Method for notification of garbage collection time

Abstract

To provide accurate information on a scheduled arrival time of a garbage collection vehicle to a user of a garbage collection point.SOLUTION: A method for notification of a garbage collection time includes the steps of: transmitting by a vehicle 70 an image containing a garbage collection point to a server 10, and identifying by the server 10 an uncollected image that contains an uncollected point, which is a collection point that a garbage collection vehicle 30 has not passed yet in a traveling route of the garbage collection vehicle 30, from a plurality of images acquired from the vehicle 70; calculating an estimated amount of garbage accumulated at the uncollected point on the basis of the identified uncollected image; when the estimated amount of garbage is large, calculating a scheduled arrival time at a collection point that the garbage collection vehicle will pass after the uncollected point in the traveling route so that the scheduled arrival time at the collection point will be later than when the estimated amount of garbage is small; and notifying a user terminal 50 of the scheduled arrival time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for notifying garbage collection times. [Background technology]

[0002] The garbage collection truck disclosed in Patent Document 1 travels around a plurality of garbage collection points in order according to a preset driving route, and collects garbage at each collection point. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-170463 A Summary of the Invention [Problem to be solved by the invention]

[0004] In a technique such as that of Patent Document 1 in which a garbage truck visits each collection point in turn, the time at which the garbage truck arrives at each collection point may vary depending on the situation. Under such circumstances, users who dump garbage at a collection point cannot know the arrival time of the garbage truck at the collection point they use. Therefore, the garbage truck may arrive at the collection point earlier than the predicted time that the user had anticipated. In this case, the garbage truck will finish collecting garbage at the collection point and depart for the next collection point before the predicted time. As a result, users are unable to dump their garbage. [Means for solving the problem]

[0005] A method of notifying a garbage collection time to solve the above problem is carried out by a garbage collection truck equipped with a receiver that acquires information on its current location, an information processing device capable of communicating with the garbage collection truck, the information processing device storing location information of a plurality of collection points where the garbage collection truck collects garbage and a driving route when the garbage collection truck visits the plurality of collection points, and a vehicle capable of communicating with the information processing device, the vehicle equipped with a camera that captures images of the surroundings and a receiver that acquires information on its current location, the vehicle transmitting an image captured by the camera that shows the collection point together with imaging information indicating the imaging location and imaging date and time of the image to the information processing device, the information processing device acquiring the image and the imaging information transmitted by the vehicle, and the garbage collection vehicle acquires information about its current location; identifies, from among a plurality of images acquired from the vehicle, an unaccumulated image that shows an unaccumulated location, which is a collection location on the travel route that the garbage collection vehicle has not yet passed, based on the information about the current location of the garbage collection vehicle and the imaging information; calculates an estimated garbage amount, which is an estimate of the amount of garbage accumulated at the unaccumulated location, based on the identified unaccumulated image; calculates an estimated arrival time at a specified collection location, which is a collection location that the garbage collection vehicle will pass after the unaccumulated location on the travel route, so that the estimated arrival time at the specified collection location is later when the estimated garbage amount is large than when the estimated garbage amount is small; and outputs notification information to notify the user of the estimated arrival time at the specified collection location. Effect of the Invention

[0006] The above technical concept allows users of the waste collection point to be provided with accurate information about the expected arrival time of the waste collection vehicle. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a schematic diagram of a notification system. [Diagram 2] FIG. 2 is a flowchart showing the procedure of the notification process. [Diagram 3] FIG. 3 is a diagram showing an example of a notification image. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] <Overall composition> An embodiment of a method for notifying users of waste collection times will be described below with reference to the drawings. As shown in Fig. 1, a notification system 100 includes a plurality of vehicles 70, a plurality of user terminals 50, a plurality of waste collection trucks 30, and a server 10. Note that Fig. 1 shows one of the plurality of vehicles 70 as a representative. The same applies to the user terminals 50 and the waste collection trucks 30.

[0009] <Vehicle> The vehicles 70 are registered in advance in the server 10 as vehicles that provide images of garbage collection points. The vehicle 70 includes a control device 72, a camera 82, and a receiver 84. The control device 72 includes a CPU 74, a memory 76, a communication module 78, and a real-time clock 79. The memory 76 stores in advance various programs that describe the processes to be executed by the CPU 74. There are three types of memory 76: RAM, ROM, and electrically rewritable non-volatile type. In this embodiment, these three types are collectively referred to as memory. This point is the same for other control devices and the server 10 described later. The communication module 78 is a communication circuit for wirelessly communicating with the outside via an external communication line network. The communication module 78 receives a command from the CPU 74 to transmit information to the outside, and acquires information from the outside and passes it to the CPU 74. The real-time clock 79 is a circuit that generates date and time information.

[0010] The camera 82 is, for example, an imaging device constituting a drive recorder. The camera 82 captures an image of the surroundings of the vehicle 70, focusing on the outside of the vehicle 70. The camera 82 then generates an image obtained by the capture. The camera 82 repeats capturing an image of the surroundings of the vehicle 70 and generating an image while the vehicle 70 is traveling.

