Estimation device, estimation method, and estimation program

Abstract

To provide an estimation device, estimation method and estimation program which estimate a position of a caution plate attached to a vehicle even for the vehicle in which a database does not store position information of the caution plate.SOLUTION: An estimation device includes: a storage device which stores data associating attribute information including attributes of multiple vehicles with position information about the attachment position of a caution plate in each vehicle; and a processing device which executes acquisition processing of acquiring the attribute information of the target vehicle, estimation processing of estimating the attachment position of the caution plate in the target vehicle on the basis of the attribute information of the target vehicle and the attribute information of each vehicle included in the data, and output processing of outputting the estimated attachment position of the caution plate in the target vehicle.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an estimation device, an estimation method, and an estimation program. [Background technology]

[0002] In the automobile recycling process, when dismantling an automobile and recovering its parts, the work can be made more efficient by narrowing down the parts that are valuable at the time of sale, or the parts that are in low stock and that the operator wants to actively recover. Therefore, when dismantling an automobile, immediately grasping the information about the vehicle to be dismantled is an important process for improving the efficiency of the work. One of the things necessary for grasping the information about the vehicle is a plate called a caution plate, which contains information specific to the vehicle. By grasping the information written on the caution plate, the dismantling work and the information management in the sale of the dismantled parts can be made easier. However, the attachment position of the caution plate varies depending on the manufacturer, the model, the year of manufacture, or the model. Furthermore, the caution plate may be attached to a position that is difficult to confirm by visual inspection alone, such as the back of the seat. Therefore, if a worker visually searches for the position of the caution plate, the search takes a lot of work time, and the work load becomes heavy.

[0003] In response to such issues, Patent Document 1 discloses that in the pre-processing of each process in an automobile recycling system, the position of the caution plate, which lists information such as the model year, which varies depending on the vehicle model, is indicated from a database. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2003-44573 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology described in Patent Document 1 has a problem that the position of the caution plate cannot be specified for vehicle models that are not stored in the database in advance. Therefore, when dismantling a vehicle model that is not stored in the database, a worker must visually search for the position of the caution plate inside the vehicle, which takes up a lot of work time. While the position information of the caution plate is made public for vehicles that are widely distributed, it is difficult to obtain the position information of the caution plate for vehicles that are old or have a low distribution number, such as luxury cars. Considering the current situation in which vehicle models are diversifying, it is expected that rare vehicle models will increase in the future, and there will be more opportunities to dismantle vehicles that are not stored in the database.

[0006] The present invention has been made in consideration of the above circumstances, and its purpose is to provide an estimation device, an estimation method, and an estimation program that are capable of estimating the position of a caution plate attached to a vehicle, even if the position information of the caution plate is not stored in the database. [Means for solving the problem]

[0007] One aspect of the present invention for solving the above problem is an estimation device that includes a storage device that stores data that corresponds attribute information including attributes of each of a plurality of vehicles with positional information regarding the installation position of a caution plate on each vehicle, and a processing device that executes an acquisition process that acquires attribute information of a target vehicle, an estimation process that estimates the installation position of a caution plate on the target vehicle based on the attribute information of the target vehicle and the attribute information of each vehicle included in the data, and an output process that outputs the estimated installation position of the caution plate on the target vehicle. Effect of the Invention

[0008] According to the present invention, even if the position information of a caution plate of a vehicle is not stored in the database, the position of the caution plate attached to the vehicle can be estimated. Configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0009] [Figure 1] FIG. 2 is a diagram illustrating an example of the configuration of an estimation device according to the present embodiment. [Diagram 2] FIG. 11 is a diagram illustrating an example of a shape information table. [Diagram 3] FIG. 13 is a diagram illustrating an example of a correspondence table. [Figure 4] FIG. 11 is a diagram illustrating an example of a position information table. [Diagram 5] FIG. 4 is a diagram showing an example of an installation position of a caution plate. [Figure 6] FIG. 2 is a process flow diagram illustrating an overview of the process performed by the estimation device. [Figure 7] FIG. 13 is a diagram showing an example of an attribute information input screen. [Figure 8] FIG. 11 is a process flow diagram illustrating details of the estimation process. [Figure 9] FIG. 13 is a diagram showing an example of a position candidate display screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. The following description and drawings are examples for explaining the present invention, and appropriate omissions and simplifications are made for clarity of explanation. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings. In the following explanation, various information may be explained using expressions such as "table," "list," and "queue," but the various information may be expressed in other data structures. To indicate independence of data structure, "XX table," "XX list," and the like may be referred to as "XX information." When explaining identification information, expressions such as "identification information," "identifier," "name," "ID," and "number" are used, but these are interchangeable. When there are multiple components having the same or similar functions, they may be described by using the same reference numerals with different subscripts. However, when there is no need to distinguish between these multiple components, the subscripts may be omitted. In addition, in the following description, the processing performed by executing a program may be described, but the program is executed by a processor (e.g., CPU, GPU) to perform a defined process using storage resources (e.g., memory) and / or interface devices (e.g., communication ports) as appropriate, so the subject of the processing may be the processor. Similarly, the subject of the processing performed by executing a program may be a controller, device, system, computer, or node having a processor. The subject of the processing performed by executing a program may be a calculation unit, and may include a dedicated circuit (e.g., FPGA or ASIC) that performs a specific process. A program may be installed in a device such as a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server may include a processor and a storage resource for storing a program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. In the following description, two or more programs may be realized as one program, and one program may be realized as two or more programs.

