JP5768526B2 - Traffic jam prediction device and traffic jam forecast data - Google Patents

Description

  The present invention relates to a traffic jam prediction device and traffic jam forecast data.

  Conventionally, Patent Documents 1 and 2 describe techniques for predicting the degree of congestion (degree of congestion) on a road. In patent document 1, the traffic jam prediction data of each road for each time created by statistically processing past traffic information is stored in advance, and based on this traffic jam prediction information, a point far from the current position is stored. Predicts the degree of future traffic in

  Further, in Patent Document 2, a certain traffic volume is set for a certain link, the traffic volume is distributed at a branching portion of the road, and the simulated travel is performed while partially erasing the traffic volume at the link. The passing traffic volume is calculated.

JP 2007-232727 A JP 2002-049984 A

  However, in the technique of Patent Document 1, as shown in FIG. 11, only a target link can be predicted such that a certain link is statistically vacant at a specific date and time. The degree of congestion is not predicted in consideration of temporary and sudden traffic conditions. For this reason, if a large number of vehicles flow into the link because of a special event in the vicinity of another link connected to the link, the link will be overcrowded beyond the statistical information forecast. May end up. That is, in the method of Patent Document 1, the degree of congestion is not predicted in consideration of the temporary and sudden traffic situation that is actually occurring.

  Moreover, although the technique of patent document 2 considers the temporary and sudden traffic volume currently occurring, according to inventors' examination, it is because traffic volume and the degree of congestion do not necessarily correspond. There's a problem. Even with the same traffic volume, the road may or may not be congested due to external factors such as road gradients and traffic lights. For example, as shown in FIG. 12, even if the traffic volume per unit time is constant, the traffic volume is more than a certain threshold, and traffic jam occurs due to conditions such as uphill and traffic lights. Therefore, in order to accurately predict traffic congestion using the passing traffic volume, it is necessary to calculate information such as gradients and traffic lights using external parameters without excess or deficiency. Become very large.

  In view of the above points, the present invention can reflect a temporary and sudden traffic situation that is actually occurring in a traffic jam prediction device that predicts the traffic jam level of a link, and easily reflects the influence other than the traffic volume in the traffic jam level. The purpose is to provide such technology.

In order to achieve the above object, the inventions according to claims 1, 3, and 5 provide various types of congestion of the entry link for each factor, for each node, and for each combination of all entry links and exit links for the node. Based on the traffic jam prediction data acquired by the acquisition unit (15) and the traffic jam prediction data acquired by the acquisition unit (15). Predicting means (17) for predicting the degree of congestion of the link, and the predicting means (17) is a range limiting means (105) for limiting the congestion prediction range that is the target range of the link for predicting the degree of congestion. The actual congestion degree at the first time point of the first link is acquired from the outside, the factor that conforms to the first time point, the approach link corresponding to the first link, and the acquired first Rin By applying the actual traffic congestion degree at the first time point to the traffic congestion prediction data, the second link entering the second link in the traffic congestion prediction range from the first link through the first node. A traffic jam prediction device comprising: a traffic jam level acquisition means (120 to 145) for acquiring the traffic jam level of the second link at a time point.

  In this way, as the traffic jam prediction data, for each factor, for each node, and for each combination of all the incoming links and outgoing links for the relevant node, various predictions of the outgoing link resulting from various congestion levels of the incoming link. By using the traffic jam prediction data including the traffic jam degree, the above object can be achieved.

  Specifically, the actual congestion degree at the first time point of the first link is acquired from the outside, the factor that conforms to the first time point, the approach link corresponding to the first link, and the acquired first The actual traffic congestion degree at the first time point of the link is applied to the traffic congestion prediction data.

  Then, the second link (exit link) that can enter from the first link (entrance link) via the first node can be specified on the structure of the traffic jam prediction data as described above, and further, the first link of the first link can be specified. Various predicted congestion levels for the second link resulting from various congestion levels can be identified, and the second time of the second time resulting from the actual congestion level at the first time of the first link can be identified. It is possible to obtain the predicted congestion level of the link.

  Such a technique actually occurs in that the actual congestion degree at the first time point of the first link is acquired from the outside, and the predicted congestion of the second link is calculated based on the acquired congestion degree. It can reflect temporary and sudden traffic conditions. Moreover, instead of the conventional traffic volume simulation, the causal relationship between the traffic congestion degree of the first link and the traffic congestion degree of the second link in the traffic congestion prediction data is used directly without passing through the traffic volume calculation. The congestion degree of the second link is calculated. Therefore, the influence other than the traffic volume can be easily reflected in the congestion degree.