[0011] The receiver 84 acquires information on its current position from a global positioning satellite. The information on the current position is specifically the current position coordinates indicated by latitude and longitude. The current position of the receiver 84 corresponds to the current position of the vehicle 70.

[0012] The CPU 74 repeatedly acquires the current position of the vehicle 70 from the receiver 84. Then, when the current position of the vehicle 70 coincides with any one of the multiple registered points, that is, when the vehicle 70 passes through a registered point, the CPU 74 performs a data generation process. The registered point is a collection point for which an image is to be provided. The memory 76 stores the position coordinates of the multiple registered points in advance. In the data generation process, the CPU 74 acquires from the camera 82 an image captured by the camera 82 when the vehicle 70 passes through a registered point. The CPU 74 handles this image as an image in which the collection point is shown. That is, the CPU 74 acquires from the camera 82 an image in which the collection point is shown among the images captured by the camera 82 at any time. Then, the CPU 74 generates a set of data in which the captured image is provided with the imaging information and the vehicle number. The imaging information is the imaging position and imaging date and time of the acquired image. The CPU 74 handles the position coordinates of the registered point that triggered the execution of the data generation process as the imaging position of the acquired image. Furthermore, the CPU 74 treats the date and time when the vehicle 70 passed the registered point that triggered the execution of the data generation process as the image capture date and time of the acquired image. The vehicle number is an individual identification number assigned to each vehicle 70. After generating the provided data, the CPU 74 transmits the generated provided data to the server 10.

[0013] <User terminal> The multiple user terminals 50 are mobile terminals held by each user who throws away garbage at the collection point. The user terminal 50 is, for example, a smartphone or a tablet terminal. The user terminal 50 includes a control device 52 and a user interface 60. The control device 52 includes a CPU 54, a memory 56, and a communication module 58. The memory 56 stores in advance various programs in which the processes to be executed by the CPU 54 are described. The function of the communication module 58 is the same as the function described for the control device 72 of the vehicle 70. The user interface 60 includes, for example, a touch panel display screen. That is, when a user performs an input operation on the user interface 60, information corresponding to the input operation is input to the control device 52.

[0014] The CPU 54 can execute a display process in response to an operation by a user. The display process is a process for displaying a scheduled arrival timetable of the garbage truck 30 on the user interface 60. That is, when the CPU 54 is instructed to execute the display process, it outputs a transmission request for notification information related to the scheduled arrival time of the garbage truck 30 to the server 10. When the CPU 54 acquires notification information from the server 10 in response to the transmission request, it displays an image corresponding to the acquired notification information on the user interface 60.

[0015] <Garbage truck> A plurality of garbage trucks 30 are registered in advance in the server 10 as vehicles whose positions are to be tracked by the server 10. The garbage truck 30 includes a control device 32 and a receiver 39. The control device 32 includes a CPU 34, a memory 36, and a communication module 38. The memory 36 stores in advance various programs in which the processes to be executed by the CPU 34 are described. The function of the communication module 38 is the same as that described for the control device 72 of the vehicle 70. The receiver 39 acquires information on its own current position from a global positioning satellite. The current position of the receiver 39 corresponds to the current position of the garbage truck 30. The CPU 34 repeatedly transmits a set of data in which the garbage truck number is added to the current position acquired by the receiver 39 to the server 10 as position data. The garbage truck number is an individual identification number assigned to each garbage truck 30.

[0016] As is well known, the garbage collection truck 30 is a dedicated vehicle used for collecting garbage. The garbage collection truck 30 is driven by an operator and travels around multiple collection points in sequence. The garbage collection truck 30 collects garbage at each collection point through the work of the operator.

[0017] <Server> The server 10 is an information processing device owned by a local government or a business operator commissioned by a local government. The server 10 includes a CPU 12, a memory 14, a communication module 16, and a real-time clock 18. The memory 14 stores in advance various programs in which the processes to be executed by the CPU 12 are described. The functions of the communication module 16 and the real-time clock 18 are the same as those described for the control device 72 of the vehicle 70.

[0018] The memory 14 prestores the vehicle numbers of the multiple garbage trucks 30 whose positions are to be tracked. The CPU 12 constantly receives the position data transmitted from these garbage trucks 30.

[0019] The memory 14 prestores the vehicle numbers of a plurality of vehicles 70 that provide images of garbage piles. The CPU 12 constantly receives the provided data transmitted from the plurality of vehicles 70. When the CPU 12 acquires the provided data from each vehicle 70, the CPU 12 continues to store the acquired provided data in the memory 14 only for the day on which the provided data was acquired.

