Arrival time prediction method, arrival time prediction device, and arrival time prediction program

Abstract

To provide an arrival time prediction method and the like with which it is possible to accurately predict the arrival time per lane of a road that leads to at least one of a first destination point and a second destination point in a region (rectification region) which is within a prescribed distance centering on the first destination point and in which the second destination point also exists.SOLUTION: Provided is a method for calculating the time required per lane of a road that leads to at least one of a first destination point and a second destination point, the method for predicting an arrival time including an arrival time prediction step for estimating, on the basis of the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, the number of vehicles per lane between the first detection device and the second detection device, and predicting an arrival time per lane at least at one of the first destination point and the second destination point.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] In recent years, in port districts and logistics areas, traffic congestion occurs frequently due to the formation of queues on specific lanes on roads for vehicles to enter destinations such as terminals or warehouses. In such cases, it is difficult to determine whether a vehicle has joined the queue or has slowed down due to traffic congestion using only the probe information from the onboard devices installed in each vehicle. Therefore, if the speed is calculated by simply taking the average speed of passing vehicles, there is a problem that the arrival time at the destination cannot be accurately predicted. Summary of the Invention [Problem to be solved by the invention]

[0003] The present invention aims to solve the above-mentioned problems in the prior art and to achieve the following objective: That is, the present invention aims to provide an arrival time prediction method, an arrival time prediction device, and an arrival time prediction program that can accurately predict the arrival time for each lane of a road leading to at least one of a first destination and a second destination in an area within a predetermined distance from a first destination and in which a second destination is also present (hereinafter, sometimes referred to as a "rectification area"). [Means for solving the problem]

[0004] The means for solving the above problems are as follows. <1> A method for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of the first destination point and the second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, the method comprising: Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; The method for predicting arrival time includes an arrival time predicting step of predicting an arrival time for each lane to at least one of the first destination point and the second destination point. <2> The arrival time prediction for each lane in the arrival time prediction step is deriving the number of vehicles for each lane between the first detection device and the second detection device based on current probe information for each lane, and predicting an arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles for each lane and past probe information for each lane; <1> This is the arrival time prediction method described in the above. <3> The arrival time prediction for each lane in the arrival time prediction step is deriving the number of vehicles and the traveling speed for each lane between the first detection device and the second detection device based on current probe information for each lane, and predicting the arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles and the traveling speed for each lane and past probe information for each lane; <1> This is the arrival time prediction method described in the above. <4> In the arrival time prediction step, when deriving a travel speed based on current probe information for each lane, the current probe information exists in a plurality of pieces, Included in the plurality of pieces of probe information, Location information X1 at time X in the first probe information is substantially identical to location information X2 at a time substantially the same as the time X in second probe information different from the first probe information, and When location information Y1 at a time Y sampled next to the time X in the first probe information and location information Y2 at a time substantially the same as the time Y in the second probe information are substantially identical, determining that the first probe information and the second probe information are related to the same moving object; <3> This is the arrival time prediction method described in the above. <5> 1. A device for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of the first destination point and the second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; The arrival time prediction device is characterized by comprising an arrival time prediction means for predicting an arrival time for each lane to at least one of the first destination point and the second destination point. <6> A program for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of a first destination point and a second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; The arrival time prediction program causes a computer to perform an arrival time prediction process for predicting an arrival time for each lane to at least one of the first destination point and the second destination point. Effect of the Invention

[0005] According to the present invention, it is possible to solve the above-mentioned problems in the conventional art and achieve the above-mentioned object, and the present invention can provide an arrival time prediction method, an arrival time prediction device, and an arrival time prediction program that can accurately predict the arrival time for each lane of a road leading to at least one of a first destination and a second destination in an area (a flow-regulating area) within a predetermined distance from a first destination point, where the second destination point is also present in the area. [Brief description of the drawings]

[0006] [Figure 1] FIG. 1 is a schematic diagram explaining the rectification region and the general-purpose region. [Diagram 2] FIG. 2 is a schematic diagram showing the mounting position of the on-board device D or the on-board device G in each vehicle. [Diagram 3] FIG. 3 is a schematic diagram showing the "complete" state of the present invention. [Figure 4] FIG. 4 is a diagram showing an example of a hardware configuration of the arrival time prediction device of the present invention. [Diagram 5] FIG. 5 is a diagram showing an example of a functional configuration of the arrival time prediction device of the present invention. [Figure 6] FIG. 6 is an explanatory diagram for explaining the definitions of terms in the embodiments. [Figure 7] FIG. 7 is a diagram illustrating an example of a road state in the first embodiment. [Figure 8] FIG. 8 is a flowchart illustrating an example of a process of the arrival time prediction method according to the first embodiment. [Figure 9] FIG. 9 is a graph showing the relationship between traffic density and speed. [Figure 10] FIG. 10 is a graph showing a linear equation that is a simplification of the “Greenberg model” of FIG. [Figure 11] FIG. 11 is a diagram showing the total number of vehicles in each of the nine sections: [PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)]. [Figure 12]FIG. 12 is a flowchart illustrating an example of a process of the arrival time prediction method according to the second embodiment. [Figure 13] FIG. 13 is a diagram for explaining the "set" process in the second embodiment. [Figure 14] FIG. 14 shows container cars and cars traveling alone categorized by the presence or absence of on-board equipment. [Figure 15] FIG. 15 is a diagram classifying container cars according to whether or not they have on-board equipment in the head section and in the chassis section of the container cars. [Figure 16] FIG. 16 is a diagram illustrating a state in which each vehicle is identified in a vehicle identification section in the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] (Arrival time prediction method and arrival time prediction device) The arrival time prediction method of the present invention is a method for calculating the required time to reach at least one of a first destination and / or a second destination for each lane of a road leading to at least one of the first destination and / or the second destination in an area within a predetermined distance from a first destination, where a second destination is also located within the area, and includes an arrival time prediction step and further includes other steps as necessary.

[0008] The arrival time prediction device of the present invention is a device that calculates the required time to reach at least one of a first destination and / or a second destination for each lane of a road leading to at least one of the first destination and / or the second destination in an area within a predetermined distance from a first destination, where a second destination is also located within the area, and has an arrival time prediction means and further has other means as necessary.

[0009] The arrival time prediction method of the present invention can be suitably implemented by the arrival time prediction device of the present invention, and the arrival time prediction step can be performed by the arrival time prediction means, and the other steps can be performed by other means.

[0010] In the present invention, in a flow-stabilizing area, which is an area within a predetermined distance from a first destination point and in which a second destination point is also present, it is possible to accurately predict the arrival time for each lane of a road leading to at least one of the first destination point and the second destination point.