  As a result, as shown in FIG. 13 (a), when the vehicle travels by calculating the route indicated by the bold line, it will encounter traffic jam on the route as shown in FIG. 13 (b). The likelihood of an event occurring is reduced.

The invention according to claim 1, wherein the predicted congestion degree obtaining means (120-145) further, the second matches the time factors, entering link corresponding to the second link, and was obtained By applying the predicted traffic congestion degree at the second time point of the second link to the traffic congestion prediction data, the second link enters the third link in the traffic congestion prediction range via the second node. The first predicted congestion degree of the third link at the third time point is acquired, and when the fourth link is connected to the second node, the third link time from the third time point A factor that matches a fourth time point that is back by the transit time of four links, an approach link corresponding to the fourth link, and a predetermined congestion degree of the fourth link are applied to the congestion prediction data. And the fourth phosphorus The second predicted congestion degree of the third link at the third time point entering the third link from the first node through the second node is obtained, and the first prediction of the third link is obtained. Of the congestion degree and the second predicted congestion degree of the third link, the higher congestion degree is set as the predicted congestion degree of the third link.

  In this way, when the second link and the fourth link are connected to the third link via the same second node, the third link generated as a result of the congestion degree of the second link If the traffic congestion degree and the traffic congestion degree of the third link resulting from the traffic congestion degree of the fourth link are different from each other, the higher traffic congestion degree is adopted instead of adding together like a traffic flow. This is because the congestion degree prediction data of the third link resulting from the congestion degree of the second link is created as data reflecting the influence of inflows from other links to the third link, etc. This is because the congestion degree prediction data of the third link generated as a result of the congestion degree of the link 4 is created as data reflecting the influence of inflow from other links to the third link.

  Claims2, 3The invention described in,The predicted congestion degree acquisition means (120 to 145) further includes a factor that matches the first time point, an approach link corresponding to the first link, and the first time point of the acquired first link. By applying the actual traffic congestion degree to the traffic congestion prediction data, the traffic congestion prediction range within the traffic congestion prediction range from the first link through the first node.Re N PAt the second time of enteringLink PA predicted traffic congestion degree of the second time point, and a factor that conforms to the second time point,Link PThe approach link corresponding toLink PBy applying the predicted traffic congestion degree at the second time point to the traffic congestion prediction data,Link PFromNode AThrough the traffic jam forecast rangeLink QEnterTime VIn the aboveLink QThe first predicted congestion degree of theNode AWithin the traffic jam forecast rangeLink RWhen connectedTime VDifferent fromTime point WFactors that conform to the aboveLink RAnd the entry link corresponding toTime point WIn the aboveLink RBy applying the degree of traffic congestion to the traffic congestion prediction data,Link RFrom the aboveNode AThrough the aboveLink QEnterTime XIn the aboveLink QTo obtain the second predicted congestion degree of Link QSaid first predicted congestion degree and saidLink QOf the second predicted congestion levels, the higher congestion level isLink QIt is characterized by the predicted traffic congestion degree.

  In this way, when the fifth link and the seventh link are connected to the sixth link via the same third node, the sixth link generated as a result of the degree of congestion of the fifth link When the congestion level and the congestion level of the sixth link resulting from the congestion level of the seventh link are different from each other, the higher congestion level is adopted instead of adding together like a traffic flow. This is because the congestion degree prediction data of the seventh link resulting from the degree of congestion of the fifth link is created as data reflecting the influence of inflow from other links to the sixth link, etc. This is because the traffic congestion degree prediction data of the sixth link generated as a result of the traffic congestion degree of the link 7 is created as data reflecting the influence of the inflow from other links to the sixth link.

Further, the invention according to claim 4 is the traffic jam prediction device according to claim 2 or 3 , wherein the predicted traffic jam level acquisition means (120 to 145) further includes the first predicted traffic jam level of the link Q and , among the second prediction congestion degree of the links Q, the higher degree of congestion, as well as the predicted congestion degree of the links Q, the earlier of the point V and the point W to the link Q and entry time, also factor compatible with entry time into the link Q, entering link corresponding to the link Q, and the predicted traffic jam degree of the entry time of the acquired links Q, fitted to the traffic jam prediction data it is, at the time of entering the link S in the traffic jam prediction range through the node B from the link Q, and acquires the prediction congestion degree of the links S. By doing so, it is possible to unify the time of entering the link and advance the calculation of the predicted traffic congestion degree of the node.