[0011] <System configuration> Fig. 1 is a diagram showing an example of the configuration of an estimation device 1 in this embodiment. The estimation device 1 is an information processing device that estimates the position of a caution plate attached to a vehicle (hereinafter referred to as a "target vehicle") that is to be dismantled in, for example, an automobile recycling process. A caution plate is a metal plate on which information specific to each vehicle is printed. For example, the vehicle manufacturer, chassis number, body color number, etc. are printed on the caution plate.

[0012] The estimation device 1 includes a storage device 11 such as an HDD (Hard Disk Drive) or SSD (Solid State Drive), a processing device 12 consisting of a CPU (Central Processing Unit) and memories such as RAM (Random Access Memory) and ROM (Read Only Memory), an input device 13 such as a keyboard, mouse, or touch panel, an output device 14 such as a display or printer, and a communication device 15 consisting of a NIC (Network Interface Card), a wireless communication module, a USB (Universal Serial Interface) module, or a serial communication module, etc.

[0013] The storage device 11 stores a shape information table 200, a correspondence table 300, and a position information table 400. The shape information table 200 is a database that stores information about the shape of each vehicle. The correspondence table 300 is a database that associates attribute information of each vehicle with the position of a caution plate attached to each vehicle. The position information table 400 is a database that stores information about the position of the caution plate.

[0014] (Shape information table) FIG. 2 is a diagram showing an example of a shape information table 200. The shape information table 200 has data items of a shape information ID 201 in which a shape information ID for identifying each shape of the vehicle is set, a shape image 202 in which a file name or address of image information of the shape is set, and a shape name 203 in which a name of the shape is set. The shape image information is, for example, data of an image of a vehicle. In the example shown in the figure, one piece of image information is set in the shape image 202 for one record, but this is not limited thereto, and multiple pieces of image information may be set. For example, multiple pieces of image information obtained by capturing images of the vehicle from different directions, such as the front, side, back, top, and bottom of the vehicle, may be set in the shape image 202.

[0015] The shape of the vehicle in this embodiment refers to the external shape of the entire vehicle. For example, the types of vehicle shapes include a sedan, a station wagon, a light passenger car, a minivan, a compact car, a sports car, a coupe, an SUV (Sport Utility Vehicle), an open car, a station wagon, a commercial vehicle, etc.

[0016] A sedan is characterized by having an engine room, a passenger compartment, and a luggage space that are all separate. A station wagon is a variation of a sedan, characterized by a longer passenger compartment, a door at the rear, and a luggage compartment at the rear. The shape information table 200 may include such information.

[0017] A light passenger car is a vehicle that satisfies the various conditions regarding total displacement, overall length, overall width, overall height, and passenger capacity, and is characterized by a relatively small body. A minivan has an engine room that is separate from the interior space (passenger space and luggage space), and the passenger space and luggage space are integrated. A minivan is characterized by having three rows of seats in the interior space, and the seats can be arranged according to the purpose by sliding or reclining. The shape information table 200 may include such information.

[0018] A compact car is a vehicle with a relatively small body and a small displacement engine. A compact car is a two-box type in which the engine room is separate from the interior space, and the passenger space and the luggage space are integrated. The shape information table 200 may include such information.

[0019] Sports cars are manufactured with a primary focus on driving performance, and are characterized by the use of engines that produce large power and engines that run smoothly even at high revolutions. Coupes are a type of two-door passenger car, and are characterized by a low roof and sporty design. Coupes often have a low vehicle height, a wide wheelbase, and wide tread to give a sense of stability. The shape information table 200 may include such information.

[0020] An SUV is a two-box type vehicle in which the engine room is separate from the interior space, and the passenger space and luggage space are integrated. SUVs are characterized by a high ground clearance and a relatively large body. The shape information table 200 may include such information.

[0021] A convertible car is characterized by having no roof. Convertible cars generally have a retractable hardtop with a removable roof or a folding soft top. A station wagon is a vehicle with a large luggage compartment at the rear of the vehicle, and is characterized by being able to carry more luggage than vehicles such as sedans and hatchbacks. A commercial vehicle is a vehicle designed primarily for carrying luggage, and is characterized by being designed to accommodate a larger amount of luggage and to be easy to load. The shape information table 200 may include such information.

[0022] In this embodiment, the shape of the vehicle refers to the overall shape and external appearance of the vehicle, but is not limited thereto and may refer to the shape of a portion of the vehicle or the shape of the interior.