[0011] There are no particular limitations on the first destination point, and it can be appropriately selected depending on the purpose. Examples of the first destination point include a container base, a shopping mall, and an event venue. There are no particular limitations on the second destination point, and it can be appropriately selected depending on the purpose, and an example of the second destination point is a construction site. The "predetermined distance" is not particularly limited and can be appropriately selected depending on the purpose, but is preferably, for example, 2 km or more, and more preferably 2 km or more and 10 km or less.

[0012] As shown in FIG. 1, the rectification area 10 is an area within a predetermined distance from a first destination point 12, in which a second destination point 13 is also present, and refers to an area in which, when vehicles are concentrated toward a specific destination point, the area is rectified with lanes on which vehicles can travel, and the time required to reach the destination point can be predicted using minor and sophisticated tools (methods). The general-purpose area 11 is an area where vehicles heading to various destinations can travel freely without lane restrictions, and is an area where general-purpose tools (methods) for calculating travel time from traffic volume can be used, such as urban areas. Most of the vehicles depart from the general area 11, enter the stabilization area 10, and travel toward multiple destinations.

[0013] In the rectification area 10, there are multiple lane roads for accessing multiple destinations, and it is possible to determine the lane for each destination. For example, vehicles bound for a container base use the first lane, vehicles bound for a construction site use the third lane, and general vehicles passing through the rectification area mainly use the second lane. Table 1 shows the types of vehicles passing through the rectification area.

[0014] [Table 1]

[0015] <Arrival time prediction process and arrival time prediction means> The arrival time prediction process is a process of estimating the number of vehicles for each lane between a first detection device and a second detection device based on the number of vehicles detected for each lane by a first detection device installed on a road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, and predicting the arrival time for each lane to at least one of the first destination point and the second destination point, and is performed by an arrival time prediction means.

[0016] The first detection device is not particularly limited as long as it is installed on a road and can detect the number of vehicles per lane, and can be appropriately selected according to the purpose, and examples thereof include imaging devices such as ultrasonic sensors, infrared sensors, and cameras. Among these, cameras are preferred, and 5G-compatible 360° AI cameras are more preferred. With a camera, for example, the number of vehicles per lane, the number of vehicles per vehicle type, the speed per vehicle type, license plate identification, video information, etc. can be obtained.

[0017] There are no particular limitations on the second detection device as long as it is installed next to the first detection device, and the same device as the first detection device can be used. The term "next to" the first detection device means that the detector may be placed in the "next" position of a consecutive first detection device, or in the "next" position of a first detection device spaced a predetermined distance apart, but it is preferable to place the detector in the "next" position of a consecutive first detection device. For example, a camera serving as a first detection device is installed near a gate within the flow control area where traffic congestion is likely to occur, and then a camera serving as a second detection device is installed at an appropriate interval (e.g., 1 km) from the gate to supplement the probe information.

[0018] <<First embodiment>> In the first embodiment, a method for estimating the number of vehicles for each lane between the first detection device and the second detection device based on the number of vehicles detected for each lane by a first detection device in the rectification area and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, is, for example, preferably to derive the number of vehicles for each lane between the first detection device and the second detection device based on current probe information for each lane, and to predict the arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles for each lane and the past probe information for each lane.

[0019] The current probe information means real-time information obtained from various on-board devices mounted on a vehicle currently traveling. Using the current probe information enables real-time processing. Examples of the current probe information include a vehicle ID, a travel route ID, a position (longitude, latitude), a speed, a departure date and time, a departure point (longitude, latitude), an arrival date and time, a destination (longitude, latitude), a travel distance, and a required time. The past probe information refers to a large amount of accumulated past probe information, so-called "travel history data." Processing using past probe information is sometimes called batch processing.

[0020] Specifically, the number of vehicles is grasped at regular time intervals based on the current various types of probe information for each distance section on the roads in the rectification area. Note that the current various types of probe information cannot distinguish between lanes, so the number of vehicles in each lane cannot be grasped. Next, in order to grasp the number of vehicles in each lane, a camera installed at a predetermined position is used to capture the number of vehicles in each lane within two distance sections. Next, based on the number of vehicles per lane in the two distance sections captured by the cameras, the number of vehicles per lane for each distance section within the "measurement section," which is the section where two consecutive cameras are installed, is estimated. Next, based on the current probe information for each distance section on the road, the time required from the starting point to the first destination point and the time required from the starting point to the second destination point are predicted using a function. The function can be calculated based on past probe information and traffic density (number of vehicles) for each lane as a parameter in accordance with the Greenberg model, etc. In the field of traffic engineering, the Greenberg model is well known, but it is common to create a function based on reality.

[0021] <<Second embodiment>> In a second embodiment, in the rectification area, it is preferable to predict the arrival time for each lane in the arrival time prediction process by deriving the number of vehicles and the traveling speed for each lane between the first detection device and the second detection device based on current probe information for each lane, and predicting the arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles for each lane and the traveling speed and past probe information for each lane, since this allows for a more accurate prediction of the arrival time taking into account the traveling speed. The current probe information and the past probe information have the same meanings as those in the first embodiment.

[0022] In the second embodiment, when deriving a traveling speed based on current probe information for each lane, if there are a plurality of pieces of current probe information, and position information X1 at time X in first probe information included in the plurality of pieces of probe information is substantially identical to position information X2 at substantially the same time as the time X in second probe information different from the first probe information, and position information Y1 at time Y sampled next to the time X in the first probe information is substantially identical to position information Y2 at substantially the same time as the time Y in the second probe information, the first probe information and the second probe information are determined to be probe information related to the same moving object.

[0023] In the second embodiment, a "set" is defined as the first probe information and the second probe information being probe information related to the same vehicle. The above-mentioned "complete set" means that the head part of the container vehicle and the chassis part of the container vehicle are connected, and the head part of the container vehicle is equipped with an on-board unit for probe D, which is a digital tachograph (D) (collectively referred to as "on-board D" and on-board ID as "D-ID"), and the chassis part of the container vehicle is equipped with an on-board unit for probe G, which is a GPS capture device (G) (collectively referred to as "on-board G" and "G-ID"). In order to accurately grasp the state of the "complete set," it is necessary to grasp all vehicle D and vehicle G and determine whether their movements are substantially the same (nearby) during a certain time period and on a certain road section.