Further, the invention according to claim 5, congestion prediction data and the acquisition means (15) obtains the hierarchy has a multilevel structure as the links and nodes are thinned out becomes higher, each factor, each layer For each node, and for each combination of all incoming and outgoing links for that node, various predicted congestion levels for that outgoing link resulting from different traffic levels for that incoming link are included, and said predicted traffic congestion The degree acquisition means (120 to 145) includes a factor that conforms to the first time point, a highest hierarchy among the hierarchies including both the first link and the second link, and the first By applying the incoming link corresponding to the link and the actual traffic congestion degree at the first time point of the acquired first link to the traffic congestion prediction data, In the second time of entering the second link in the first of the traffic jam prediction range through the node obtains the predicted congestion degree of the second link, characterized by having a.

  Thus, by using the traffic jam prediction data divided into hierarchies, it is possible to reduce the processing load for calculating the predicted traffic jam degree.

  In addition, the code | symbol in the bracket | parenthesis in the said and the claim shows the correspondence of the term described in the claim, and the concrete thing etc. which illustrate the said term described in embodiment mentioned later. .

1 is a configuration diagram of a navigation device 1 according to an embodiment of the present invention. It is a figure which shows the structural example of traffic congestion prediction data. It is a flowchart of a traffic jam prediction process. It is a figure of the links L1-L6 from the starting point Po to the destination Pd. It is a table | surface showing the relationship of the traffic congestion degree of the entrance link L1 and the exit links L2 and L5 in the time T1 and the node N1. It is a table | surface showing the relationship of the traffic congestion degree of the entrance link L2 and the exit link L3 in the time T1 + T2 and the node N2. It is a table | surface showing the relationship of the traffic congestion degree of the approach link L4 in the node N2, and the exit link L3. It is a table | surface showing the relationship of the traffic congestion degree of the entrance link L5 and the exit link L6 in the time T1 + T2 and the node N3. It is a table | surface showing the relationship of the traffic congestion degree of the entrance link L3 and the exit link L6 in the time T1 + T2 + T3 and the node N3. It is a figure showing the calculated optimal path | route. It is a figure which shows the traffic congestion prediction data used by patent document 1. FIG. It is a figure which illustrates the difference of traffic volume and congestion degree. It is a figure which shows the example which runs along the calculated route and encounters a traffic jam.

  Hereinafter, an embodiment of the present invention will be described. FIG. 1 shows an example of a navigation device 1 (corresponding to an example of a traffic jam prediction device) according to the present embodiment. This navigation device may be mounted on the vehicle, or the user may be able to carry it outside the vehicle.

  The navigation device 1 includes a position detector 11, an image display device 12, an operation unit 13, a speaker 14, a communication unit 15, and a control circuit 17. The position detector 11 is a sensor that outputs a signal for specifying the current position of the vehicle to the control circuit 17 such as a GPS receiver, a vehicle speed sensor, an acceleration sensor, and a gyro sensor. The image display device 12 is a device such as a liquid crystal display that displays characters and images to the user under the control of the control circuit 17. The operation unit 13 is a device that receives a user operation and outputs a signal corresponding to the received operation to the control circuit 17.

  The communication unit 15 wirelessly communicates with an external traffic jam prediction data distribution server and receives traffic jam prediction data from the traffic jam prediction data distribution server. The received traffic jam prediction data is recorded in the storage medium of the navigation device 1. Details of the traffic jam prediction data will be described later. Further, the communication unit 15 receives traffic information data including a degree of congestion indicating the current degree of congestion of each link from a traffic information distribution device (for example, VICS transmission station, VICS beacon, probe vehicle information distribution server) outside the vehicle. It is supposed to be.

  The map data acquisition unit 16 is a device that acquires map data under the control of the control circuit 17 and outputs the map data to the control circuit 17. The map data acquisition method may be a method of reading map data from a storage medium (HDD, flash memory, DVD medium, etc.) of the navigation device 1 or providing a map outside the vehicle 1 by wireless communication using the communication unit 15. A method of downloading map data from a server may be used.

  The map data acquired by the map data acquisition unit 16 includes road shape data for map display, facility data for destination search, road network data for guidance route calculation, and the like. The road network data includes information such as connection relations between links and nodes, link costs, and the like.