[0023] (Corresponding table) 3 is a diagram showing an example of the correspondence table 300. The correspondence table 300 has data items of No. 301 in which the record number is set, CP position ID 302 in which the CP position ID for identifying the position of the caution plate is set, vehicle model 303 in which the vehicle model is set, era 304 in which the vehicle was manufactured or sold is set, manufacturer name 305 in which the name of the vehicle manufacturer is set, shape 306 in which the shape name of the vehicle is set, displacement 307 in which the vehicle displacement is set, vehicle use country 308 in which the country in which the vehicle was used is set, manufacturing country 309 in which the country in which the vehicle was manufactured is set, manufacturing plant 310 in which the factory in which the vehicle was manufactured is set, and sibling vehicle model 311 in which the vehicle's sibling vehicle model is set. Information set in each data item of the vehicle model 303, era 304, manufacturer name 305, shape 306, displacement 307, vehicle use country 308, manufacturing country 309, manufacturing plant 310, and sibling vehicle model 311 corresponds to vehicle attribute information.

[0024] The country of manufacture 309 may be the country where a part of the vehicle is manufactured, the country where the vehicle is finally assembled, or the country where the vehicle manufacturer is based. Here, there is a high possibility that there is a correlation between the design policy at the time of manufacturing the vehicle and the installation position of the caution plate. In addition, there is a correlation between the design policy for manufacturing and the country of manufacture. For example, the design (shape) or function of the vehicle is greatly influenced by the environment of the country where the manufacturer is located. For example, each evaluation element such as safety, driving stability, practicality, etc. may differ depending on the country of manufacture, and such evaluation information may be included in the attribute information.

[0025] The manufacturing plant 310 may be a plant where a part of the vehicle is manufactured, or may be a plant where the vehicle is finally assembled. Here, there is a high possibility that the manufacturing process of the vehicle and the installation position of the caution plate are correlated. In addition, it is considered that there is a correlation between the manufacturing process of the vehicle and the manufacturing plant. Therefore, for example, by setting the plant where the vehicle is finally assembled as the manufacturing plant 310, the manufacturing process of the vehicle at that plant can be reflected in the attribute "manufacturing plant".

[0026] A sibling car is a vehicle that is sold as a different product with a different product name or appearance. Sibling cars are almost identical in basic structure, but are sold as different models with differences in appearance such as body color variations, bumpers, front grilles, lighting shapes, body exterior panels, emblems, etc. Sibling cars are likely to have similar shapes and manufacturing methods, so the locations of their caution plates will also be similar.

[0027] (Location information table) 4 is a diagram showing an example of a position information table 400. The position information table 400 has the following data items: CP position ID 401 in which a CP position ID is set, position name 402 in which a name of a position is set, and position image 403 in which a file name or address of image information for guiding a worker to the position is set.

[0028] FIG. 5 is a diagram showing an example of the installation positions of the caution plate. In the illustrated example, the following positions are shown as examples of the installation positions of the caution plate: (1) driver's seat side center pillar, (2) upper left of the engine room, (3) upper center of the engine room, (4) upper right of the engine room, (5) passenger seat side front pillar, (6) passenger seat side center pillar, and (7) under the driver's seat. The installation positions of the caution plate vary depending on the manufacturer, model, year of manufacture, model type, etc. Also, as shown in the figure, the caution plate may be installed in a position that is difficult to confirm by visual inspection alone.

[0029] The storage device 11 also stores in advance programs for the processing device 12 to realize each of the functions of the information acquisition unit 120, the shape estimation unit 123, the position estimation unit 124, and the output unit 125. The processing device 12 realizes each of the functions of the information acquisition unit 120, the shape estimation unit 123, the position estimation unit 124, and the output unit 125 by reading and executing the programs from the storage device 11.

[0030] The information acquisition unit 120 executes an acquisition process to acquire attribute information of the target vehicle. The information acquisition unit 120 includes a vehicle information acquisition unit 121 and an image information acquisition unit 122. The vehicle information acquisition unit 121 displays an attribute information input screen for accepting input of attribute information of the target vehicle on the output device 14, and accepts input of the attribute information of the target vehicle from a user on the input device 13. Details of the attribute information input screen will be described later. The image information acquisition unit 122 acquires one or more pieces of image information of the target vehicle. For example, the image information acquisition unit 122 receives image information (image data) of the target vehicle captured from a predetermined direction from an imaging device such as a camera. The image information acquisition unit 122 may acquire a plurality of pieces of image information of the target vehicle captured from a plurality of different directions, such as the front, side, back, top, and bottom of the target vehicle.

[0031] The shape estimation unit 123 estimates the shape of the target vehicle by comparing the image information acquired by the image information acquisition unit 122 with the image information set in the shape image 202 of the shape information table 200. Note that in this embodiment, the shape estimation unit 123 estimates and acquires the shape of the target vehicle from the image information of the target vehicle, but this is not limiting, and other attribute information (model, year, manufacturer name, shape, engine displacement, country where the vehicle is used, country of manufacture, manufacturing plant, or sibling model) may be acquired from the image information of the target vehicle.

[0032] The position estimation unit 124 performs an estimation process to estimate the position of the caution plate attached to the target vehicle based on the attribute information of the target vehicle acquired by the vehicle information acquisition unit 121, the shape of the target vehicle estimated by the shape estimation unit 123, and the attribute information of each vehicle stored in the correspondence table 300.