[0024] Examples of the vehicle include the head portion of a container vehicle, the chassis portion of a container vehicle, a truck, a ready-mix concrete vehicle, a dump truck, a regular automobile, a large special-purpose automobile, and a small special-purpose automobile. The same vehicle means one vehicle. For example, if the head of a container vehicle and the chassis of the container vehicle are connected, they are one vehicle and the same vehicle, but if the head of the container vehicle and the chassis of the container vehicle are separated, the head of the container vehicle and the chassis of the container vehicle become two vehicles and are no longer the same vehicle. The same vehicle having multiple pieces of probe information means that one vehicle has at least a first probe information and a second probe information, and may have three or more types of probe information.

[0025] In the "single set" discrimination method, the number of samples of the first probe information and the second probe information may be the same, but it is preferable for the number of samples of the first probe information and the number of samples of the second probe information to be different from each other in order to increase the accuracy of discriminating that they are the same vehicle, and it is more preferable for the number of samples of the first probe information to be greater than the number of samples of the second probe information.

[0026] The first probe for collecting the first probe information may be, for example, a digital tachograph. Examples of the second probe that collects the second probe information include a GPS capture device and a smartphone.

[0027] The features of the digital tachograph, GPS trap, and smartphone are as shown in Table 2 below.

[0028] [Table 2]

[0029] Digital tachographs (sometimes abbreviated as "DigiTach") are certified by the Ministry of Land, Infrastructure, Transport and Tourism's Road Transport Bureau in cooperation with the Ministry of Health, Labor and Welfare, with the aim of visualizing the working conditions of truck and bus operators and making it mandatory to keep operation records. Approximately 20 private companies have shipped certified equipment, including Transtron Co., Ltd. and Yazaki Energy Systems Co., Ltd. In this invention, a digital tachograph made by Transtron Co., Ltd., which is strong in cloud-based systems, is used, but there are no particular limitations as long as it is cloud-based. The digital tachograph has a power supply and the GPS acquisition interval is one second.

[0030] GPS tracking devices do not require a power supply and can only function when towed by a towed vehicle (trailer, chassis of a container car, double-linked vehicle, etc.), so in recent years, grasping location information has become an issue for logistics companies. For this reason, several private companies have begun to commercialize the devices. The GPS tracking devices do not require a power supply, and the GPS acquisition interval can be changed from 5 seconds to 1 hour depending on the application, depending on the battery life.

[0031] In the method of determining "a set," it is determined that the first probe information and the second probe information are probe information related to the same moving object. For example, when there is "first probe information D" and "second probe information G," the installation location of the on-board probe device that collects the probe information is usually determined by the design of the on-board device based on the purpose and characteristics of the on-board device. Specifically, on-board D is generally mounted on the head of a container truck or the front part in front of the seats of a ready-mix concrete truck, and on-board G is generally mounted near the front of the loading platform of a dump truck or the chassis part of a container truck, and their mounting locations are generally as shown in Figures 2 and 3. The "set" corresponds to Figure 3.

[0032] Here, the terms "substantially the same time" and "substantially the same position information" will be explained separately for a moving state of a moving body and a stopped state of a moving body. As shown in Figure 3, the head of a container vehicle is located in front of the chassis of the container vehicle in the direction of travel. Therefore, when the vehicle is moving, vehicle D passes before vehicle G passes. When the head of a container vehicle is alone, it often moves backwards to connect with the chassis of the container vehicle, but when the head of the container vehicle and the chassis of the container vehicle are connected, it rarely moves backwards.

[0033] In order to fuse multiple types of probe information, it is necessary to know whether multiple vehicles are substantially the same (neighborhood). In the present invention, whether vehicles are substantially the same (neighborhood) is determined by the "space-time distance" as described below. In the case of the multiple-type probes, probe D and probe G, the head of the container vehicle is located forward in the direction of travel relative to the chassis of the container vehicle, and when probe D is in front of probe G, the difference in the probe acquisition times (time G - time D) is not a negative value. Therefore, the space-time distance is defined by the following formula 1. Note that 1 second of longitude and latitude is approximately 25 m.

[0034] [Formula 1] TIFF0007613739000003.tif17165

[0035] When the fact that the position information X1 at time X in the first probe information is substantially identical to the position information X2 at substantially the same time as the time X in the second probe information different from the first probe information, and the fact that the position information Y1 at time Y sampled next to the time X in the first probe information is substantially identical to the position information Y2 at substantially the same time as the time Y in the second probe information, are expressed as a spatiotemporal distance, the spatiotemporal distance between the first probe information and the second probe information is positive or zero, and the probes with the minimum spatiotemporal distance are regarded as substantially identical (nearby). Specifically, the spatiotemporal distance between the first probe information and the second probe information in a moving state is preferably 1.5 (sec·sec) or less, more preferably 1.0 (sec·sec) or less.

[0036] <Other steps and other means> The other steps are not particularly limited and can be appropriately selected depending on the purpose. Examples of the other steps include a communication step and an input step. The other means is not particularly limited and can be appropriately selected depending on the purpose. Examples of the other means include communication means and input means.

[0037] The communication unit is not particularly limited as long as it is capable of communicating with the arrival time prediction device, and any known communication unit may be used as appropriate, such as a transceiver, an information and communication network, or the Internet.

[0038] The input unit is not particularly limited as long as it can accept various requests to the arrival time prediction device, and any known device can be used as appropriate, such as a keyboard, a mouse, a touch panel, or a microphone.

[0039] (Arrival time prediction program) The arrival time prediction program of the present invention is a program for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of the first destination point and the second destination point, in an area within a predetermined distance from a first destination point, the area including a second destination point, Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; The computer is caused to perform an arrival time prediction process for predicting the arrival time for each lane to at least one of the first destination point and the second destination point.

[0040] The arrival time prediction program of the present invention can be, for example, a program for causing a computer to execute the arrival time prediction method of the present invention. Also, a preferred embodiment of the arrival time prediction program of the present invention can be, for example, similar to a preferred embodiment of the arrival time prediction method of the present invention.

[0041] The arrival time prediction program of the present invention can be created using various known programming languages ​​depending on the configuration of the computer system used and the type and version of the operating system.

[0042] The arrival time prediction program of the present invention may be recorded on a recording medium such as an internal hard disk or an external hard disk, or on a recording medium such as a CD-ROM, a DVD-ROM, an MO disk, or a USB memory. Furthermore, when the arrival time prediction program of the present invention is recorded on the above-mentioned recording medium, it can be used by installing it directly or on a hard disk via a recording medium reading device of the computer system as necessary. The arrival time prediction program of the present invention may also be recorded in an external storage area (such as another computer) accessible from the computer system via an information and communication network. In this case, the arrival time prediction program of the present invention recorded in the external storage area can be used by installing it directly or on a hard disk via an information and communication network from the external storage area as necessary. The arrival time prediction program of the present invention may be divided into any number of processes and recorded on a plurality of recording media.