  Further, the road network data of the present embodiment has a hierarchical structure having a plurality of hierarchies with different degrees of detail of the link structure. Of the multiple hierarchies, the data of the lowest hierarchy includes link and node data for all roads, and the higher the hierarchy is, the thinner the data of links and nodes included in the data of that hierarchy. It is burned. For example, if the road network data has two layers, an upper layer and a lower layer, the lower layer data includes link and node data for both main roads and non-main roads. The data includes only data of links and nodes that constitute only the main road out of the main road and the non-main road.

  The control circuit 17 is a device (e.g., a microcontroller) that includes a CPU, RAM, ROM, flash memory, and the like, and the CPU executes a program recorded in the ROM or flash memory so that various processes are realized. It has become.

  Specific processing performed by the control circuit 17 executing the program includes current position specifying processing, map display processing, guidance route calculation processing, route guidance processing, and the like. The current position specifying process is a process for specifying the current position and direction of the vehicle based on a signal from the position detector 11 using a known technique such as map matching.

  The map display process is a process for causing the image display device 12 to display a map of a specific area such as the vicinity of the current position of the vehicle. At this time, information used for map display is acquired from road shape data.

  The guidance route calculation process is a process of receiving an input of a destination by the user from the operation unit 13 and calculating an optimum guidance route from the current position to the destination.

  The route guidance process is a process that guides the travel along the calculated guidance route, sequentially monitoring the positional relationship between the guidance route and the vehicle position, and before a guidance point such as a right-left turn intersection on the guidance route. When the host vehicle arrives, a guidance voice instructing a right turn, a left turn, etc. is output to the speaker 14, and an enlarged view of the guidance point is displayed on the image display device 12, thereby driving the vehicle along the guidance route. It is a process to guide.

  Details of the guidance route calculation process will be described below. In the present embodiment, in the guidance route calculation process, the congestion degree of each link is first predicted by executing a congestion prediction process, and a cost corresponding to the degree of congestion is allocated to each link, and then the road network data is used. The route with the lowest total cost is adopted as the guidance route by the Dijkstra method. In this traffic jam prediction process for predicting the traffic jam degree of each link, the traffic jam forecast data acquired from the communication unit 15 is used. FIG. 2 shows a configuration example of this traffic jam prediction data.

  As shown in this figure, the traffic jam prediction data includes, for each factor (factors relating to time, day of the week, etc., the same applies hereinafter), each layer, each node included in the layer, and all incoming links to the node ( For each combination of the link included in the hierarchy) and the exit link (limited to that included in the hierarchy), it includes various predicted congestion levels of the exit link resulting from various congestion levels of the entry link. . In addition, the traffic jam prediction data includes, for each factor, each hierarchy, each node, and each incoming link to the node, various traffic levels of the incoming link and the passing time of the incoming link at the traffic level. Yes.

  More specifically, the traffic jam prediction data has a plurality of factor records R1 each corresponding to one factor, and each factor record R1 is factor data representing a factor corresponding to the factor record R1. 21 and a plurality of hierarchy records R2 each corresponding to one hierarchy. In addition, a hierarchy is a hierarchy equivalent to the above-mentioned road network data. Therefore, the traffic jam prediction data also has a multi-hierarchy structure in which links and nodes are thinned out as the hierarchy becomes higher.

  Each hierarchical record R2 has data 22 of the layer number corresponding to the hierarchical record R2, and each of the hierarchical records R2 is limited to one node (however, the road network data is included in the hierarchical level of the hierarchical record R2). ) Has a plurality of node records R3 corresponding to the number of nodes included in the hierarchy of the hierarchy record R2.

  Each node record R3 has node number data 23 corresponding to the node record R3, and each node record R3 corresponds to one entry link (however, only a link entering the node of the node record R3). There are one or a plurality of incoming link records R4 corresponding to the number of links entering the node of the node record R3.

  Each entry link record R4 includes link number data 24 of the entry link corresponding to the entry link record R4. Further, each entry link record R4 includes a congestion degree record R5 corresponding to one congestion degree. Only have more than one. The degree of congestion is, for example, a numerical value from 1 to 6 that increases as the degree of congestion increases (specifically 1: very vacant, 2: empty, 3: light congestion, 4: heavy congestion, 5 : Light traffic jam, 6: heavy traffic jam).

