster. Therefore, when the same type of work vehicle is mounted with the same type of battery having different degrees of deterioration and the same remaining amount on each of the work vehicles, only the vehicle with the battery with the higher degree of deterioration is performed. During the operation, the battery may run out. In addition, since the method of reducing the battery varies depending on the work load, the battery may not run out when performing a work with no load, while the battery may run out when performing a work with a load. Therefore, when it is determined that the operation can be performed without battery replacement or charging based only on the battery remaining amount, the battery remaining amount decreases rapidly during the operation depending on the degree of deterioration of the battery, the workload, and the like. There was a risk of cutting.

  Also, if it is decided to replace or recharge the battery as soon as possible in order to prevent the battery from running out during the work, it is possible that the battery will be frequently replaced or recharged even though the remaining battery capacity is sufficient. Invite. Such a situation is not preferable because it causes a situation such as a decrease in work efficiency due to an increase in the number of times of battery replacement or the number of times of charge, and an early shortening of the battery life. Conventionally, it is also performed to judge whether battery replacement or charging is necessary at the time before the execution of work based on the rule of thumb of the driver. It is difficult to replace or charge the battery.

  One aspect of the present invention is to determine whether or not the work can be completed without running out of battery during the execution of the work without relying on the empirical rules of the driver at a time before the execution of the work. It is an object of the present invention to realize a determination device or the like that can avoid running out of battery during the execution of.

  In order to solve the above-described problem, a determination device according to an aspect of the present invention includes a reduction degree of the battery remaining amount in a work performed using a work vehicle driven by a battery, a state of the battery, and a state of the battery. Using a learned model obtained by machine-learning the correlation with the content of the work, a reduction degree specifying unit that specifies the reduction degree of the battery remaining amount in the scheduled work to be performed using the work vehicle; An execution determination unit that determines whether the scheduled work is to be performed using the work vehicle based on the degree of decrease identified by the degree identification unit.

  According to another embodiment of the present invention, there is provided a determination method using a determination device, wherein the battery remaining amount in a work performed using a work vehicle driven by a battery is determined. Using a learned model obtained by machine learning the correlation between the degree of decrease of the battery and the state of the battery and the content of the work, and the degree of decrease in the remaining battery level in the scheduled work to be performed using the work vehicle. And an execution determination step of determining whether the scheduled work using the work vehicle is to be performed based on the reduction degree identified in the reduction degree identification step.

  According to one embodiment of the present invention, it is possible to avoid running out of battery during execution of a task without relying on a rule of thumb of a driver.

It is a block diagram showing an example of the important section composition of the judging device concerning Embodiment 1 of the present invention. It is a figure showing the outline of the judgment by the above-mentioned judging device. It is a flowchart which shows an example of the process which the said determination apparatus performs. It is a block diagram showing an example of the important section composition of the judging device concerning Embodiment 2 of the present invention.

[Embodiment 1]
(Overview of the judgment device)
An outline of the determination device of the present embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating an outline of the determination by the determination device 1. The determination device 1 is a device that determines whether or not a scheduled work can be performed by the work vehicle based on the remaining battery level of the work vehicle driven by the battery. In the example shown in FIG. 2, the forklift 3A sequentially executes a plurality of works (A, B, C,...) Shown in the work schedule A, and the forklift 3B performs a plurality of works (X, Y, Z,...) Are to be sequentially executed. Each of the forklifts 3A and 3B is an example of a work vehicle driven by a battery. When it is not necessary to distinguish these, a forklift 3 is described.

  When the remaining battery level of the forklift 3A is low, the determination device 1 specifies the state of the battery of the forklift 3A, and specifies the scheduled work to be performed next by the forklift 3A from the work schedule A. Then, the determination device 1 specifies the degree of decrease in the remaining battery level of the forklift 3 when performing the scheduled work, and determines whether the scheduled work can be performed based on the specified degree of decrease. Similarly, the determination device 1 specifies the state of the battery for the forklift 3B and also specifies the next scheduled work to be performed by the forklift 3B from the work schedule table B. Note that the scheduled work may be a single operation of unloading one piece of luggage from the shelf by the forklift 3 or a work of a plurality of operations of unloading a group of luggage from the truck and storing the same on the shelf. There may be.