[0043] <Computer-readable recording medium> A computer-readable recording medium according to the present invention records the arrival time prediction program of the present invention. The computer-readable recording medium related to the present invention is not particularly limited and can be appropriately selected depending on the purpose. Examples of the computer-readable recording medium include an internal hard disk, an external hard disk, a CD-ROM, a DVD-ROM, an MO disk, and a USB memory. Furthermore, the computer-readable recording medium related to the present invention may be a plurality of recording media on which the arrival time prediction program of the present invention is recorded in a divided form for each arbitrary process.

[0044] An example of the technology disclosed in the present invention will be described in more detail below using an example of the configuration of an apparatus and a flow chart. FIG. 4 shows an example of the hardware configuration of the arrival time prediction device of the present invention. In the arrival time prediction device 100, for example, a control unit 101, a main memory device 102, an auxiliary memory device 103, an I / O interface 104, a communication interface 105, an input device 106, an output device 107, and a display device 108 are connected via a system bus 109.

[0045] The control unit 101 performs calculations (arithmetic operations, comparison operations, etc.), controls the operation of hardware and software, etc. The control unit 101 may be, for example, a CPU (Central Processing Unit), or may be a part of a machine used in the arrival time prediction method, or may be a combination of these. The control unit 101 realizes various functions by, for example, executing a program (such as the arrival time prediction program of the present invention) loaded into the main storage device 102 or the like. The process performed by the determining means (determining unit) in the arrival time prediction device of the present invention can be performed by the control unit 101, for example.

[0046] The main memory device 102 stores various programs and also stores data necessary for executing the various programs. As the main memory device 102, for example, one having at least one of a Read Only Memory (ROM) and a Random Access Memory (RAM) can be used. The ROM stores various programs such as a BIOS (Basic Input / Output System), etc. The ROM is not particularly limited and can be appropriately selected according to the purpose, and examples thereof include a mask ROM and a PROM (Programmable ROM). The RAM functions as a working area in which various programs stored in, for example, the ROM or the auxiliary storage device 103 are deployed when they are executed by the control unit 101. There are no particular limitations on the RAM, and it can be appropriately selected depending on the purpose, and examples of the RAM include DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory).

[0047] The auxiliary storage device 103 is not particularly limited as long as it can store various information, and can be appropriately selected according to the purpose, and examples of the auxiliary storage device 103 include a solid state drive (SSD), a hard disk drive (HDD), etc. The auxiliary storage device 103 may also be a portable storage device such as a CD drive, a DVD drive, or a BD (Blu-ray (registered trademark) Disc) drive. Moreover, the arrival time prediction program of the present invention is stored in, for example, the auxiliary storage device 103, loaded into the RAM (main memory) of the main storage device 102, and executed by the control unit 101.

[0048] The I / O interface 104 is an interface for connecting various external devices. The I / O interface 104 enables input and output of data from, for example, a CD-ROM (Compact Disc ROM), a DVD-ROM (Digital Versatile Disk ROM), an MO disk (Magneto-Optical disk), a USB memory (Universal Serial Bus (USB) flash drive), and the like.

[0049] There are no particular limitations on the communication interface 105, and any known interface may be used as appropriate, such as a wireless or wired communication device.

[0050] The input device 106 is not particularly limited as long as it can accept various requests and information input to the arrival time prediction device 100, and any known device can be used as appropriate, such as a keyboard, a mouse, a touch panel, a microphone, etc. In addition, when the input device 106 is a touch panel (touch display), the input device 106 can also function as the display device 108.

[0051] There are no particular limitations on the output device 107, and any known device may be used as appropriate, such as a printer. There are no particular limitations on the display device 108, and any known display device may be used as appropriate, such as a liquid crystal display or an organic EL display.

[0052] FIG. 5 shows an example of the functional configuration of the arrival time prediction device of the present invention. As shown in FIG. 5, the arrival time prediction device 100 includes a communication function unit 120, an input function unit 130, an output function unit 140, a display function unit 150, a storage function unit 160, and a control function unit 170.

[0053] The communication function unit 120 transmits and receives various data to and from an external device, for example. The communication function unit 120 may receive data such as vehicle attribute data and driving history data from an external device, for example. The input function unit 130, for example, receives various instructions for the arrival time prediction device 100. In addition, the input function unit 130 receives, for example, a plurality of pieces of probe information. The output function unit 140, for example, prints out the arrival time predicted by the arrival time prediction device. The display function unit 150 displays, for example, the arrival time predicted by the arrival time prediction device on a display.

[0054] The storage function unit 160 stores various programs, for example, and also has a probe information DB 161, a camera information DB 162, and a predicted result DB 163. The probe information DB 161 is a DB that stores the probe information collected by the vehicle-mounted device. The probe information includes the following data items, as shown in Table 3 below: "Vehicle ID," "Route ID," "Location (longitude, latitude)," "Speed," "Departure date and time," "Departure point (longitude, latitude)," "Arrival date and time," "Destination (longitude, latitude)," "Distance traveled," and "Travel time required."

[0055] [Table 3]

[0056] The data item "Vehicle ID" is data for identifying the vehicle in which the on-board device is installed, and is set in advance. The data item "driving route ID" is used to identify a driving route, which is a unit of travel for moving from a certain starting point to a certain destination with a purpose. The data items "date and time" and "location (longitude, latitude)" are obtained by a GPS (Global Positioning System) unit installed in the vehicle-mounted device. The "speed" data item is obtained from the vehicle's axles using a speed sensor installed in the vehicle and synchronized with the GPS unit. The data items "departure date and time" and "departure point (longitude, latitude)" are the departure date and time and the longitude and latitude of the departure point of the travel route. The data items "Arrival date and time" and "Destination (longitude, latitude)" are the arrival date and time and the longitude and latitude of the destination on the travel route. The data item "travel distance" is the travel distance from the starting point to the current location. The data item "travel time" is the time required from the starting point to the current location.

[0057] The camera information DB 162 is a DB that stores image data of vehicles captured by a camera. Based on the data in the camera information DB, the number of vehicles detected for each lane can be estimated.

[0058] The predicted result DB 163 is a DB that stores the results of the arrival time predicted by the arrival time prediction device.

[0059] The control function unit 170 has an arrival time prediction unit 171 as an arrival time prediction means, and a determination unit 172. The control function unit 170 executes various programs stored in the storage function unit 160, for example, and controls the operation of the entire arrival time prediction device 100.