[0020] The memory 14 stores map data in advance. The map data has a plurality of nodes and a plurality of links. Each node indicates a position coordinate expressed by latitude and longitude. A plurality of pieces of additional information are attached to each node. An example of the additional information is the type of the node, such as an intersection or a branch point. Each link is defined as a line segment connecting adjacent nodes. Each link indicates a road. A plurality of pieces of additional information are attached to each link. An example of the additional information is the type of road, such as a national road or a prefectural road.

[0021] The map data includes location information of a plurality of collection points. That is, in the map data, a node corresponding to each collection point is assigned, as supplementary information, an identification value indicating that the node is a collection point, a collection point number assigned to each collection point, and an identification value indicating the district name of each collection point. The collection point number corresponds to the order in which one garbage collection truck 30 collects garbage. That is, the collection point to which the garbage collection truck 30 first heads is assigned the collection point number "1". Then, the collection point number for each collection point increases according to the order of the collection points to which the garbage collection truck 30 heads. The last collection point to which the garbage collection truck 30 heads is assigned an identification value indicating that it is the last collection point. The collection point number is determined for each collection area. A collection area is an area including a plurality of collection points to which one garbage collection truck 30 travels on a specific day of the week.

[0022] The map data includes information on the travel route when the garbage truck 30 travels around a plurality of collection points. That is, in the map data, an identification value indicating that the road is a road on which the garbage truck 30 travels and a route number indicating the number of the travel route are attached to the link corresponding to the travel route as supplementary information. The route number can also be said to be an identification number indicating the collection area. In a certain collection area, the travel route starts from the base station of the garbage truck 30 and connects the nodes of the collection points existing in the collection area in ascending order of collection point number. The end point of the travel route is the base station of the garbage truck 30. In this embodiment, a garbage truck 30 in charge is assigned in advance to each collection area, that is, each travel route. The collection vehicle number of the garbage truck 30 in charge is reflected in the route number.

[0023] <Notification process> The CPU 12 can execute a notification process. The notification process is a process for notifying each user terminal 50 of the scheduled arrival time of the garbage truck 30 at each collection point. The CPU 12 executes the notification process for each collection area. The content of the notification process will be described below for one collection area. When the garbage collection day in the target collection area is reached, the CPU 12 identifies the garbage truck 30 that travels around the collection area based on the route number included in the map data. Then, the CPU 12 repeatedly acquires information on the current position from the garbage truck 30. Then, when the garbage truck 30 departs from the base station, the CPU 12 starts the notification process. The CPU 12 repeats the notification process until the garbage truck 30 reaches the last collection point in the collection area. The CPU 12 starts the notification process at a predetermined time interval, for example, 10 minutes. Note that, when the garbage collection day arrives, the CPU 12 performs the following before executing the first notification process. First, the CPU 12 identifies a travel route to which a route number corresponding to the target collection area is assigned from the map data. After identifying the travel route, the CPU 12 identifies the direction of travel of the garbage truck 30 on this travel route based on the collection point number of the garbage collection point. After this, the CPU 12 identifies the collection point number of the collection point immediately before the last collection point on this travel route as the analysis final number. The CPU 12 uses this identified information in each notification process. Although a detailed explanation is omitted, the CPU 12 stores the information it identifies and generates in the memory 14 as appropriate.

[0024] As shown in Fig. 2, when the CPU 12 starts the notification process, it resets the target location number to zero and then executes the process of step S1. In step S1, the CPU 12 updates the target location number. Specifically, the CPU 12 adds "1" to the previous value, which is the target location number currently set. Then, the CPU 12 sets the obtained value as the new target location number. After this, the CPU 12 advances the process to step S2.

[0025] In step S2, the CPU 12 judges whether or not the collection point corresponding to the target location number is an uncollected collection point. An uncollected collection point is a collection point on the travel route that the garbage truck 30 has not yet passed. As a specific process of this step S2, the CPU 12 first identifies a collection point corresponding to the target location number on the travel route specified in advance. Then, the CPU 12 compares the position of the identified collection point on the travel route with the current position of the garbage truck 30. If the current position of the garbage truck 30 on the travel route is a position further ahead than the collection point corresponding to the target location number, the CPU 12 judges that the collection point corresponding to the target location number is not an uncollected collection point (step S2: NO). In this case, the CPU 12 advances the process to step S12. Note that the CPU 12 also judges step S2 as NO when the current position of the garbage truck 30 coincides with the collection point corresponding to the target location number. Then, the CPU 12 advances the process to step S12. The process of step S12 will be described later.

[0026] On the other hand, in step S2, if the current position of the garbage collection vehicle 30 is located before the collection point corresponding to the target location number on the travel route, the CPU 12 determines that the collection point corresponding to the target location number is an uncollected location (step S2: YES). In this case, the CPU 12 advances the process to step S3. Hereinafter, the uncollected location corresponding to the target location number is referred to as the analysis target location.