[0033] The output unit 125 executes an output process to output information on the mounting position of the caution plate in the target vehicle estimated by the position estimation unit 124. For example, the output unit 125 displays a position candidate display screen on the output device 14 to guide the user to the mounting position of the caution plate. Details of the position candidate display screen will be described later. Alternatively, the output unit 125 may print the information on the mounting position of the caution plate on a paper medium or the like by the output device 14. Alternatively, the output unit 125 may transmit the information on the mounting position of the caution plate to a terminal used by a dismantling worker via the communication device 15 and cause the terminal to display the information on the mounting position of the caution plate. The terminal used by the dismantling worker is, for example, an information processing device such as a personal computer, a tablet terminal, or a smartphone.

[0034] A program for realizing each function of the information acquisition unit 120, the shape estimation unit 123, the position estimation unit 124, and the output unit 125 can be recorded on, for example, a portable or fixed recording medium and distributed. Next, the processing performed by the estimation device 1 will be described.

[0035] <Processing Overview> 6 is a process flow diagram outlining the process performed by the inference device 1. The process shown in the diagram is executed, for example, when a predetermined input is made to the inference device 1 by a user, or at a predetermined timing (for example, at a predetermined time or at a predetermined time interval).

[0036] First, the vehicle information acquisition unit 121 displays an attribute information input screen on the output device 14 and accepts input of attribute information from the user (S100).

[0037] (Attribute information input screen) 7 is a diagram showing an example of an attribute information input screen 500. The attribute information input screen 500 includes an area 501 for receiving an input of the model of the target vehicle from the user, an area 502 for receiving an input of the year of the target vehicle from the user, an area 503 for receiving an input of the manufacturer name of the target vehicle from the user, an area 504 for receiving an input of the engine displacement of the target vehicle from the user, an area 505 for receiving an input of the country of use of the target vehicle from the user, an area 506 for receiving an input of the country of manufacture of the target vehicle from the user, an area 507 for receiving an input of the manufacturing plant of the target vehicle from the user, and an area 508 for receiving an input of sibling models of the target vehicle from the user.

[0038] Next, the image information acquisition unit 122 acquires image information of the target vehicle (S200). At this time, the image information acquisition unit 122 may acquire a plurality of pieces of image information obtained by capturing images of the target vehicle from a plurality of different directions (for example, image information obtained by capturing an image of the target vehicle from the front, image information obtained by capturing an image of the target vehicle from the side, and image information obtained by capturing an image of the target vehicle from the back).

[0039] Next, the shape estimation unit 123 estimates the shape of the target vehicle by comparing the image information of the target vehicle with the image information set in the shape image 202 of the shape information table 200 (S300). For example, the shape estimation unit 123 uses the image information set in the shape image 202 of the shape information table 200 as a template image, and performs image processing such as pattern matching on the image information to extract a shape that is the same as or similar to the target vehicle from the shape information table 200.

[0040] Next, the shape prediction unit 123 judges whether or not prediction of the shape of the target vehicle in S300 has been successful (S400). For example, when a shape identical or similar to that of the target vehicle can be extracted from the shape information table 200 as a result of image processing, the shape prediction unit 123 judges that prediction of the shape has been successful, and when a shape identical or similar to that of the target vehicle cannot be extracted from the shape information table 200, the shape prediction unit 123 judges that prediction of the shape has failed.

[0041] If the vehicle information acquisition unit 121 fails to estimate the shape of the target vehicle (S400: No), it displays a screen for accepting input of information regarding the shape of the target vehicle and accepts input of the shape of the target vehicle from the user (S500). Then, it proceeds to the process of S600.

[0042] On the other hand, if the prediction of the shape of the target vehicle is successful (S400: Yes), or following the processing of S500, the position prediction unit 124 determines whether or not the position information (CP position ID) of the caution plate of the target vehicle is stored in the correspondence table 300 (S600). Specifically, the position prediction unit 124 determines whether or not there is a record in the correspondence table 300 whose vehicle type 303 matches that of the target vehicle.

[0043] If the position information of the caution plate of the target vehicle is stored in the correspondence table 300 (S600: Yes), the position estimation unit 124 reads out the CP position ID 302 of the record of the vehicle type 303 of the target vehicle, proceeds to processing of S800, and outputs the position information indicated by the read CP position ID 302 to the output unit 125.

[0044] On the other hand, if the position information of the caution plate of the target vehicle is not stored in the correspondence table 300 (S600: No), the position estimation unit 124 executes an estimation process to estimate the position of the caution plate based on the attribute information of the target vehicle and the attribute information of each vehicle stored in the correspondence table 300 (S700).

[0045] <Inference processing> FIG. 8 is a process flow diagram illustrating the details of the inference process S700.

[0046] First, the position estimation unit 124 reads one record from the correspondence table 300 (S701).