[0060] The arrival time prediction unit 171 estimates the number of vehicles for each lane between the first detection device and the second detection device based on the number of vehicles detected for each lane by a first detection device installed on a road in a stabilization area and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, and predicts the arrival time for each lane to at least one of the first destination point and the second destination point. In addition, the arrival time prediction unit 171 derives a traveling speed based on past probe information for each lane of the road in the rectification area, and predicts an arrival time for each lane to at least one of the first destination point and the second destination point from the derived traveling speed and the number of vehicles for each lane.

[0061] The discrimination unit 172 performs a process of discriminating that the first probe information and the second probe information are probe information related to the same moving body when location information X1 at time X in a first probe information included in the plurality of pieces of probe information and location information X2 at substantially the same time as the time X in a second probe information different from the first probe information are substantially identical, and when location information Y1 at time Y sampled next to the time X in the first probe information and location information Y2 at substantially the same time as the time Y in the second probe information are substantially identical. EXAMPLES

[0062] Examples of the present invention will be described below, but the present invention is not limited to these examples. The definitions of terms used in the following examples will be explained.

[0063] <Terminology> As shown in FIG. 6, for road R (e.g., 3 km from starting point o) in a flow area within a predetermined distance from first destination point a, where second destination point b is also present, the lane number is j, where j = 1 to 3. The distance section number based on the distance from second destination point b (e.g., in units of 100 m) is i, where i = 1 to 30. The corresponding part of road R is represented as R(i,j). When lane numbers on road R are not taken into consideration, it is represented as R(i,).

[0064] The cameras are installed near gates on road R within the rectification area where traffic congestion (queues) are likely to occur, and are installed at predetermined intervals (e.g., 1 km) to capture probe information from near the gates. The locations where the cameras are installed can be appropriately selected based on past statistical information to ensure that probe information can be easily captured. The location of a camera is represented as C(i). In Fig. 6, the cameras are installed at C(1), C(8), C(16), and C(24). Each camera can capture two distance intervals (boxed areas). The section where two consecutive cameras are installed is called the "measurement section." In Figure 6, these are "measurement section 1," "measurement section 2," and "measurement section 3." The "measurement section" is a parameter of the function that predicts the required time.

[0065] As shown in FIG. 6, on a road R in a stabilization area, a "vehicle identification section" (for example, 300 m) for identifying the type of each vehicle is provided immediately before a starting point o. The type of vehicle can be identified based on information from the onboard device installed in each vehicle. For example, if the probe acquisition time intervals of the onboard device are digital tachograph: 1 second, GPS acquisition device: 5 seconds, and smartphone: 30 seconds, when traveling at a speed of 60 km / h, information can be acquired at intervals of digital tachograph: 16 m, GPS acquisition device: 83 m, and smartphone: 500 m.

[0066] The digital tachograph D is mounted on the head of a container truck or on a ready-mix concrete truck. Since the digital tachograph D is permanently mounted on each vehicle, it can be identified by an on-board ID (hereinafter referred to as "D-ID"). There is a strong demand to keep the D-ID secret. With the digital tachograph D, it is possible to predict the destination after passing the starting point o based on the driving history data.

[0067] The GPS trap G is mounted on the chassis of a dump truck or a container truck. The GPS trap G is easily attached to each vehicle and is frequently mounted on different vehicles, so there is little need to conceal the vehicle-mounted ID (hereinafter referred to as "G-ID"). Therefore, when the GPS trap is attached to the vehicle, information such as the license plate, destination, or driving route of the vehicle to which it is installed is linked to the vehicle-mounted ID and registered.

[0068] For the relevant portion R(i,j) of road R, the current speed of the digital tachograph is VD(i,j), the current speed of the GPS tracking device is VG(i,j), and the current speed of the smartphone is VS(i,j), and the sum of these is the current speed VP(i,j) based on the probe information. The current speed based on the camera is defined as VC(i,j), and when it is confirmed that a queue has formed in a particular lane (high traffic density and low speed), the speed of movement within the queue is defined as VQ(i,j).

[0069] The speed of each vehicle is determined by comparing the GPS positions of two points, except when a method of obtaining the speed directly from the vehicle, such as a digital tachograph, is used. For example, the current speed of the GPS capture device, VG(i,), uses the GPS position information of R(i,) and R(i+1,). The speed in the "vehicle identification section" can be determined from information from the onboard device in the "vehicle identification section", except for smartphones.

[0070] The traffic density (number of vehicles) based on the probe information is expressed as KP(i,j), the traffic density (number of vehicles) based on the camera is expressed as KC(i,j), and the true traffic density (number of vehicles) is expressed as KE(i,j).

[0071] The time required T(a←0) from the starting point o to the destination point a (a←o) and the time required T(b←o) from the starting point o to the destination point b (b←o) can be calculated from the functions Fa and Fb. The functions Fa and Fb can be calculated in advance based on the Greenberg model or the like.

[0072] Example 1 <Method of predicting approximate number of vehicles per lane by supplementing probe information with cameras> As shown in Figure 7, on road R within the stabilization area, there is a starting point o (distance section No. i = 30), a first destination point Gate a (distance section No. i = 7), and a second destination point Gate b (distance section No. i = 1). In each distance section (i=1-30) on road R, three lanes (lane No. j=1-3) are actually travelling with the number of vehicles circled in Fig. 7. In this state, although multiple types of probe information are acquired, information on all the lanes on road R cannot be grasped.

[0073] First, the number of vehicles is determined based on various probe information in each distance section (i = 1 to 30) on road R at regular time intervals. Note that the various probe information does not allow distinction between lane numbers, so the number of vehicles per lane cannot be determined. The intervals at which the various probe information is obtained are fixed and known, as described above. Next, in order to grasp the number of vehicles in each lane, the number of vehicles in each lane within two distance sections (framed areas) is captured using cameras installed at C(1), C(7), C(17), and C(27).

[0074] Next, based on the number of vehicles per lane in the two distance sections captured by the cameras, the number of vehicles per lane for each distance section within the "measurement section," which is the section where two consecutive cameras are installed, is estimated.

[0075] Next, based on past probe information for each distance section on road R, the travel time T(a←0) from starting point o to destination point a (a←o) and the travel time T(b←o) from starting point o to destination point b (b←o) are predicted using functions Fa and Fb. The functions Fa and Fb can be calculated in advance based on past probe information, using PM(k,j), which is the traffic density (number of vehicles) for each lane, as a parameter.

[0076] Next, the required time T(a←o) and the required time T(b←o) can be calculated as shown in FIG. In step S1, the number of vehicles KP(i,) based on probe information is determined for each distance section (total of three lanes) every 10 seconds from position and speed information from vehicles equipped with onboard devices traveling on road R within the rectification area, and then the process moves to step S2.