[0027] In step S3, the CPU 12 specifies an analysis target image, which is an image showing an analysis target location, from among images included in the multiple provided data stored in the memory 14. The analysis target image corresponds to a non-accumulated image, which is an image showing a non-accumulated location. The CPU 12 specifies the analysis target image as follows. First, the CPU 12 refers to the imaging position in the imaging information attached to the image for each provided data. Then, the CPU 12 specifies an image, the imaging position of which matches the position coordinates of the analysis target location grasped from the map data, as a candidate image. At this time, there may be multiple candidate images, or there may be only one candidate image. When there are multiple candidate images, the CPU 12 refers to the imaging date and time attached to the candidate image. Then, the CPU 12 specifies the image with the latest imaging date and time among the multiple candidate images as the analysis target image. On the other hand, when there is only one candidate image, the CPU 12 specifies the candidate image as it is as the analysis target image. When the CPU 12 specifies the analysis target image, the CPU 12 advances the process to step S4. In this way, through the process of step S3 and the process of step S2, the CPU 12 identifies an analysis target image that shows the analysis target location based on the current position of the garbage collection truck 30 and the image capture position and image capture date and time attached to each provided data. Depending on the collection state of the provided data, it is possible that an analysis target image does not exist. In this case, the CPU 12 skips steps S4, S5, and S6, which will be described later. Then, in step S7, the CPU 12 sets a predetermined standard collection time as the required collection time, which will be described later. The memory 14 stores the standard collection time in advance.

[0028] In step S4, the CPU 12 calculates an estimated amount of waste based on the image to be analyzed identified in step S3. The estimated amount of waste is an estimate of the amount of waste collected in the non-accumulated area. The CPU 12 calculates the estimated amount of waste using a first model previously stored in the memory 14. The first model is a trained model that was trained by supervised learning before the first model was implemented in the server 10. The first model takes the non-accumulated image as input and outputs the estimated amount of waste. The CPU 12 inputs the image to be analyzed to this first model, thereby calculating the estimated amount of waste in the area to be analyzed.

[0029] Here, data in which a certain amount of garbage is added as correct answer information to an unaccumulated image with a certain amount of garbage is called first teacher data. The first model is learned using such first teacher data. That is, when learning the first model, an infinite number of first teacher data generated from various unaccumulated images with different amounts of garbage are prepared in advance. Then, the parameters of the first model are repeatedly adjusted so that the output of the first model when the first model is executed for each first teacher data approaches the correct answer information included in the first teacher data. In this way, the first model is learned. The amount of garbage in the correct answer information is determined as the number of garbage bags. That is, the more the number of garbage bags, the greater the amount of garbage.

[0030] When the CPU 12 calculates the estimated amount of garbage in step S4, the process proceeds to step S5. In step S5, the CPU 12 calculates the litter level based on the analysis target image identified in step S3. Here, the garbage bags collected at the collection point may be destroyed by, for example, wild animals. As a result, the garbage bags may break and the garbage inside may spill out. In this case, the garbage that has spilled out of the garbage bags is in a littered state at the collection point. The litter level is a level obtained by dividing the degree of littering of the garbage that has spilled out of the garbage bags into multiple stages. The CPU 12 calculates the litter level by using a second model that is stored in advance in the memory 14. The second model is a learned model that has been learned by supervised learning before the second model is implemented in the server 10. The second model inputs an uncollected image and outputs a litter level. The CPU 12 inputs the analysis target image to the second model to calculate the litter level at the analysis target point.

[0031] Here, data in which a certain level of clutter is added as correct answer information to an unaccumulated image with a certain level of clutter is called second teacher data. The second model is learned using such second teacher data. That is, when learning the second model, an infinite number of second teacher data generated from various unaccumulated images with different levels of clutter are prepared in advance. Then, the parameters of the second model are repeatedly adjusted so that the output of the second model when the second model is executed on each second teacher data approaches the correct answer information included in the second teacher data. In this way, the second model is learned. The clutter level of the correct answer information is higher as the number of garbage sticking out of the garbage bag increases.

[0032] When the CPU 12 calculates the messiness level in step S5, the CPU 12 advances the process to step S6. In step S6, the CPU 12 sets a time flag. The time flag is a flag for distinguishing between a time when the amount of garbage in the analysis target location is particularly large and other times. As a premise for the CPU 12 to perform step S6, the memory 14 stores in advance a period division list for each collection area. When a year is divided into 24 periods in half-month units, each period is called an individual period. In one period division list, the following contents are defined in correspondence with each of a plurality of collection point numbers included in the target collection area. That is, in one period division list, an individual period that corresponds to the first period and an individual period that corresponds to the second period are defined for each collection point. The first period is a period in which the amount of garbage collected at the target collection point in one garbage collection is equal to or greater than a set value. The second period is a period in which the amount of garbage collected at the target collection point in one garbage collection is less than a set value. As described above, the amount of garbage is the number of garbage bags. The set value is, for example, predetermined as twice the standard waste amount, which is, for example, the annual average amount of waste across the various collection areas.