[0047] Next, the location estimation unit 124 compares the attribute information of the target vehicle with the attribute information of the read record, and calculates the similarity therebetween (S702). Similarity is the degree of similarity between the values ​​of attributes. First, the location estimation unit 124 calculates the similarity between the attribute of the target vehicle and the attribute of the record for each attribute. Specifically, for the attributes "manufacturer name", "shape", "vehicle use country", "manufacturing country", and "manufacturing country", the location estimation unit 124 increases the similarity when the attribute values ​​match, and decreases the similarity when the attribute values ​​do not match. On the other hand, for the attributes "year" and "engine displacement", the location estimation unit 124 increases the similarity the smaller the difference in the attribute values, and decreases the similarity the larger the difference in the attribute values. In addition, with regard to the attribute "sibling model," the position estimation unit 124 increases the similarity if the attribute values ​​match as a result of comparing the model of the target vehicle with the sibling model of the read record, or as a result of comparing the sibling model of the target vehicle with the model or sibling model of the read record, and decreases the similarity if the attribute values ​​do not match.

[0048] Next, the position estimation unit 124 calculates the similarity of the overall attributes of the vehicle based on the similarity of each attribute. Here, the position estimation unit 124 may weight each attribute when calculating the similarity of the overall attributes. For example, the position estimation unit 124 may weight the shape or sibling car model more heavily than the other attributes. The weight of each attribute is set in advance by the user.

[0049] Next, the position estimation unit 124 determines whether or not the attribute information of all records in the correspondence table 300 has been compared with the attribute information of the target vehicle (S703). That is, the position estimation unit 124 determines whether or not the similarity has been calculated for all records in the correspondence table 300. If the comparison has not been made with all records (S703: No), the process returns to S701 and the next record is obtained from the correspondence table 300.

[0050] On the other hand, when the position estimation unit 124 compares with all records in the correspondence table 300 (S703: Yes), it extracts all CP position IDs of records whose similarity satisfies a predetermined condition (S704). The predetermined condition is, for example, that the similarity is higher than a predetermined reference value, or that the similarity is the highest, etc. The reference value for similarity is set in advance by the user.

[0051] Next, the position estimation unit 124 calculates an evaluation value for each of the acquired CP position IDs (S705). Specifically, the position estimation unit 124 calculates an evaluation value for each CP position ID by tallying up the number of extractions for each CP position ID. For example, the position estimation unit 124 sets the total number of CP position IDs "1" extracted in S704 as the evaluation value for the CP position ID "1". In other words, the position estimation unit 124 sets the number of records whose CP position ID 302 is "1" among records whose similarity satisfies a predetermined condition as the evaluation value for the CP position ID "1". Note that the method of calculating the evaluation value is not limited to this, and a method of calculating an evaluation value based on the type of attribute with high similarity may also be used. For example, the position estimation unit 124 may calculate an evaluation value such that the evaluation value of the CP position ID of a record with high similarity in the attribute "brother car model" is the highest.

[0052] Next, the position estimation unit 124 outputs one or more CP position IDs with high evaluation values ​​to the output unit 125 together with their evaluation values ​​as candidates for the position of the caution plate of the target vehicle (S706). A CP position ID with a high evaluation value may be a CP position ID whose evaluation value is higher than a predetermined reference value, or may be a predetermined number of CP position IDs (e.g., the top two) with the highest evaluation values. Thereafter, the position estimation unit 124 ends this estimation process.

[0053] The estimation process executed by the position estimation unit 124 is not limited to the above-mentioned flow, and may be any process that estimates the position of the caution plate of the target vehicle based on the attribute information of the target vehicle. For example, the position estimation unit 124 may estimate the position of the caution plate of the target vehicle using a trained model constructed by machine learning the data stored in the correspondence table 300 as training data. Specifically, the position estimation unit 124 generates a trained model by using the attributes of each of the multiple vehicles as training data of input data and the position information on the mounting position of the caution plate of each of the multiple vehicles as training data of output data. The trained model is constructed based on the algorithms of, for example, a neural network, a decision tree, a random forest, and a support vector machine (SVM). The neural network is a neural network having an input layer that receives input data, one or more intermediate layers (hidden layers) that extract and output feature amounts from the input data, and an output layer that outputs output data, and is, for example, a convolution neural network (CNN). For example, the trained model uses attribute information as input data and a CP position ID as output data. Therefore, the position estimation unit 124 estimates the position of the caution plate of the target vehicle by inputting the attribute information of the target vehicle into the learned model and obtaining the output data.

[0054] Next, the output unit 125 outputs information on the position of the caution plate estimated by the position estimation unit 124 (S800). For example, the output unit 125 displays on the output device 14 a position candidate display screen that guides the user to the estimated position of the caution plate.

[0055] (Position candidate display screen) FIG. 9 is a diagram showing an example of a position candidate display screen 600. The position candidate display screen 600 has a display area for displaying the estimated position of the caution plate as a candidate for the position of the caution plate in the target vehicle. In the illustrated example, the position candidate display screen 600 has a first display area 610 for displaying a first candidate and a second display area 620 for displaying a second candidate. In the first display area 610, a position name 611 of the first candidate and a position image 612 of the first candidate are displayed. In the second display area 620, a position name 621 of the second candidate and a position image 622 of the second candidate are displayed. The output unit 125 reads out the position name 402 and the position image 403 corresponding to each of one or more CP position IDs 401 input from the position estimation unit 124 from the position information table 400, and displays the read position name 402 and the position image 403 in each display area of ​​the position candidate display screen 600.