[0077] In step S2, the number of vehicles based on the camera, KC(i,j), is determined from the number of vehicles per lane in two analyzable distance sections from the location C(i) where the camera is installed, and the process proceeds to step S3.

[0078] In step S3, the actual number of vehicles, KE(i, j), is estimated for each lane based on the "measurement section" of two consecutive cameras, and then the process proceeds to step S4. Specifically, as shown in Table 4, we estimate KE(i,j), which is the true number of vehicles by lane in the “measurement section.”

[0079] [Table 4]

[0080] The process of estimating the number of vehicles in each lane in the distance section (i=7 to 16) in the "measurement section 2" between camera C (7) and camera C (17) will be described. As shown in Table 4, the number of vehicles based on the cameras, KC(7,1), KC(8,1), KC(17,1), and KC(18,1), are all “6”, so the true number of vehicles, KE(i,1) (i=7~16), are all “6”. Since the number of vehicles based on the cameras, KC(7,3) and KC(8,3), are both "8," and the number of vehicles based on the cameras, KC(17,3) and KC(18,3), are both "5," it is estimated that the true number of vehicles, KE(i,3) (i=9~16), will decrease monotonically. The maximum value for the number of vehicles based on the probes, KP(i,) (i=9~16), is "15" for i=11, so the true number of vehicles, KE(11,1), can be estimated as "6", the true number of vehicles, KE(11,3), as "8", and the true number of vehicles, KE(11,2), as "1". Since the true number of vehicles, KE(11,3), can be estimated as "8", the true number of vehicles, KE(9,3) and the true number of vehicles, KE(10,3), can also be estimated as "8". Regarding the true number of vehicles, KE(i,3) (i = 12 to 16), we estimate that it simply decreases from "8" to "5". The true number of vehicles, KE(i,2), is estimated by taking into account the number of vehicles based on probe information, KP(i,). From the above, it is possible to estimate the true number of vehicles in each lane, KE(i,j), in the distance section (i=7 to 16) in "measurement section 2" between camera C(7) and camera C(17). Note that in the same manner, it is possible to estimate the true number of vehicles in each lane, KE(i,j), in "measurement section 1" and "measurement section 3".

[0081] In step S4, the number of vehicles in each lane in each of the measurement sections 1 to 3 is totaled to obtain the number of vehicles in each lane PM(k,j), and then the process proceeds to step S5. From Table 4, the number of vehicles per lane, PM(k,j), is PM(1,1) = 36, PM(1,2) = 2, PM(1,3) = 0, PM(2,1) = 60, PM(2,2) = 8, PM(2,3) = 68, PM(3,1) = 38, PM(3,2) = 4, and PM(3,3) = 34.

[0082] In step S5, function Fa [PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] is determined (once a month), and function Fb [PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] is determined (once a month), and then the process proceeds to step S6. In step S5, functions Fa and Fb for calculating the required time T(a←0) and the required time T(b←0) are calculated in advance based on past probe information, using PM(k,j), which is the number of vehicles per lane, as a parameter.

[0083] In step S6, the number of vehicles per lane, PM(k,j), is calculated every 10 seconds, and the required time T(a←0) is calculated from function Fa and the required time T(b←0) is calculated from function Fb. After notifying these, the process ends. Specifically, the process of calculating the required time T(a←0) from the function Fa and the process of calculating the required time T(b←0) from the function Fb is as follows. In the field of traffic engineering, the "Greenberg model" shown in Figure 9 is well-known as a relationship between spatial average speed and traffic density (see Figure 1.7 on page 14 of "Traffic Flow Theory" by Kuwahara Masao, published by the Japan Society of Traffic Engineers). However, it is common to simplify this model in accordance with reality, for example by converting it into a first-order linear equation as shown in Figure 10.

[0084] Here, the total number of vehicles in a jammed state (maximum) is assumed for the nine sections shown in Figure 11: [PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)]. For example, it is set to 80 vehicles in PM(2,3). The speed limit is set to 60 km / h in all sections, and vehicles can travel freely when there are no vehicles, so the image is of a linear equation as shown in Figure 10. The number of vehicles "68" in FIG. 10 is the result estimated in the above steps S3 to S4 in the first embodiment.

[0085] [Table 5]

[0086] Required time T(a←o)=Fa[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] Time required: T(a←o) = Fa(PMa) = 5.111 × PMa + 138 Here, PMa is a linear function consisting of PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), and PM(3,3).

[0087] PMa=1+ 0.1PM(1,1) +0.02PM(1,2)+0.05PM(1,3)+0.8PM(2,1)+0.2PM(2,2)+PM(2,3)+0.7PM(3,1)+0.2PM(3,2)+PM(3,3) Here, the coefficient by which each PM(i,j) is multiplied is the weight that the measurement section has on the required time.

[0088] PMa jam=1+0.1×36+0.02×60+0.05×48+0.8×60+0.2×100+80+0.7×60+0.2×100+80=298 However, PMa·jam is an extreme state of traffic congestion where the speed is 5km / hour.

[0089] Required time T(b←o)=Fb[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] Required time T(b←o)=Fb(PMb)=4.440×PMb+180 Here, PMb is a linear function consisting of PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), and PM(3,3).

[0090] PMb=1+ PM(1,1) +0.2PM(1,2)+0.8PM(1,3)+PM(2,1)+0.6PM(2,2)+PM(2,3)+PM(3,1)+0.4PM(3,2)+PM(3,3) Here, the coefficient by which each PM(i,j) is multiplied is the weight that the measurement section has on the required time.

[0091] PMb jam=1+36+0.2×60+0.8×48+60+0.6×100+80+60+0.4×100+80=467 However, PMb·jam is an extreme traffic jam condition where the speed is 5 km / h.

[0092] Note that the coefficients of PM(i,j) differ between PMa and PMb in the above (for example, PM(1,1) in PMa is "0.1", while PM(1,1) in PMb is "1.0"; see the underlined part), because the degree of influence on the required time T(a←o) and the required time T(b←o) differs.

[0093] [Table 6] From the results in Table 6, in Example 1, the required time T(a←0) calculated from the function Fa is 1,077 seconds, and the required time T(b←0) calculated from the function Fb is 1,214 seconds.

[0094] Example 2 <Prediction method based on the number of vehicles per lane, taking into account the vehicle speed by vehicle type after reinforcing probe information with a camera> This Example 2 is the same as Example 1, except that in Example 1, in order to determine the number of vehicles per lane in the measurement section, the section speed for each vehicle type is calculated to predict the number of vehicles per lane. Therefore, a description of the same configuration as Example 1 already described will be omitted.