[0033] As a specific process of step S6, first, the CPU 12 identifies the period-division list of the target collection area from among the period-division lists for each collection area. The above-mentioned route number is attached to each period-division list. The CPU 12 identifies the period-division list for the target collection area based on this route number. When the CPU 12 identifies the period-division list, it identifies a collection point that corresponds to the collection point number of the analysis target point from among the multiple collection points defined in the period-division list. Then, the CPU 12 refers to the first period and the second period related to this collection point. Then, the CPU 12 compares these periods with the image capture date and time attached to the analysis target image identified in step S3, that is, the image capture date and time of the unaccumulated image on which the estimated amount of waste is calculated in step S4. Then, the CPU 12 determines whether the image capture date and time of the analysis target image is within the first period or the second period. If the image capture date and time of the analysis target image is within the first period, the CPU 12 sets the time flag to "1". On the other hand, if the imaging date and time is within the second period, the CPU 12 sets the time flag to “2.” After that, the CPU 12 advances the process to step S7.

[0034] In step S7, the CPU 12 calculates the required collection time. The required collection time is the time required for the worker to collect the garbage in the garbage truck 30 at the analysis target location. The CPU 12 calculates the required collection time based on the estimated garbage amount calculated in step S4, the clutter level calculated in step S5, and the time flag calculated in step S6. If the clutter level and the time flag are the same, the CPU 12 calculates the required collection time so that the required collection time is longer as the estimated garbage amount increases. If the estimated garbage amount and the time flag are the same, the CPU 12 calculates the required collection time so that the required collection time is longer as the clutter level increases. If the estimated garbage amount and the clutter level are the same, the CPU 12 calculates the required collection time so that the required collection time is longer when the time flag is "1" than when it is "2". The CPU 12 calculates the required collection time based on, for example, the following relational expression. That is, the CPU 12 calculates the required collection time by adding together the estimated amount of garbage multiplied by the first coefficient, the clutter level multiplied by the second coefficient, and the correction value. When the timing flag is "1", the CPU 12 sets the correction value to a predetermined positive value. The memory 14 stores the correction value in advance. When the timing flag is "2", the CPU 12 sets the correction value to zero. The first coefficient is a coefficient for converting the estimated amount of garbage into the garbage collection time. The second coefficient is a coefficient for converting the clutter level into the garbage collection time. After calculating the required collection time, the CPU 12 advances the process to step S8.

[0035] In step S8, the CPU 12 calculates the travel time from the analysis target location to the next collection point. The next collection point is an uncollected location where the garbage truck 30 will head next after the analysis target location on the travel route of the garbage truck 30. That is, the next collection point corresponds to a predetermined collection point that is an uncollected location where the garbage truck 30 will pass after the analysis target location. For example, the CPU 12 calculates the travel time based on the travel distance of the garbage truck 30 from the analysis target location to the next collection point and the estimated travel speed of the garbage truck 30. In this case, the CPU 12 calculates the travel time so that the travel time is shorter as the travel distance is shorter. Also, the CPU 12 calculates the travel time so that the travel time is shorter as the estimated travel speed is higher. The CPU 12 can calculate the travel distance, which is the distance from the analysis target location to the next collection point, based on the travel route on the map, for example. The CPU 12 can calculate the estimated travel distance depending on the presence or absence of congestion between the analysis target location and the next collection point, for example. In this case, the CPU 12 may obtain the latest traffic information from the traffic information distribution center to determine whether there is a traffic jam between the analysis target location and the next collection point. After calculating the travel time from the analysis target location to the next collection point, the CPU 12 advances the process to step S9.

[0036] In step S9, the CPU 12 determines whether the currently targeted analysis location is the initial non-accumulation location identified in the current notification process. Hereinafter, this initial non-accumulation location will be referred to as the initial accumulation location. If the analysis location is not the initial accumulation location (step S9: NO), the CPU 12 skips the subsequent step S10 and proceeds to step S11. On the other hand, if the analysis location is the initial accumulation location (step S9: YES), the CPU 12 proceeds to step S10.

[0037] In step S10, the CPU 12 calculates the estimated time of arrival at the first collection point. First, the CPU 12 calculates the travel time from the current position of the garbage collection vehicle 30 to the first collection point in the same manner as in step S8. After calculating the travel time, the CPU 12 sets the time that is advanced from the current time by the calculated travel time as the estimated time of arrival at the first collection point. After calculating the estimated time of arrival at the first collection point, the CPU 12 performs the following. That is, the CPU 12 stores a set of the collection point number of the first collection point and the estimated time of arrival at the first collection point as the initial data of the calculation result information in the memory 14. The estimated times of arrival at other uncollected collection points are sequentially written into this calculation result information through the processing of the subsequent step S11. The collection point number of the first collection point is the target location number at the current time. After this, the CPU 12 advances the processing to step S11.