  Each congestion degree record R5 includes the data 25 indicating the congestion degree corresponding to the congestion degree record R5 and the passage required to travel the approach link of the approach link record R4 including the congestion degree record R5 with the congestion degree. Further, each of the exit link records R6 corresponding to one exit link (however, limited to the link exiting from the node of the node record R3 including the congestion degree record R5), One or a plurality of links are provided corresponding to the number of links leaving the node of the node record R3 including the congestion degree record R5.

  Further, each exit link record R6 has data 27 of the link number of the exit link corresponding to the exit link record R6, and when entering the entrance link of the entrance link record R4 including the exit link record R6 Occurs in the exit link when entering the exit link from the entrance link as a result of the congestion level of the entrance link being the congestion level of the congestion level record R5 including the exit link record R6 Then, the data 28 indicating the predicted traffic congestion degree is included.

  The relationship between the degree of congestion of the approach link when entering an entry link and the predicted degree of congestion of the exit link when entering the exit link from the entry link via a node is statistically determined beforehand by field surveys, etc. After determination, traffic jam prediction data is created and stored in the traffic jam prediction data distribution server.

  Here, the hierarchy of traffic jam prediction data will be described. Similar to the road network data hierarchy, the congestion forecast data hierarchy includes the link and node data for all roads in the lowest hierarchy data, and the higher the hierarchy, Data of links and nodes included in the data of the hierarchy is thinned out.

  For example, the road network has four layers, the lowest layer 0 includes links and nodes to be searched for all routes, the layer 1 includes only the links of the major local roads and higher and the nodes that intersect, and the layer 2 includes the prefectural roads. Only the links with the above ranks and the nodes where they intersect are included, the hierarchy 3 includes only the links with the ranks higher than the national road and the nodes with which they intersect, and the hierarchy 4 includes only the highways and the nodes with which they intersect.

  Each hierarchical record R2 of a certain hierarchy has a node record R3 only for the nodes included in that hierarchy, has an incoming link record R4 only for the incoming links included in that hierarchy, and an outgoing link included in that hierarchy. Only the exit link record R6.

  When receiving the input of the destination in the guidance route calculation process, the control circuit 17 first determines a search route range from the current position to the destination and various search conditions by a known method. Subsequently, the control circuit 17 receives traffic information data including information on the degree of congestion of each link at the present time (that is, time T1 at the time of route search) from the traffic information distribution device via the communication unit 15. The current information on the degree of congestion of each link is adopted as an initial value when the congestion is predicted.

  Subsequently, the control circuit 17 starts execution of the traffic jam prediction process shown in FIG. 3, and in step 105, in order to shorten the calculation time of the entire traffic jam prediction process, the target range of the link for calculating the traffic jam degree (hereinafter referred to as traffic jam prediction). Range). The method for limiting the traffic jam prediction range may be, for example, all links included in the above-described search route range, or the total cost of the guidance route from the current position to the destination without performing traffic jam prediction. A plurality (for example, six) of the lowest ones may be extracted, and the extracted route and its surrounding links may be used as the traffic jam prediction range. In this case, as a link around the extracted route, for example, the range surrounded by the extracted route may be expanded so that the area increases by 10%, and the link within the range newly included by the expansion may be used. .

  For example, as an example, as shown in FIG. 4, it is assumed that links L2, L3, L5, and L6 from the departure point (current position) Po to the destination Pd are in the traffic jam prediction range. In addition, since the vehicle is already at the end Po of the link L1, it is sufficient to adopt the current congestion level and it is not a target for predicting the congestion level. In addition, it is assumed that the above extraction route is a route that travels along the links L1, L2, L5, and L6 in this order and a route that travels along the links L1, L3, and L6 in this order.

  In this example, it is assumed that the calculated links L2, L3, L5, and L6 are all included in only the lowest hierarchy, but the links included in the higher hierarchy among the links L1 to L6. There may be. It is assumed that the congestion levels of the links L1 to L6 at the current time T1 received via the communication unit 15 are the congestion level 5 for the link L1 and the congestion level 2 for the links L2 to L6.

  Subsequently, in step 110, the level of the traffic congestion degree calculation target is calculated. The same level as the hierarchy that is calculated when the route calculation without considering the congestion degree is performed in advance may be used as the congestion level calculation target hierarchy. In this example, the links L2, L3, L5, and L6 are all assumed to be the lowest hierarchy data, so the hierarchy for calculating the degree of congestion is only the lowest hierarchy, and the hierarchy record R3 used in the congestion prediction data. Is only the hierarchy record R3 of the lowest hierarchy.