[0095] As shown in FIG. 12, in step S21, the number of vehicles KP(i,) based on the current probe information is determined for each distance section every 10 seconds from the position and speed information from vehicles equipped with onboard devices traveling on road R in the rectification area, and then the process proceeds to step S22.

[0096] In step S22, when the destination points of all digital tachograph vehicle IDs (D-IDs) and GPS-equipped device vehicle IDs (G-IDs) present in the vehicle identification section are estimated using GPS information obtained every 10 seconds, processing proceeds to step S23.

[0097] In step S23, the number of vehicles based on the camera, KC(i,j), is determined from the number of vehicles per lane in an analyzable distance section (two distance sections) from the location C(i) where the camera is installed, and the process proceeds to step S24.

[0098] In step S24, after determining the number of ready mixed concrete trucks "Nad", the number of dump trucks "Nag", the number of container vehicle heads "Nbd", and the number of container vehicle chassis "Nbg" within the vehicle identification section within 10 seconds, processing transitions to step S25.

[0099] In step S25, when "Nbc", which is the number of container cars (complete set) present in the vehicle identification section within 10 seconds, is determined, the process proceeds to step S26. In the second embodiment, since it is necessary to accurately grasp the speed, the combination (complete set) of the container vehicle head section and the container vehicle chassis section is detected in the vehicle identification section shown in Figure 13 from multiple types of current probe information collected from the traveling vehicle. Case [a], which is the combination (complete) state of the container car head section and the container car chassis section shown in FIG. 3, is detected from among the vehicles in various states shown in FIG. "Nbd" is the number of container car heads alone, and "Nbg" is the number of container car chassis alone. Note that the chassis of a container car cannot run alone, but there is no on-board unit at the head of the container car. In FIG. 15, the case [a] is determined to be in the "complete" state.

[0100] In order to combine multiple types of probe information, it is necessary to grasp the proximity (substantially identical) of multiple moving objects. Therefore, the proximity is determined based on the "spatiotemporal distance (location information, time)" expressed by the following formula 1.

[0101] [Formula 1] TIFF0007613739000008.tif17165

[0102] Among the various types of probe information, probe D and probe G are moving forward, so probe D is always ahead of probe G. Each second of longitude and latitude is approximately 25 m. The proximity (substantially the same) is determined as the proximity of probe information whose spatiotemporal distance is positive or zero and has a minimum value.

[0103] As shown in Fig. 16, each vehicle is identified in a "vehicle identification section" (for example, a section of 300 m). In the vehicle identification section, those captured in a time period (10 seconds from t31 to t33) can be represented by "D-ID" for probe D and "G-ID" for probe G. Assume that there are five vehicles. D1+G1, D3+G3, and D4+G4 are the “set” states of [a]. G2 is in state [c] (the chassis of a container vehicle is towed by a head unit without a digital tachograph, or a dump truck). D5 is in state [b] (the head of a container vehicle alone, the head of a container vehicle towing the chassis of a container vehicle without a GPS tracking device, or a ready-mix concrete truck).

[0104] Next, consider the case of determining the "complete" state for D3+G3. To simplify the calculation, the vehicle identification section of road R shown in FIG. 16 is 170 m north of the 300 m, runs due north at 135 degrees east longitude, and point o is located at 35 degrees north latitude. The speed limit on this road R is 60 km / h = 16.7 m / s, and each vehicle is traveling at the speed limit. The southern end of this section is at latitude 34:59:53 N. That is, 170 / 25=6.8, which is 7 seconds south of point o. The calculation is made assuming that the "G3-1" probe was detected at latitude 34:59:56 North (all East longitudes are maintained at 135 degrees) at time t32 (=10:59:40). Since probe information collected before "G3-1" is limited to information collected before t32, the spatiotemporal distances between "D4-2", "D4-1", "D3-4", "D3-3", "D3-2", "D1-9", "D1-8", "D1-7", "D1-6", and "D1-5" and G3-1 were calculated using Equation 1 above and are shown in Table 7.

[0105] [Table 7] From the results in Table 7, if the discrimination standard for the time-space distance of the "complete" state when traveling at 60 km / h is 1.5 (sec·sec) or less, then the case where this condition is met is "D3-4", and D3 is recognized as being in the "complete" state. Note that the process of determining this "complete" state must be performed in real time and must be determined immediately, so if the condition is met, it will be estimated.

[0106] In step S26, the vehicle speeds VP(i,j) included in "Nad", the number of ready mixed concrete trucks in the confirmed vehicle identification section, "Nag", the number of dump trucks, "Nbd", the number of container vehicle heads, "Nbg", the number of container vehicle chassis, and "Nbc", the number of container vehicles (complete set), are calculated by setting j=3 for "Nad+Nag" and j=1 for "Nbd+Nbg-Nbc", and then processing proceeds to step S27.

[0107] In step S27, the number of vehicles in each lane is estimated based on the average speed for each lane and distance section for the measurement sections of two consecutive cameras as the true number of vehicles, KE(i,j), and then the process proceeds to step S28. Specifically, we estimate KE(i,j), which is the true number of vehicles per lane in the “measurement section”, as shown in Table 8.

[0108] [Table 8]

[0109] The process of estimating the number of vehicles in each lane in the distance section (i=7 to 16) in the "measurement section 2" between the two cameras C (7) and C (17) will be described. As shown in Table 8, the number of vehicles based on the cameras, KC(7,1), KC(8,1), KC(17,1), and KC(18,1), are all “6”, and the current speeds based on the probe information, VP(7,1), VP(8,1), VP(17,1), and VP(18,1), are all “5”, so the true number of vehicles, KE(i,1) (i=7~16), are all “6”. If the current speed VP based on the probe information is less than or equal to the moving speed in the queue VQ (e.g. "5"), the vehicle is determined to be in the queue. Basically, the true number of vehicles, KE(i,1), is calculated as a linear function of the current speed, VP(i,1), based on the probe information. Similarly, the true number of vehicles, KE(i,3), is calculated as a linear function of the current speed, VP(i,3), based on the probe information. From the above, it is possible to estimate the true number of vehicles in each lane, KE(i,j), in the distance section (i=7 to 16) in "measurement section 2" between camera C(7) and camera C(17). Note that in the same manner, it is possible to estimate the true number of vehicles in each lane, KE(i,j), in "measurement section 1" and "measurement section 3".

[0110] In step S28, the number of vehicles in each lane in each measurement section is totaled to obtain the number of vehicles in each lane PM(k,j), and then the process proceeds to step S29. From Table 8, the number of vehicles per lane, PM(k,j), is PM(1,1) = 36, PM(1,2) = 2, PM(1,3) = 0, PM(2,1) = 60, PM(2,2) = 8, PM(2,3) = 68, PM(3,1) = 38, PM(3,2) = 4, and PM(3,3) = 34.