[0038] In step S11, the CPU 12 calculates the estimated time of arrival at the next collection point. First, the CPU 12 reads out the estimated time of arrival at the non-collection point immediately before the next collection point, i.e., the currently targeted point number, from the calculation result information stored in the memory 14. Then, the CPU 12 calculates the estimated time of arrival at the next collection point based on this estimated time of arrival and the travel time to the next collection point calculated in step S8. That is, the CPU 12 sets the time that is advanced from the estimated time of arrival by the travel time calculated in step S8 as the estimated time of arrival at the next collection point. After this, the CPU 12 adds the estimated time of arrival at the next collection point to the calculation result information in association with the collection point number of the next collection point. After this, the CPU 12 advances the process to step S12.

[0039] In step S12, the CPU 12 determines whether or not analysis has been completed for all collection locations for which estimated arrival times are to be calculated. Specifically, in step S12, the CPU 12 determines whether or not the current target location number matches the above-mentioned analysis final number identified before the start of the notification process. If the current target location number does not match the analysis final number (step S12: NO), the CPU 12 returns to step S1. On the other hand, if the current target location number matches the analysis final number (step S12: YES), the CPU 12 advances the process to step S13.

[0040] In step S13, the CPU 12 generates a notification image. The notification image is an image for notifying users of the collection points of the estimated arrival time of the garbage truck 30 at each collection point. The CPU 12 generates the notification image in a form that reflects the contents of the calculation result information. As shown in FIG. 3, the notification image is, for example, a list of necessary information for each collection point arranged from the top in ascending order of collection point number. The necessary information includes the collection point number, the district name of the collection point, and the estimated arrival time. For the district name of the collection point, the CPU 12 converts an identification value that indicates the district name of the collection point, which is included in the map data, into a character string of the actual district name and describes it in the notification image. The CPU 12 handles the estimated arrival time as follows. That is, for collection points among the multiple collection points for which the estimated arrival time has been calculated, that is, for uncollected collection points, the CPU 12 describes the estimated arrival time listed in the calculation result information in the map image. On the other hand, for collection points for which the estimated arrival time has not been calculated, the CPU 12 describes in the notification image that collection has already been completed. As shown in FIG. 2, after generating the notification image, the CPU 12 advances the process to step S14.

[0041] The collection point corresponding to the analysis final number, that is, the collection point one before the last collection point on the travel route, is called the analysis final point. Here, regarding step S13, if the current position of the garbage truck 30 when the notification process is executed is between the analysis final point and the last collection point on the travel route, the CPU 12 will reach step S13 without calculating the estimated arrival time at the last collection point, taking into account steps S2 and S12. Therefore, the CPU 12 performs the following process in step S13 only if the current position of the garbage truck 30 when the notification process is executed is between the analysis final point and the last collection point. That is, the CPU 12 calculates the estimated arrival time at the last collection point taking into account the travel distance from the current position of the garbage truck 30 to the last collection point, etc. Then, the CPU 12 generates a notification image in which the calculated time is written as the estimated arrival time at the last collection point. In this notification image, it is written that all other collection points have been collected.

[0042] In step S14, the CPU 12 goes into a standby state waiting for a transmission request from the user terminal 50. When the CPU 12 receives a transmission request from the user terminal 50, the CPU 12 outputs the notification image generated in step S13 to the user terminal 50 as notification information. When the user terminal 50 makes a transmission request to the server 10, the user terminal 50 includes in the transmission request the route number of the collection area in which the owner of the user terminal 50 resides. Based on this route number, the CPU 12 can determine which collection area the notification image is being requested for. In step S14, the CPU 12 outputs the notification image when there is a transmission request for the target collection area. After the start of the process of step S14, when it is time to execute the next notification process, the CPU 12 deletes the calculation result information and the notification image from the memory 14 and ends the process of step S14. The CPU 12 then ends the notification process for the time being.

[0043] As described above, the CPU 12 does not execute notification processing during the time period before the garbage collection truck 30 departs from the base station. If the user terminal 50 requests an arrival time during this time period, the CPU 12 outputs to the user terminal 50 an image indicating that garbage collection has not yet started. The CPU 12 also does not execute notification processing during the time period after the garbage collection truck 30 has reached the last collection point. If the user terminal 50 requests an arrival time during this time period, the CPU 12 outputs to the user terminal 50 an image indicating that garbage collection has ended.

[0044] <Actions and Effects of the Embodiments> (1) When a large amount of garbage is collected at a collection point, it takes time to collect the garbage at that collection point. Therefore, in this case, the arrival time at each subsequent non-collecting point will be delayed. The CPU 12 calculates the expected arrival time at each subsequent non-collecting point taking into account the amount of garbage collected at each collection point. Specifically, when calculating the expected arrival time at the next collection point, the CPU 12 calculates the expected arrival time so that when the estimated amount of garbage at the analysis target point is large, the expected arrival time at the next collection point will be later than when the estimated amount of garbage at the analysis target point is small. In this way, by calculating the expected arrival time taking into account the amount of estimated garbage, the CPU 12 can accurately calculate the expected arrival time at each non-collecting point. Then, the CPU 12 outputs a notification image reflecting this expected arrival time to the user terminal 50. This allows the CPU 12 to provide the user with more accurate information as the expected arrival time at each non-collecting point. With such information, the user is less likely to be late for putting out the garbage.