[0056] At this time, the output unit 125 displays candidates (CP position IDs) with higher evaluation values ​​preferentially over candidates with lower evaluation values. "Displaying with higher evaluation values ​​preferentially" includes displaying candidates in order from the highest evaluation value, or displaying candidates with higher evaluation values ​​in an emphasized manner. "Displaying with emphasis" includes changing the display color, changing the character size, changing the background color, or adding a mark or the like. In this embodiment, the output unit 125 displays candidates in order from the highest evaluation value. For example, the output unit 125 displays the candidate with the highest evaluation value (position name "driver's seat side center pillar") in the first display area 610, and displays the candidate with the next highest evaluation value (position name "engine room inside") in the second display area 620.

[0057] In addition, by displaying position images 612, 622 that guide the mounting position together with position names 611, 621 of candidate positions for the caution plate, the worker can easily identify the candidate positions where the caution plate is mounted. In the illustrated example, the position images 612, 622 that guide the mounting position indicate the candidate mounting positions with arrows 6121, 6221.

[0058] In the illustrated example, the output unit 125 displays two candidates for the location of the caution plate of the target vehicle on the location candidate display screen 600, but this is not limited to this, and the number of candidates displayed may be one, or three or more.

[0059] Furthermore, when performing the estimation process using a trained model, the estimation device 1 may receive an input of the actual position of the caution plate of the target vehicle from the user on the position candidate display screen 600, and use the input position and attribute information of the target vehicle as learning data for the trained model. Alternatively, the estimation device 1 may associate the input position with the attribute information of the target vehicle, and add it to the correspondence table 300 as a new record.

[0060] As described above, the estimation device 1 of this embodiment stores data (correspondence table 300) that associates attribute information including the attributes of each of a plurality of vehicles with position information (CP position ID) regarding the mounting position of the caution plate on each vehicle, and executes an acquisition process that acquires the attribute information of a target vehicle, an estimation process that estimates the mounting position of the caution plate on the target vehicle based on the attribute information of the target vehicle and the attribute information of each vehicle included in the correspondence table 300, and an output process that outputs the estimated mounting position of the caution plate on the target vehicle.

[0061] That is, the estimation device 1 of the present embodiment can estimate the position of the caution plate attached to the target vehicle even if the position information of the caution plate of the target vehicle is not stored in the database. This saves the worker the trouble of visually searching for the position of the caution plate, and reduces the number of man-hours of the worker.

[0062] The attribute information also includes the shape of the vehicle, and the estimation device 1 of this embodiment estimates the attachment position of the caution plate on the target vehicle based on the shape of the target vehicle and the shapes of each vehicle included in the correspondence table 300.

[0063] The general rule for caution plates is that they should be installed in places that are visible without dismantling the vehicle, and the shape of the vehicle affects where the caution plates are installed. For example, the positions where a caution plate can be installed, where it is easy to install, where it is difficult to install, or where it is impossible to install, vary depending on the shape of the vehicle. Therefore, by estimating the installation position of the caution plate based on the shape of the vehicle, it is possible to more accurately estimate the installation position of the caution plate for the target vehicle.

[0064] In addition, the attribute information includes the country of use of the vehicle, and the estimation device 1 of this embodiment estimates the installation position of the caution plate on the target vehicle based on the country of use of the target vehicle and the countries of use of each vehicle included in the correspondence table 300.

[0065] As described above, there is a high possibility that the shape of a vehicle and the mounting position of a caution plate are correlated. In addition, since the shape of a vehicle depends on the road laws of each country, it is considered that there is also a correlation between the country of use and the mounting position of the caution plate. Therefore, by estimating the mounting position of the caution plate based on the country of use of the vehicle, it is possible to more accurately estimate the mounting position of the caution plate of the target vehicle.

[0066] In addition, the attribute information includes the vehicle's manufacturing plant, and the estimation device 1 of this embodiment estimates the installation position of the caution plate on the target vehicle based on the manufacturing plant of the target vehicle and the manufacturing plants of each vehicle included in the correspondence table 300.

[0067] There is a high possibility that the vehicle manufacturing process and the conditions for installing the caution plate are correlated. For example, it is considered that the installation position of the caution plate will differ depending on which step in the manufacturing process the caution plate is installed at. It is also considered that there is a correlation between the manufacturing process and the vehicle manufacturing plant. Therefore, by estimating the installation position of the caution plate based on the vehicle manufacturing plant, it is possible to more accurately estimate the installation position of the caution plate of the target vehicle.

[0068] In addition, the attribute information includes the country of manufacture of the vehicle, and the estimation device 1 of this embodiment estimates the installation position of the caution plate on the target vehicle based on the country of manufacture of the target vehicle and the country of manufacture of each vehicle included in the correspondence table 300.