[0111] In step S29, function Fa[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] is determined (e.g., once a month) and function Fb[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] is determined (e.g., once a month), and then the process proceeds to step S30. In step S29, functions Fa and Fb for finding the required time T(a←0) and the required time T(b←0) are found in advance using PM(k,j) as a parameter.

[0112] In step S30, the number of vehicles per lane, PM(k,j), is calculated every 10 seconds, and the required time T(a←0) is calculated from function Fa and the required time T(b←0) is calculated from function Fb. After notifying these, the process ends. Specifically, the process of calculating the required time T(a←0) from the function Fa and the process of calculating the required time T(b←0) from the function Fb is as follows. In the field of traffic engineering, the "Greenberg model" shown in Figure 9 is well-known as a relationship between spatial average speed and traffic density (see Figure 1.7 on page 14 of "Traffic Flow Theory" by Kuwahara Masao, published by the Japan Society of Traffic Engineers). However, it is common to simplify this model in accordance with reality, for example by converting it into a first-order linear equation as shown in Figure 10. The total number of vehicles (maximum) in a jam state is assumed for the nine sections shown in Figure 11 [PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)]. For example, it is set to 80 vehicles in PM(2,3). The speed limit is set to 60 km / h for all sections, and vehicles can freely travel at that speed when there are no vehicles, so it is like a linear equation for the straight line shown in Figure 10. In the second embodiment, the number of vehicles "68" in FIG. 10 is a result obtained in steps S27 to S28 above, taking into consideration the speed VP(i,j) in steps S24 to S26 above.

[0113] [Table 9]

[0114] Required time T(a←o)=Fa[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] Time required: T(a←o) = Fa(PMa) = 5.111 × PMa + 138 Here, PMa is a linear function consisting of PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), and PM(3,3).

[0115] PMa=1+ 0.1PM(1,1) +0.02PM(1,2)+0.05PM(1,3)+0.8PM(2,1)+0.2PM(2,2)+PM(2,3)+0.7PM(3,1)+0.2PM(3,2)+PM(3,3) Here, the coefficient by which each PM(i,j) is multiplied is the weight that the measurement section has on the required time.

[0116] PMa jam=1+0.1×36+0.02×60+0.05×48+0.8×60+0.2×100+80+0.7×60+0.2×100+80=298 However, PMa·jam is a traffic jam condition with a speed of 5 km / h.

[0117] Required time T(b←o)=Fb[PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), PM(3,3)] Required time T(b←o)=Fb(PMb)=4.440×PMb+180 Here, PMb is a linear function consisting of PM(1,1), PM(1,2), PM(1,3), PM(2,1), PM(2,2), PM(2,3), PM(3,1), PM(3,2), and PM(3,3).

[0118] PMb=1+ PM(1,1) +0.2PM(1,2)+0.8PM(1,3)+PM(2,1)+0.6PM(2,2)+PM(2,3)+PM(3,1)+0.4PM(3,2)+PM(3,3) Here, the coefficient by which each PM(i,j) is multiplied is the weight that the measurement section has on the required time.

[0119] PMb jam=1+36+0.2×60+0.8×48+60+0.6×100+80+60+0.4×100+80=467 However, PMb·jam is a traffic jam state where the speed is 5 km / h.

[0120] Note that the coefficients of PM(i,j) differ between PMa and PMb in the above (for example, PM(1,1) in PMa is "0.1", while PM(1,1) in PMb is "1.0"; see the underlined part), because the degree of influence on the required time T(a←o) and the required time T(b←o) differs.

[0121] [Table 10] From the results of Table 10, in Example 2, the required time T(a←0) calculated from the function Fa is 1,077 seconds, and the required time T(b←0) calculated from the function Fb is 1,214 seconds. [Explanation of symbols]

[0122] 10 Rectification area 11 General Area 100 Arrival time prediction device 101 Control section 102 Main storage 103 Auxiliary storage device 104 I / O Interface 105 Communication Interface 106 Input Devices 107 Output Device 108 Display device 109 System Bus 120 Communication function unit 130 Input function unit 140 Output function section 150 Display function section 160 Memory function unit 170 Control Function Section 171 Arrival Time Prediction Department 172 Discrimination part

Claims

1. A method for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of the first destination point and the second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, the method comprising: Using a computer, estimate the number of vehicles per lane between the first detection device and the second detection device based on the number of vehicles detected per lane by a first detection device installed on the road and the number of vehicles detected per lane by a second detection device installed next to the first detection device; An arrival time predicting method comprising: an arrival time predicting step of predicting, using a computer, an arrival time for each lane to at least one of the first destination point and the second destination point.

2. The arrival time prediction for each lane in the arrival time prediction step is 2. The arrival time prediction method according to claim 1, further comprising: deriving the number of vehicles for each lane between the first detection device and the second detection device based on current probe information for each lane; and predicting the arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles for each lane and past probe information for each lane.

3. The arrival time prediction for each lane in the arrival time prediction step is 2. The arrival time prediction method according to claim 1, further comprising: deriving the number of vehicles and the traveling speed for each lane between the first detection device and the second detection device based on current probe information for each lane; and predicting the arrival time for each lane to at least one of the first destination point and the second destination point based on the number of vehicles and the traveling speed for each lane and past probe information for each lane.

4. In the arrival time prediction step, when deriving a travel speed based on current probe information for each lane, the current probe information exists in a plurality of pieces, Included in the plurality of pieces of probe information, Location information X1 at time X in the first probe information is substantially identical to location information X2 at a time substantially the same as the time X in second probe information different from the first probe information, and When location information Y1 at a time Y sampled next to the time X in the first probe information and location information Y2 at a time substantially the same as the time Y in the second probe information are substantially identical, The method for predicting an arrival time according to claim 3 , further comprising determining that the first probe information and the second probe information are related to the same moving object.

5. 1. A device for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of the first destination point and the second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; 2. An arrival time predicting device comprising: an arrival time predicting means for predicting an arrival time for each lane to at least one of the first destination point and the second destination point.

6. A program for calculating a required time to at least one of a first destination point and a second destination point for each lane of a road leading to at least one of a first destination point and a second destination point, the required time being within a predetermined distance from a first destination point, the second destination point also being present within the said area, Based on the number of vehicles detected for each lane by a first detection device installed on the road and the number of vehicles detected for each lane by a second detection device installed next to the first detection device, estimating the number of vehicles for each lane between the first detection device and the second detection device; An arrival time prediction program that causes a computer to perform an arrival time prediction process for predicting an arrival time for each lane to at least one of the first destination point and the second destination point.

Images