  In the example of FIG. 2, the determination device 1 determines that the forklift 3A cannot perform the scheduled operation (specifically, the remaining battery power is insufficient to complete the scheduled operation). 3B determines that the scheduled task can be executed (specifically, the remaining battery power is sufficient to complete the scheduled task). Therefore, the determination device 1 notifies the driver of the forklift 3A of stopping the execution of the scheduled work. On the other hand, such notification is not given to the forklift 3B. That is, the forklift 3B is caused to execute the scheduled work as it is.

  As described above, according to the determination device 1, at the time before the execution of the scheduled work, whether or not the scheduled work can be completed without running out of the battery during the execution of the scheduled work is determined based on the rule of thumb of the driver. It can be determined without resorting to it. Thus, it is possible to prevent the forklift 3A from running out of battery during scheduled work.

(Main configuration of the judgment device)
The configuration of the main part of the determination device 1 will be described with reference to FIG. FIG. 1 is a block diagram illustrating an example of a main configuration of the determination device 1. As illustrated, the determination device 1 includes a control unit 10 that controls each unit of the determination device 1 in an integrated manner, and a storage unit 20 that stores various data used by the determination device 1. Further, the determination device 1 includes an input unit 30 that receives an input operation performed on the determination device 1 and a communication unit 40 that allows the determination device 1 to communicate with another device. Further, the control unit 10 includes a work information acquisition unit 101, a battery information acquisition unit 102, an input data generation unit 103, a decrease degree identification unit 104, and an execution availability determination unit 105. The storage unit 20 stores the battery information 201.

  The work information acquisition unit 101 determines a scheduled work that is to be executed next by a forklift (forklift 3A or 3B in the example of FIG. 2, hereinafter referred to as a target forklift) for which the degree of decrease in the remaining battery level is to be specified. Identify. Specifically, the work information acquisition unit 101 specifies a scheduled work from the work schedule of the target forklift. For example, in the example of FIG. 2, when specifying the scheduled work of the forklift 3A, the work information acquisition unit 101 may specify the planned work from the current time and the work schedule A in which the work for each time is specified. . Needless to say, the method for specifying the scheduled work is not limited to this example. For example, the work information acquisition unit 101 receives the input of the scheduled work and its ID (identification information) via the input unit 30 to specify the scheduled work. May be. Also, for example, when the scheduled work can be specified from an image of the place where the work is performed by the forklift 3A (for example, when the truck has arrived at the unloading place, the unloading has been decided). May specify the scheduled work by analyzing the image.

  The battery information acquisition unit 102 detects the remaining battery level of the target forklift. The remaining battery level may be detected, for example, by acquiring an output value of a remaining level sensor mounted on the target forklift through communication with the target forklift. Then, the battery information acquisition unit 102 determines whether the remaining battery level of the target forklift is equal to or less than a predetermined threshold.

  Further, the battery information acquisition unit 102 acquires battery information indicating the state of the battery of the target forklift. Specifically, the battery information acquisition unit 102 acquires the battery information 201 stored in the storage unit 20. Although details will be described later, the battery information 201 is information correlated with the degree of decrease in the remaining battery level.

  The input data generation unit 103 generates input data to be input to the decrease degree identification unit 104. Although the details will be described later, the input data is data corresponding to the battery information 201 acquired by the battery information acquisition unit 102 and the contents of the scheduled work identified by the work information acquisition unit 101.

  The degree-of-reduction specifying unit 104 specifies the degree of reduction in the remaining battery level of the target forklift in the scheduled work using a trained model generated by supervised machine learning. Although the details will be described later, the learned model causes the neural network to machine-learn the correlation between the degree of decrease in the remaining battery level in the work performed using the forklift powered by the battery, the state of the battery, and the work content. It is generated by this.