  However, in the case of multi-layer data, when the first layer is used for the links L1 and L6 and the second layer is used for the links L2, L3, and L5 in the prior route calculation calculation, the link L1, L6 is calculated in the first hierarchy, and links L2, L3, and L5 are calculated in the second hierarchy. Subsequently, in step 115, the congestion degree prediction of the links L1 to L6 is started from the own vehicle position node Po where the own vehicle is currently located. Subsequently, in step 120, it is determined whether or not there is a link for which the congestion degree is not calculated among the congestion degree calculation target links (links within the congestion prediction range) L2, L3, L5, and L6.

  At first, there is a link for which the degree of congestion is not calculated. Therefore, it is determined that there is “Yes”, and the process proceeds to step 125 to determine a target link for calculating the degree of congestion. The order of determination is basically the order close to the link L1 where the host vehicle is currently located. Accordingly, first, of the links L2 and L5 that enter from the link L1 (first link) connected to the vehicle position node Po through one node N1 (corresponding to an example of the first node), the link L2 ( This corresponds to an example of the second link).

  Subsequently, at step 130, the arrival time of the host vehicle at the target link L2 is estimated. As for the link L1, since the host vehicle has already arrived at the end Po, the estimated time is the time obtained by adding the transit time T2 of the link L1 to the current time T1 (corresponding to an example of the first time point). T1 + T2 (corresponding to an example of the second time point).

  Here, the passage time T2 is determined using the traffic jam prediction data. Specifically, a factor that conforms to the current time T1, a hierarchy corresponding to the link L1 where the vehicle is present (the lowest hierarchy), an entry link corresponding to the link L1 where the vehicle is present, and a node N1 entering from the link L1 By applying the current traffic congestion degree of the link L1 (5 in this example) to the traffic congestion prediction data, one traffic congestion degree record R5 corresponding to them is determined, so the transit time t corresponding to the traffic congestion degree record R5 Is determined. Therefore, the passage time t is adopted as the passage time T2.

  Subsequently, in step 135, it is determined whether or not there is a link for which the degree of congestion is not calculated among the external links that enter the target link L2 from the node N1 that is the starting point of the target link L2. Here, the external link means a link that is not in the traffic jam prediction range. In this example, since there is no external link entering the target link L2, it is determined that there is no link, and the process proceeds to step 145.

  In step 145, the predicted congestion degree of the target link L2 is calculated. Specifically, the factor that matches the time T1, the lowest hierarchy, the incoming link corresponding to the link L1, and the actual congestion level at the time T1 of the link L1 (5 in this example) are applied to the congestion prediction data. Thus, one congestion degree record R5 corresponding to them is determined, and one exit link record R6 of the link L2 included in the congestion degree record R5 is also determined. Therefore, the predicted congestion degree corresponding to the exit link record R6 is 1. Determined.

  FIG. 5 is a table in which the correspondence relationship between various traffic congestion levels of the incoming link L1 and traffic congestion levels of the exit links L2 and L5 generated as a result of the traffic congestion level at the time T1, the lower layer, and the node N1 is extracted from the traffic congestion prediction data. Show.

  As shown in this table, when the traffic congestion level of the incoming link L1 is 5 at time T1, the predicted traffic congestion level of the outgoing link L2 resulting therefrom is 6. Therefore, the predicted congestion degree of the link L2 at time T1 + T2 is determined to be 6.

  As described above, when the predicted congestion level of the link L2 at the time T1 + T2 is determined, the process returns to Step 120, and if there is a link for which the congestion level is not calculated again in the links L2, L3, L5, and L6 in the congestion prediction range. In step 125, the link L5 is set as the target link.

  In step 130, the arrival time of the host vehicle at the target link L5 is estimated. Similar to the link L2 connected to the same node N1, this arrival time is a time T1 + T2 (an example of the second time point) obtained by adding the transit time T2 of the link L1 to the current time T1 (corresponding to an example of the first time point). Equivalent).