[0045] (2) The amount of garbage collected at the collection point gradually increases as time passes. Therefore, when calculating the estimated amount of garbage based on an image, by using an image taken as late as possible, the actual amount of garbage at the time when the garbage truck 30 arrives at the collection point is less likely to deviate from the estimated amount of garbage. Therefore, when identifying an image to be analyzed, if there are multiple candidate images for the same non-collected point, the CPU 12 selects the image with the latest image capture date and time as the image to be analyzed. Then, the CPU 12 calculates the estimated amount of garbage based on this image to be analyzed. That is, the CPU 12 calculates the estimated amount of garbage at the analysis target point based on the most recent information at the analysis target point. By calculating the estimated arrival time based on such an estimated amount of garbage, the CPU 12 can more accurately calculate the estimated arrival time at each non-collected point.

[0046] (3) Garbage collected at a collection point may be destroyed by, for example, wild animals. In this case, the garbage is scattered at the collection point. If garbage is scattered at a collection point, it takes time to collect the garbage at that collection point. The CPU 12 calculates the estimated arrival time at each uncollected collection point taking into account the degree of littering of the garbage. Specifically, when calculating the estimated arrival time at the next collection point, the CPU 12 calculates the estimated arrival time so that the estimated arrival time at the next collection point is later when the littering level at the analysis target point is high than when the littering level at the analysis target point is low. In this way, by calculating the estimated arrival time taking into account the littering level of the garbage, the CPU 12 can more accurately calculate the estimated arrival time at each uncollected collection point.

[0047] (4) The CPU 12 calculates the estimated time of arrival at each uncollected collection point by taking into account the traffic volume from the current position of the garbage collection truck 30 to each uncollected collection point. Specifically, when calculating the estimated time of arrival at the next collection point, the CPU 12 calculates the estimated time of arrival at the next collection point so that if the traffic volume from the analysis target point to the next collection point is high, the estimated time of arrival at the next collection point will be later than if the traffic volume from the analysis target point to the next collection point is low. In this way, by calculating the estimated time of arrival by taking the traffic volume into account, the CPU 12 can more accurately calculate the estimated time of arrival at each uncollected collection point.

[0048] (5) For example, at collection points near apartments, the amount of garbage may be greater in early spring than in other periods due to people moving in to new homes in the spring. The CPU 12 calculates the estimated time of arrival at each uncollected collection point taking into account such differences in the amount of garbage depending on the season. Specifically, when calculating the estimated time of arrival at the next collection point, the CPU 12 calculates the estimated time of arrival so that, when the image to be analyzed is captured on a date and time within a first period, the estimated time of arrival at the next collection point is later than when the image to be analyzed is captured on a date and time within a second period. In this way, by calculating the estimated time of arrival taking into account differences in the amount of garbage depending on the season, the CPU 12 can more accurately calculate the estimated time of arrival at each uncollected collection point.

[0049] <Example of change> The above embodiment can be modified as follows: The above embodiment and the following modified examples can be combined with each other to the extent that there is no technical contradiction.

[0050] The length of the individual period is not limited to the example in the above embodiment. The individual period may be any length that allows the average change in the amount of waste to be captured. The method of reflecting the first and second periods in the required collection time and thus the estimated arrival time is not limited to the example of the above embodiment. The above reflection method may be any method that satisfies the following: In other words, when the image to be analyzed is captured during the first period, the estimated arrival time at the next collection point is later than when the image to be analyzed is captured during the second period.

[0051] It is not necessary for the estimated time of arrival to reflect the first and second periods. It is not necessary to use a pre-trained model to calculate clutter levels. It is not necessary to define the clutter level by the number of pieces of garbage spilling out of the garbage bag. The clutter level may be defined by any appropriate definition so that the higher the degree of clutter in terms of the amount of garbage spilling out of the garbage bag, the higher the clutter level.

[0052] The method of reflecting the clutter level in the required collection time and therefore the estimated arrival time is not limited to the example in the above embodiment. The above reflection method may be any method that satisfies the following requirements. That is, when the clutter level at the location to be analyzed is high, the estimated arrival time at the next collection point will be later than when the clutter level at the location to be analyzed is low.

[0053] · It is not necessary for the estimated time of arrival to reflect the clutter level. Information indicating the amount of traffic is not limited to the presence or absence of congestion. For example, traffic volume may be indicated by a number of levels.

[0054] The method of reflecting traffic volume in travel time and thus estimated arrival time is not limited to the example of the above embodiment. The above reflection method may be any method that satisfies the following: In other words, when the traffic volume from the analysis target location to the next collection point is high, the estimated arrival time at the next collection point will be later than when the traffic volume from the analysis target location to the next collection point is low.