  The execution possibility determination unit 105 determines whether or not the scheduled work using the target forklift is executable based on the reduction degree specified by the reduction degree specifying unit 104. Specifically, the execution possibility determination unit 105 determines that the value obtained by subtracting the reduction degree specified by the reduction degree specifying unit 104 from the remaining battery level of the target forklift immediately before the execution of the scheduled work is equal to or greater than a predetermined lower limit value. It is determined that execution is possible, and if it is less than the lower limit, it is determined that execution is not possible. The lower limit may be, for example, a value obtained by adding a predetermined margin to the remaining amount of battery necessary for the forklift 3 to move to a place where the battery can be replaced or charged.

  As described above, the battery information 201 is data indicating the state of the battery of the target forklift, and is updated according to a change in the state of the battery. When a plurality of forklifts can be the target forklifts as in the example of FIG. 2, the battery information 201 is stored for each forklift.

(About generation of trained model and input data)
The generation of the learned model used by the reduction degree specifying unit 104 of the present embodiment to specify the reduction degree of the battery remaining amount in the scheduled work will be described. In the following, an example of generating a learned model of a neural network (NN: Neural Network) will be described. However, if a learned model capable of specifying the degree of decrease in the remaining battery power can be constructed, Other algorithms can be applied. When generating a NN learned model, it is preferable to generate a multilayer NN learned model that can be expected to have high specific accuracy.

  The generation of the trained model includes, as teacher data, the state of the battery immediately before the work, the work content of a certain work, and the degree of decrease in the remaining battery level in the certain work when a work with a forklift was performed in the past. Can be used. Of these data, the state of the battery and the contents of the work immediately before the work are input data, and the degree of decrease in the remaining battery power is the correct answer data.

  The above “battery state” and “work content” are both pieces of information correlated with the degree of decrease in the remaining battery level. For example, at least one of the degree of aging of the battery, the remaining amount of the battery, and the elapsed time since the last charge may be used as the “battery state”. Alternatively, for example, a time-series change in the remaining battery level may be recorded, and a pattern of the pattern may be used as the “battery state”.

  The “work content” includes, for example, movement of luggage, loading and unloading of luggage from a transport vehicle such as a truck, and the like. Further, information indicating the degree of load in the work may be the “work content”. For example, at least any one of the cumulative moving distance of the forklift, the total weight of the load to be conveyed, the total number of times of lifting and lowering of the load, and the total number of times of acceleration of the forklift during the movement in the work (the number of times the acceleration that exceeds the threshold is performed). May be used as information indicating the degree of load in the work.

  Note that the “battery state” and “work content” are both values that are correlated to the degree of decrease in the remaining battery level in order to be input data to the neural network. For example, when the aging degree and the remaining battery level are used as the “battery state”, the “battery state” may be represented by a combination of the cumulative number of times of charging and a numerical value indicating the remaining battery level. The “work content” may be represented by, for example, an ID (identification information) set in advance for each work content, or may be represented by a numerical value indicating a degree of load in the work.

  The “reduction degree” is information indicating how much the remaining battery power has decreased. For example, the “reduction degree” may be expressed as a percentage, with the remaining battery capacity in a fully charged state being 100%. In the present embodiment, the degree of decrease is classified into the following five stages.

(1) The degree of decrease in the remaining battery level is less than 10%. (2) The degree of decrease in the remaining battery level is 10% or more and less than 20%. (3) The degree of decrease in the remaining battery level is 20% or more and less than 30%. (5) Degree of decrease in remaining battery power is 40% or more. In this case, in a work performed using a forklift in the past, for example, when the remaining battery level before the work is 30%. If the remaining battery level after the work is 15%, the correct answer data (degree of decrease) in the teacher data based on the work is classified into the above (2). Note that the correct answer data is not limited to the above example as long as it is data in a format that can determine whether the scheduled work can be performed.