  Subsequently, in step 135, it is determined that there is no link for which the degree of congestion is not calculated among the external links entering the target link L5 from the node N1 that is the starting point of the target link L5, and the process proceeds to step 145. The predicted congestion degree of the link L5 is calculated. Specifically, the factor that matches the time T1, the lowest hierarchy, the incoming link corresponding to the link L1, and the actual congestion level at the time T1 of the link L1 (5 in this example) are applied to the congestion prediction data. Thus, as shown in FIG. 5, one predicted traffic congestion degree is determined. Specifically, as shown in FIG. 5, when the traffic congestion degree of the incoming link L1 is 5 at time T1, the predicted traffic congestion degree of the outgoing link L5 resulting therefrom is 2. Therefore, the predicted congestion degree of the link L5 at time T1 + T2 is determined to be 2.

  As described above, when the predicted congestion level of the link L5 at the time T1 + T2 is determined, the process returns to Step 120, and there is a link for which the congestion level is not calculated again in the links L2, L3, L5, and L6 in the congestion prediction range. In step 125, the link L3 is set as the target link.

  Thereafter, the process returns to step 120, and it is determined again that there is a link for which the congestion degree is not calculated in the links L2, L3, L5, and L6 in the congestion prediction range. In step 125, the link L3 is set as the target link.

  In step 130, the arrival time of the host vehicle at the target link L3 is estimated. This arrival time is time T1 + T2 + T3 (corresponding to an example of the third time point) obtained by adding the transit time T3 of the link L3 to the arrival time T1 + T2 (corresponding to an example of the second time point) to the link L2. .

  Here, the passage time T2 is determined using the traffic jam prediction data. Specifically, the factor that matches the arrival time T1 + T2 of the link L2 that enters the link L3 via the node N2, the hierarchy corresponding to the link L2 (the lowest hierarchy), the entry link corresponding to the link L2, the link By applying the traffic congestion level of the node N2 entering from L2 and the current link L2 (6 in this example) to the traffic congestion prediction data, one corresponding traffic congestion level record R5 is determined. One transit time t corresponding to is determined. Therefore, the passage time t is adopted as the passage time T3.

  Subsequently, in step 135, it is determined whether there is a link for which the degree of congestion is not calculated among the external links that enter the target link L3 from the node N2 that is the starting point of the target link L3. Here, since the link L4 (corresponding to an example of the fourth link) corresponds to the external link, it is determined as “present” and the process proceeds to step 140.

  In step 140, one of the external links (here, link L4) is extracted, and the degree of congestion of the extracted link L4 is calculated. Here, in order to calculate the congestion level of the link L3 at the time T1 + T2 + T3 calculated in the immediately preceding step 130, it is desired to calculate the congestion level of the link L4. The congestion degree of the link L4 at (corresponding to an example of the fourth time point) is calculated.

  Here, the transit time T4 to be used is a factor suitable for the time T1 + T2 + T3, the lowest layer, the node N2, the incoming link L4, and the congestion level “2” of the link L4 received from the communication unit 15 at the current time T1. Adopt the transit time in the record of the corresponding congestion level.

  Further, it is assumed that the congestion degree of the link L4 at the time T1 + T2 + T3-T4 is the same as the congestion degree “2” of the link L4 at the current time T1 received from the communication unit 15. This is because the inflow traffic cannot be estimated because the link L4 is a link outside the prediction range. If the link L4 is also within the predicted range, and another link of the predicted range continues to the opposite end of the node N2 of the link L4, the congestion level at the time T1 of the link in the same hierarchy as this other link is calculated. Based on the prediction data, the predicted congestion degree of other links and link L4 is also calculated.

  After step 140, the process returns to step 135 again, and it is determined that there is no link whose traffic congestion degree has not been calculated among the external links entering the target link L3 from the node N2 at the starting point of the target link L3. move on.

  In step 145, the predicted congestion degree of the target link L3 is calculated. However, the target link L3 enters both the link L2 and the external link L4 via the node N2. Therefore, here, the predicted congestion degree of the target link L3 resulting from the congestion degree of the link L2 is calculated as the first predicted congestion degree, and the predicted congestion degree of the target link L3 resulting from the congestion degree of the link L4 is calculated as the second congestion degree. Calculated as the predicted congestion level.

  The calculation process of the 1st prediction congestion degree is as follows. By applying the factor that conforms to the time T1 + T2, the lowest hierarchy, the incoming link corresponding to the link L2, and the predicted traffic congestion level at the time T1 + T2 of the link L2 (6 in this example) to the traffic congestion prediction data Since one traffic congestion level record R5 is determined and one outgoing link record R6 of the link L3 included in the traffic congestion level record R5 is also determined, one predicted traffic congestion level corresponding to the outgoing link record R6 is determined.