[0055] It is not necessary to reflect traffic conditions in the estimated arrival time. It is not necessary to use a trained model to calculate the estimated waste volume. It is not necessary to define the amount of waste by the number of garbage bags. Any appropriate definition may be used to define the amount of waste.

[0056] The method of reflecting the estimated amount of waste in the required collection time and therefore the estimated arrival time is not limited to the example in the above embodiment. The reflection method may be any method that satisfies the following requirements. That is, when the estimated amount of waste at the location to be analyzed is large, the estimated arrival time at the next collection point will be later than when the estimated amount of waste at the location to be analyzed is small.

[0057] When there are multiple identical candidate images to identify the image to be analyzed, it is not necessary to select the image with the latest image capture date and time among these candidate images. Even if it is not the image with the latest image capture date and time, as long as it is an image captured on the day of garbage collection, it will be useful information for calculating the estimated garbage volume.

[0058] The notification information is not limited to a notification image. Any notification information that can notify the estimated time of arrival at the uncollected location may be used. The manner in which the notification information is output to the user terminal 50 is not limited to the example of the above embodiment. For example, the notification information may be automatically notified to a pre-registered user terminal 50 without waiting for a transmission request from the user terminal 50. The notification information may be output to the user terminal 50 by using e-mail or the like.

[0059] The following aspect may be adopted for the data generation process performed by the control device 72 of the vehicle 70. That is, when generating the provided data, the CPU 74 may analyze an image captured by the camera 82 when the vehicle 70 passes the registered point, and after confirming that the image shows the collection point, transmit the image to the server 10 as the provided data. [Explanation of symbols]

[0060] 10: server 30: garbage collection truck 39: receiver 70: vehicle 82: camera 84: receiver

Claims

1. A garbage collection vehicle equipped with a receiver for acquiring information on its current location; an information processing device capable of communicating with the garbage collection vehicle, the information processing device storing location information of a plurality of collection points where the garbage collection vehicle collects garbage and a travel route when the garbage collection vehicle travels around the plurality of collection points; a vehicle capable of communicating with the information processing device, the vehicle including a camera for capturing images of the surroundings and a receiver for acquiring information on the vehicle's current location; The vehicle transmits an image of the collection location among the images captured by the camera to the information processing device together with imaging information indicating an imaging position and an imaging date and time of the image, The information processing device, acquiring the image and the imaging information transmitted by the vehicle; Obtaining information on a current location of the garbage collection vehicle; Identifying, from among a plurality of images acquired from the vehicle, a non-accumulation image that shows a non-accumulation location that is a collection location on the travel route that the garbage collection vehicle has not yet passed through, based on information on the current position of the garbage collection vehicle and the imaging information; Calculating an estimated amount of garbage, which is an estimate of the amount of garbage accumulated in the non-accumulated area, based on the identified non-accumulated image; calculating an expected arrival time at a predetermined collection point, which is a collection point that the garbage collection vehicle passes through after the non-collection point on the travel route, so that the expected arrival time at the predetermined collection point is later than the expected arrival time at the predetermined collection point when the estimated garbage amount is large, compared to when the estimated garbage amount is small; and outputting notification information for notifying the expected arrival time at the predetermined collection point. How to notify garbage collection times.

2. When there are a plurality of non-accumulated images for the same non-accumulated location, The information processing device calculates the estimated amount of dust based on the non-accumulated image with the latest image capture date and time among the multiple non-accumulated images for the same non-accumulated location. The method for notifying a garbage collection time according to claim 1.

3. The information processing device, Calculating a litter level in the non-accumulated area based on the non-accumulated image; and calculating the estimated time of arrival at the predetermined collection point so that, when the clutter level is high, the estimated time of arrival at the predetermined collection point is later than when the clutter level is low. The method for notifying a garbage collection time according to claim 1.

4. The information processing device, acquiring traffic information on the travel route from the non-accumulation point where the estimated amount of waste has been calculated to the predetermined accumulation point; and calculating the estimated arrival time at the specified collection point so that the estimated arrival time at the specified collection point is later when the traffic volume between the non-collection point where the estimated amount of waste is calculated and the specified collection point is large, compared to when the traffic volume is small. The method for notifying a garbage collection time according to claim 1.

5. the information processing device stores a period division list that indicates, for each of the collection locations, a first period during which the amount of garbage is equal to or greater than a predetermined set value and a second period during which the amount of garbage is less than the set value; When the image capture date and time of the non-accumulated image on which the estimated amount of dust is calculated is within the first period corresponding to the non-accumulated location for which the estimated amount of dust is calculated, the estimated arrival time at the specified collection location is calculated so that the estimated arrival time at the specified collection location is later than when the image capture date and time is within the second period. The method for notifying a garbage collection time according to claim 1.

Images