  It is possible to generate a trained model of the degree-of-decrease identification unit 104 by generating such teacher data for each work performed in the past using a forklift and performing machine learning using the generated teacher data. it can. By inputting the input data generated by the input data generation unit 103 to the learned model, the probability that the battery level of the target forklift falls into the above five categories is output. Then, the decrease degree specifying unit 104 specifies the classification having the highest output probability as the decrease degree of the battery of the target forklift. For example, when the probability of the classification of (3) is higher than the probability of any of the other classifications, the reduction degree specifying unit 104 specifies that the degree of reduction of the remaining battery level is 20% or more and less than 30%.

  The degree of battery reduction depends on the type of battery, the type of forklift on which the battery is mounted, and the like. For this reason, in the above-described machine learning, the reduction degree specifying unit 104 generates the reduction degree based on the work performed using the same type of forklift equipped with the same type of battery as the forklift whose battery reduction degree is specified. It is desirable to use teacher data.

(Processing flow)
The flow of the process executed by the determination device 1 will be described with reference to FIG. FIG. 3 is a flowchart illustrating an example of a process (determination method) performed by the determination device 1. Hereinafter, an example in which the determination is performed for the forklift 3 will be described.

  In S1, the battery information acquisition unit 102 detects the remaining battery level of the forklift 3. Further, in S2, the battery information acquisition unit 102 determines whether or not the remaining battery level detected in S1 is equal to or smaller than a threshold. Here, when it is determined that the difference is equal to or smaller than the threshold value (YES in S2), the process proceeds to S3. On the other hand, if it is determined that the battery remaining amount exceeds the threshold (NO in S2), the process returns to S1.

  In S3, the battery information acquisition unit 102 specifies the state of the battery of the forklift 3. Specifically, the battery information acquisition unit 102 specifies the state of the battery of the forklift 3 by acquiring the battery information 201 of the forklift 3 stored in the storage unit 20.

  In S4, the work information acquiring unit 101 specifies a scheduled work of the forklift 3. For example, the work information acquisition unit 101 may specify a scheduled work from the current time and the work schedule A of the forklift 3. As the work schedule table A, for example, the one input to the determination device 1 via the input unit 30 may be used.

  In S5, the input data generation unit 103 generates input data according to the state of the battery specified in S3 and the contents of the scheduled task specified in S4. The format of the input data is the same as the teacher data used for the machine learning. For example, when using a learned model machine-learned with teacher data obtained by combining the cumulative number of charges, the numerical value indicating the remaining battery level, and the ID of the work content, the input data generating unit 103 indicates the cumulative number of charging times and the remaining battery level. Input data is generated by combining a numerical value and a work content ID. Then, the input data generation unit 103 inputs the generated input data to the decrease degree identification unit 104.

  In S6 (decrease degree identification step), when the forklift 3 executes the scheduled operation identified in S4, the decrease degree identification unit 104 identifies the degree of decrease in the remaining battery level in the work.

  In S7 (executability determination step), the execution determination unit 105 determines whether or not the forklift 3 can execute the scheduled work specified in S4 based on the degree of reduction specified by the reduction degree specifying unit 104. Here, when it is determined that the execution is possible (YES in S7), the process returns to S1. On the other hand, if it is determined that execution is not possible (NO in S7), the process proceeds to S8.

  In S8, the execution possibility determination unit 105 instructs the forklift 3 to notify the driver of the forklift 3 of the suspension of the execution of the scheduled work. Then, the forklift 3 that has received this notification notifies the driver to stop performing the scheduled work. Thus, the driver can avoid running out of battery during execution of the scheduled work. After the end of S8, the process returns to S1.

(About the timing of the judgment)
In the above example, whether the scheduled work is executable is determined when the remaining battery power is equal to or less than the threshold value. However, the trigger for determining whether the scheduled work is executable is not limited to this example. For example, the configuration may be such that when one task is completed, it is determined whether the next task, that is, the scheduled task, can be executed.