[0069] There is a high possibility that the design policy at the time of vehicle manufacture and the installation position of the caution plate are correlated. Furthermore, it is thought that there is also a correlation between the manufacturing design policy and the country of manufacture. This is because the design (shape) or function of a vehicle is greatly influenced by the environment of the country of manufacture (e.g., the country where the manufacturer is located). Therefore, by predicting the installation position of the caution plate based on the country of manufacture of the vehicle, it is possible to more accurately predict the installation position of the caution plate on the target vehicle.

[0070] In addition, the attribute information includes the model of the vehicle's sibling vehicles, and the estimation device 1 of this embodiment estimates the installation position of the caution plate on the target vehicle based on the model of the target vehicle or the model of its sibling vehicles and the model of each vehicle or the model of its sibling vehicles included in the correspondence table 300.

[0071] It is considered that sibling vehicles will have similar shapes and manufacturing methods. Therefore, it is considered that there is a correlation between the mounting positions of the caution plates between sibling vehicle models. For example, it is highly likely that the mounting positions of the caution plates are the same between sibling vehicles. Therefore, by estimating the mounting position of the caution plate based on the vehicle's sibling vehicle model, it is possible to more accurately estimate the mounting position of the caution plate of the target vehicle.

[0072] In addition, the attribute information includes the vehicle's engine displacement, and the estimation device 1 of this embodiment estimates the installation position of the caution plate on the target vehicle based on the engine displacement of the target vehicle and the engine displacement of each vehicle included in the correspondence table 300.

[0073] It is considered that vehicle models with similar engine displacement will have similar shapes and manufacturing methods. As described above, the shape or manufacturing method of a vehicle is highly likely to correlate with the mounting position of a caution plate. For this reason, it is considered that engine displacement also correlates with the mounting position of a caution plate. Therefore, by estimating the mounting position of a caution plate based on the engine displacement of a vehicle, it is possible to more accurately estimate the mounting position of the caution plate of a target vehicle.

[0074] Moreover, the estimation device 1 of the present embodiment estimates a plurality of mounting positions, and outputs the estimated plurality of mounting positions as candidates for the mounting position of the caution plate on the target vehicle.

[0075] With this configuration, multiple candidates for the mounting position of the caution plate of the target vehicle are output. As a result, even if one candidate is different from the actual position of the caution plate, the worker can search for the caution plate using the other candidates as clues. Therefore, the workload of the worker can be reduced compared to when searching without clues.

[0076] Moreover, the estimation device 1 of this embodiment acquires attribute information of the target vehicle from image information obtained by capturing an image of the target vehicle.

[0077] With this configuration, an operator can input attribute information to the estimation device 1 simply by capturing an image of the target vehicle. That is, the operator does not need to manually input the attributes of the target vehicle. Therefore, even if the operator does not know the attribute information of the target vehicle, the estimation device 1 can acquire the attribute information and estimate the position of the caution plate of the target vehicle.

[0078] In addition, the attribute information includes multiple attributes, and the estimation device 1 of this embodiment calculates the similarity between the attributes of the target vehicle and the attributes of each vehicle included in the correspondence table 300 for each attribute, and infers the location information of a vehicle for which the calculated similarity satisfies a predetermined condition as the installation location of the caution plate on the target vehicle.

[0079] With this configuration, the position of the caution plate of a vehicle with attributes similar to those of the target vehicle is estimated as the position of the caution plate of the target vehicle. Since vehicles with similar attributes are considered to have the same or similar positions of the caution plates, the mounting position of the caution plate of the target vehicle can be estimated with high accuracy.

[0080] In addition, the estimation device 1 of this embodiment calculates the similarity based on weights preset for each of the multiple attributes, and estimates the location information of a vehicle for which the calculated similarity satisfies a specified condition as the installation location of a caution plate on the target vehicle.

[0081] With this configuration, for example, the weight of an attribute that is more correlated with the mounting position of the caution plate can be increased more than the weight of other attributes to calculate the similarity of the attributes, thereby making it possible to more accurately estimate the mounting position of the caution plate on the target vehicle.

[0082] In addition, the estimation device 1 of this embodiment generates a model by using the attributes of each of the multiple vehicles as learning data for input data, and using position information regarding the installation position of the caution plate on each of the multiple vehicles as learning data for output data, and estimates the installation position of the caution plate on the target vehicle by inputting attribute information of the target vehicle into the generated model.

[0083] In this way, by using a model generated using the attributes of each of multiple vehicles as learning data for input data and positional information regarding the installation position of the caution plate on each of multiple vehicles as learning data for output data, it is possible to more accurately predict the installation position of the caution plate on the target vehicle.

[0084] The present invention is not limited to the above-described embodiment, and can be implemented using any components without departing from the scope of the present invention. The above-described embodiments and modifications are merely examples, and the present invention is not limited to these contents as long as the features of the invention are not impaired. In addition, although various embodiments and modifications have been described above, the present invention is not limited to these contents. Other aspects conceivable within the scope of the technical idea of ​​the present invention are also included in the scope of the present invention.

[0085] For example, part of the hardware included in each device of this embodiment may be provided in another device.

[0086] Furthermore, each program of the estimation device 1 may be provided in another device, a certain program may be made up of multiple programs, or multiple programs may be integrated into one program.