(Adjustment of battery replacement or charging timing)
Further, the determination device 1 may sequentially determine whether or not a plurality of works scheduled to be performed continuously can be performed. In this case, it is possible to grasp in advance which of the plurality of tasks can be executed. Further, by making such a determination for a plurality of forklifts, it is possible to specify the timing at which the battery of each forklift needs to be replaced or charged. Thereby, the timing of battery exchange or charging of each forklift can also be adjusted so that the timing of battery exchange or charging does not overlap. For example, the forklift 3A can execute the operations A to D, but cannot execute the subsequent operations, and the forklift 3B can execute the operations X to Z, but does not execute the subsequent operations. Suppose that it was impossible. In this case, if the scheduled end times of work D and Z are close to each other, the timing of battery replacement or charging of forklift 3A or 3B may be brought forward so that the timing of battery replacement or charging does not overlap.

(About notification)
Regarding the processing in S8, the mode of notification by the forklift 3 is not particularly limited. For example, when the forklift 3 is provided with an output device such as a speaker, the output device may output a message for prompting the stop of the scheduled operation. Further, for example, when the forklift 3 is provided with a display device, the display device may display the above message to notify the driver. In addition to this, for example, when the forklift 3 is provided with a lighting unit indicating that the remaining battery level is low, the driver is notified by lighting or blinking the lighting unit. It doesn't matter.

  Further, when it is determined that the scheduled work is not executable, the execution availability determination unit 105 may cause the forklift 3 to execute processing related to battery replacement or charging. According to this configuration, instead of executing the scheduled work, the process related to battery replacement or charging is executed, so that it is possible to reliably avoid running out of the battery.

  Note that, for example, when the target forklift is a manned work vehicle such as the forklift 3, the “process related to battery replacement” may be a process of transmitting a notification urging the driver to replace the battery. Good. Such a notification may be made by a device other than the forklift 3 (for example, a portable terminal owned by the driver of the forklift 3), or output to a speaker or the like in a facility where the forklift 3 works. You may. Further, when the determination target is an unmanned work vehicle, the operation of the work vehicle may be controlled to replace the battery. The same applies to the processing related to charging.

(Summary of Embodiment 1)
As described above, the determination device 1 includes the reduction degree specifying unit 104 that specifies the reduction degree of the battery remaining amount in the scheduled work. The degree-of-reduction specifying unit 104 has learned the correlation between the degree of reduction of the remaining battery level in the work performed using the battery-powered forklift 3 and the state of the battery and the content of the work by machine learning. The degree of decrease is specified using a model. In addition, the determination device 1 includes an execution possibility determination unit 105 that determines whether the scheduled work can be performed by the forklift 3 based on the reduction degree specified by the reduction degree specifying unit 104. Therefore, the necessity of battery replacement or charging at the time before the execution of the scheduled work can be determined without depending on the empirical rules of the driver. Further, it is possible to avoid running out of battery during execution of the scheduled work.

[Embodiment 2]
Another embodiment of the present invention will be described below. For convenience of description, members having the same functions as the members described in the first embodiment are denoted by the same reference numerals, and description thereof will not be repeated.

  The determination device 1 of the present embodiment differs from the determination device of the first embodiment in that, in order to more accurately specify the degree of decrease in the remaining battery level, it is determined whether the scheduled work can be performed in consideration of the driver of the forklift 3. ing. The determination device 1 of the present embodiment is also different from the determination device of the first embodiment in that teacher data is generated and a learned model is updated. This will be described with reference to FIG. FIG. 4 is a diagram illustrating a main configuration of the determination device 1 according to the second embodiment of the present invention. As illustrated in FIG. 4, the control unit 10 of the determination device 1 of the present embodiment includes a driver information acquisition unit 121, a teacher data generation unit 122, and a learning unit 123.

  The driver information acquisition unit 121 acquires driver information indicating a driver of the forklift 3. The method of obtaining the driver information is not particularly limited, and the driver information may be obtained by, for example, inputting the driver name or the driver ID into the input unit 30. Further, for example, the driver information may be obtained by analyzing an image of the driver. When acquiring driver information from an image, a convolutional neural network (CNN (Convolutional Neural Network)) machine-learned using the image of the driver as teacher data may be used.

  As described above, in the present embodiment, it is determined whether or not the scheduled operation can be performed in consideration of the driver of the forklift 3. For this reason, the input data generation unit 103 of the present embodiment generates input data including the driver information acquired by the driver information acquisition unit 121.

  In addition, the decrease degree specifying unit 104 of the present embodiment machine-learns the correlation between the decrease degree of the remaining battery level in the operation using the forklift 3, the state of the battery, the work content, and the driver of the forklift 3. The degree of decrease in the remaining battery power is specified using the learned model. That is, the decrease degree specifying unit 104 of the present embodiment uses the learned model generated using the teacher data including the driver information of the forklift 3.

  The teacher data generation unit 122 generates teacher data used for updating the learned model. Specifically, when the scheduled work using the forklift 3 is executed, the teacher data generation unit 122 determines the degree of decrease in the remaining battery level in the scheduled work, the state of the battery immediately before the execution of the scheduled work, and the Teacher data is generated in association with the contents of the scheduled work.

  The learning unit 123 updates the learned model used by the decrease degree identifying unit 104 using the teacher data generated by the teacher data generating unit 122. The generation of the teacher data and the update of the learned model will be described in detail below. Note that the learning unit 123 can also generate a learned model by using teacher data prepared in advance.

(About generating teacher data and updating learned models)
Also in the present embodiment, when it is determined that the scheduled work is executable (YES in S7 of FIG. 3), the scheduled work is performed using the forklift 3. In this case, the teacher data generation unit 122 associates the degree of reduction of the remaining battery level in the scheduled work with the state of the battery immediately before the execution of the scheduled work and the content of the scheduled work, and uses the information to update the learned model. Generate teacher data.

  More specifically, the battery information acquisition unit 102 detects the remaining battery level of the forklift 3 at the time when the scheduled work has been completed. Then, the teacher data generation unit 122 subtracts the detected remaining battery power from the remaining battery power detected in S1 of FIG. 3 to specify the degree of reduction of the remaining battery power in the scheduled work. In addition, the teacher data generation unit 122 acquires the input data generated in S5 of FIG. 3 as information indicating the state of the battery immediately before the execution of the scheduled work and the content of the scheduled work, and identifies the input data and the above specified data. Teacher data is generated in association with the degree of decrease.

  As described above, since the teacher data generating unit 122 generates the teacher data in accordance with the execution of the scheduled work, the teacher data increases each time the forklift 3 performs the work. The learning unit 123 updates the learned model of the decrease degree specifying unit 104 using the teacher data every time a predetermined number of the teacher data generated in this way is accumulated. Thereby, it is possible to maintain or improve the accuracy of specifying the reduction degree by the reduction degree specifying unit 104. Note that the determination device 1 of the first embodiment can also be configured to include the teacher data generation unit 122 and the learning unit 123.

(Summary of Embodiment 2)
As described above, the degree-of-reduction specifying unit 104 of the present embodiment determines the degree of reduction of the remaining battery level in work performed using the forklift 3, the state of the battery, the content of the work, and the driver of the forklift 3. The degree of decrease in the remaining battery level in the scheduled work according to the driver of the forklift 3 is specified using a learned model obtained by machine learning the correlation between the battery and the vehicle. Therefore, the degree of decrease in the remaining battery level can be more accurately specified in consideration of the difference in operation for each driver.

[About processing after determining whether scheduled work can be performed]
When it is determined that the scheduled work cannot be performed (NO in S7 of FIG. 3), the execution possibility determination unit 105 performs another work that can be performed using the remaining battery capacity of the forklift 3 instead of causing the forklift 3 to perform the scheduled work. It may be executed. As a result, the work can be advanced by effectively utilizing the remaining battery power. In this case, the execution possibility determination unit 105 instructs the forklift 3 to stop the execution of the scheduled work and notify the driver of the forklift 3 of another work to be performed instead. Further, when the determination target is an unmanned work vehicle, the execution availability determination unit 105 may control the operation of the work vehicle and execute another work instead of the scheduled work.