[0087] In addition, in this embodiment, the estimation device 1 acquires the shape of the vehicle from image information capturing an image of the target vehicle, but this is not limited to this, and other attribute information (for example, vehicle model, year, manufacturer name, engine displacement, country in which the vehicle is used, country of manufacture, manufacturing structure, or sibling vehicle models, etc.) may be inferred and acquired from the image information capturing an image of the target vehicle. [Explanation of symbols]

[0088] 1 Estimation device, 11 Storage device, 12 Processing device, 120 Information acquisition unit, 121 Vehicle information acquisition unit, 122 Image information acquisition unit, 123 Shape estimation unit, 124 Position estimation unit, 125 Output unit, 13 Input device, 14 Output device, 15 Communication device, 200 Shape information table, 300 Correspondence table, 400 Position information table, 500 Attribute information input screen, 600 Position candidate display screen

Claims

1. a storage device that stores data correlating attribute information including attributes of each of a plurality of vehicles with position information relating to the mounting position of a caution plate on each vehicle; An acquisition process for acquiring attribute information of a target vehicle; an estimation process for estimating a mounting position of a caution plate on the target vehicle based on attribute information of the target vehicle and attribute information of each vehicle included in the data; an output process for outputting the estimated mounting position of the caution plate in the target vehicle; The estimation device includes a processor that executes the above steps.

2. The attribute information includes a shape of a vehicle, The processing device includes: In the estimation process, an attachment position of a caution plate on the target vehicle is estimated based on a shape of the target vehicle and shapes of each vehicle included in the data. The estimation device according to claim 1.

3. The attribute information includes a country of use of the vehicle; The processing device includes: In the estimation process, an installation position of a caution plate on the target vehicle is estimated based on a country of use of the target vehicle and a country of use of each vehicle included in the data. The estimation device according to claim 1.

4. The attribute information includes a manufacturing plant of the vehicle; The processing device includes: In the estimation process, an attachment position of a caution plate on the target vehicle is estimated based on a manufacturing plant of the target vehicle and a manufacturing plant of each vehicle included in the data. The estimation device according to claim 1.

5. The attribute information includes a country of manufacture of the vehicle; The processing device includes: In the estimation process, an installation position of a caution plate on the target vehicle is estimated based on a country of manufacture of the target vehicle and a country of manufacture of each vehicle included in the data. The estimation device according to claim 1.

6. The attribute information includes a vehicle model of a sibling vehicle of the vehicle, The processing device includes: In the estimation process, an attachment position of a caution plate on the target vehicle is estimated based on the vehicle model of the target vehicle or the vehicle model of its siblings and the vehicle models of each vehicle or its siblings included in the data. The estimation device according to claim 1.

7. The attribute information includes an engine displacement of the vehicle. The processing device includes: In the estimation process, an installation position of a caution plate on the target vehicle is estimated based on the engine displacement of the target vehicle and the engine displacement of each vehicle included in the data. The estimation device according to claim 1.

8. The processing device includes: In the estimation process, a plurality of the attachment positions are estimated, In the output process, the estimated mounting positions are output as candidates for mounting positions of the caution plate in the target vehicle. The estimation device according to claim 1.

9. The processing device includes: In the acquisition process, attribute information of the target vehicle is acquired from a captured image of the target vehicle. The estimation device according to claim 1.

10. The attribute information includes a plurality of attributes, The processing device includes: In the inference process, for each attribute, a similarity between the attribute of the target vehicle and the attribute of each vehicle included in the data is calculated, and the position information of a vehicle for which the calculated similarity satisfies a predetermined condition is inferred to be the mounting position of a caution plate on the target vehicle. The estimation device according to claim 1.

11. The processing device includes: In the inference process, the similarity is calculated based on a weight previously set for each of a plurality of attributes, and the position information of a vehicle for which the calculated similarity satisfies a predetermined condition is inferred to be the mounting position of a caution plate on the target vehicle. The estimation device according to claim 10.

12. The processing device includes: In the estimation process, a model is generated by using attributes of each of a plurality of vehicles as learning data of input data, and position information on the mounting position of the caution plate of each of the plurality of vehicles as learning data of output data, and the attribute information of the target vehicle is input into the generated model to estimate the mounting position of the caution plate on the target vehicle. The estimation device according to claim 1.

13. An information processing device, An acquisition process for acquiring attribute information of a target vehicle; an estimation process for estimating the installation position of the caution plate on the target vehicle based on attribute information of each vehicle included in data correlating attribute information including attributes of each of a plurality of vehicles with position information related to the installation position of the caution plate on each vehicle, and attribute information of the target vehicle; and executing an output process for outputting the estimated installation position of the caution plate in the target vehicle. Guessing method.

14. In the information processing device, An acquisition process for acquiring attribute information of a target vehicle; an estimation process for estimating the installation position of the caution plate on the target vehicle based on attribute information of each vehicle included in data correlating attribute information including attributes of each of a plurality of vehicles with position information related to the installation position of the caution plate on each vehicle, and attribute information of the target vehicle; and outputting the estimated installation position of the caution plate on the target vehicle. Guessing program.

Images