  Further, when it is determined that the scheduled work cannot be executed (NO in S7 of FIG. 3), the execution possibility determination unit 105 detects another forklift capable of executing the scheduled work and causes the forklift to execute the scheduled work. Is also good. For example, consider a case where the execution availability determination unit 105 determines that the forklift 3A cannot execute the scheduled work D. In this case, even if the forklift 3B completes all the scheduled operations, it is determined from the output of the decrease degree identifying unit 104 that the remaining battery level is sufficient to execute the scheduled operation D. Alternatively, the scheduled work D may be executed by the forklift 3B. Thus, even if a forklift with a low battery level appears among a plurality of forklifts, the scheduled work can be performed as scheduled.

[About distributed processing]
Part of the processing performed by the determination device 1 described in each of the above embodiments may be performed by one or more devices that are communicatively connected to the determination device 1. For example, the processing executed by the decrease degree identification unit 104 may be executed by an AI server that is communicably connected to the determination device 1. In this case, the determination device 1 generates input data, transmits the input data to the AI server, receives output data from the AI server, and determines whether the scheduled work is executable.

[Example of software implementation]
The control block of the determination device 1 (particularly, each unit included in the control unit 10) may be realized by a logic circuit (hardware) formed on an integrated circuit (IC chip) or the like, or may be realized by software. .

  In the latter case, the determination device 1 includes a computer that executes instructions of a program that is software for realizing each function. This computer includes, for example, one or more processors and a computer-readable recording medium storing the program. Then, in the computer, the object of the present invention is achieved when the processor reads the program from the recording medium and executes the program. As the processor, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) can be used. Examples of the recording medium include “temporary tangible medium” such as a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, and a programmable logic circuit. Further, a RAM (Random Access Memory) for expanding the program may be further provided. Further, the program may be supplied to the computer via an arbitrary transmission medium (a communication network, a broadcast wave, or the like) capable of transmitting the program. Note that one embodiment of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the above-described program is embodied by electronic transmission.

  The present invention is not limited to the embodiments described above, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention.

1 Judgment device 104 Decrease degree identification unit 105 Executability judgment unit 122 Teacher data generation unit

Claims (5)

Using a learned model obtained by machine learning the correlation between the degree of decrease in the remaining battery level in work performed using a work vehicle driven by a battery and the state of the battery and the content of the work, A reduction degree specifying unit that specifies the reduction degree of the battery remaining amount in the scheduled work to be performed using the work vehicle,
Based on the degree of decrease identified by the degree of decrease identification unit, an execution availability determination unit that determines whether the scheduled work using the work vehicle is executable,
A determination device comprising: The execution availability determination unit, when it is determined that the scheduled work is not executable, causes the processing related to the replacement or charging of the battery,
The determination device according to claim 1, wherein: The learned model indicates a correlation between the degree of decrease in the remaining battery level in the work performed using the work vehicle, the state of the battery, the content of the work, and the driver of the work vehicle. A machine-learned model,
Using the learned model, the decrease degree identifying unit identifies the degree of decrease in the remaining battery level in the scheduled work according to the driver of the work vehicle,
The determination device according to claim 1 or 2, wherein: When the scheduled work using the work vehicle is performed, the degree of decrease in the remaining battery level in the scheduled work corresponds to the state of the battery immediately before the scheduled work and the content of the scheduled work. In addition, a teacher data generation unit that generates teacher data used for updating the learned model is provided.
The determination device according to claim 1, wherein: A determination method by a determination device,
Using a learned model obtained by machine learning the correlation between the degree of decrease in the remaining battery level in work performed using a work vehicle driven by a battery and the state of the battery and the content of the work, A degree-of-reduction specifying step of specifying the degree of reduction of the remaining battery level in the scheduled work to be performed using the work vehicle;
A determination step of determining whether or not the scheduled work using the work vehicle can be performed, based on the degree of decrease specified in the step of specifying the degree of decrease.

Images