  FIG. 6 shows a table in which the correspondence relationship between various congestion levels of the entrance link L2 and the congestion levels of the exit link L3 resulting from the congestion level at the time T1 + T2, the lower layer, and the node N2 is extracted from the congestion prediction data. As shown in this table, when the traffic congestion level of the approach link L2 is 6 at time T1 + T2, the first predicted traffic congestion level at time T1 + T2 + T3 of the exit link L3 resulting therefrom is 2.

  The calculation process of the second predicted congestion level is as follows. By applying the factor suitable for the time T1 + T2 + T3-T4, the lowest hierarchy, the incoming link corresponding to the link L4, and the congestion level (2 in this example) at the time T1 + T2 + T3-T4 of the link L4 to the congestion prediction data, Since one congestion degree record R5 corresponding to them is determined and one exit link record R6 of the link L3 included in the congestion degree record R5 is also determined, one predicted congestion degree corresponding to the exit link record R6 is determined.

  FIG. 7 is a table in which the correspondence relationship between various congestion levels of the entrance link L4 and the congestion level of the exit link L3 resulting from the congestion level at the time T1 + T2 + T3-T4, the lower layer, and the node N2 is extracted from the congestion prediction data. Show. As shown in this table, when the traffic congestion level of the incoming link L4 is 2 at time T1 + T2 + T3-T4, the second predicted traffic congestion level at the time T1 + T2 + T3 of the outgoing link L3 resulting therefrom is 3.

  Here, the first predicted congestion level is 2, the second predicted congestion level is 3, and the two are different. However, the first and second predicted congestion levels are added together like the traffic volume. is not. This is because the traffic congestion degree prediction data of the exit link L3 generated as a result of the entrance link L2 is based on statistics, and is therefore reflected for vehicles entering the link L3 from the link L4. Of the first and second predicted traffic congestion levels, the one with the higher congestion level (that is, the larger value) (the traffic congestion level 3) is adopted as the predicted traffic congestion level of the link L3 at time T1 + T2 + T3. It is done.

  After step 145, the process returns to step 120, and it is determined again that there is a link for which the congestion degree is not calculated in the links L2, L3, L5, and L6 in the congestion prediction range. In step 125, the link L6 is targeted. A link.

In step 130, the arrival time (entry time) of the host vehicle to the target link L6 is estimated. However, since the link N3 at the start point of the target link L6 is connected to the link L3 and L5 within the traffic jam prediction range, the time when the link L3 enters the link L6 (ie, the entry time) and the link Two times, that is, a time point when entering the link L6 from L5 (entry time) are calculated.

The entry time from the link L3 to the link L6 is a time T1 + T2 + T3 + T5 (corresponding to the time point X) obtained by adding the transit time T5 of the link L3 to the time T1 + T2 + T3 (corresponding to the time point W) entering the link L3 from the link L2. . Further, the entry time from the link L5 to the link L6 is a time T1 + T2 + T6 (corresponding to the time point V) obtained by adding the transit time T6 of the link L5 to the time T1 + T2 entering the link L5 from the link L1.

  Here, the passage times T5 and T6 are determined using the traffic jam prediction data. Specifically, with regard to the transit time T5 of the link L3, the factors suitable for the time T1 + T2 + T3, the lowest hierarchy, the incoming link corresponding to the link L3, the node N3 entering from the link L3, the current traffic congestion of the link L3 By applying the degree (3 in this example) to the traffic jam prediction data, one traffic jam degree record R5 corresponding to them is determined, so that one passage time t corresponding to the traffic jam degree record R5 is determined. Therefore, the passage time t is adopted as the passage time T5. As for the transit time T6 of the link L5, the factor that matches the time T1 + T2, the lowest hierarchy, the incoming link corresponding to the link L5, the node N3 entering from the link L5, the current congestion level of the link L5 (this In the example, 2) is applied to the traffic jam prediction data, so that one traffic jam degree record R5 corresponding to them is determined. Therefore, one passage time t corresponding to the traffic jam degree record R5 is determined. Therefore, the passage time t is adopted as the passage time T6.

  In this case, the time T1 + T2 + T3 + T5 and the time T1 + T2 + T6 may or may not match. In the case of coincidence, the coincidence time may be set as the entry time to the target link L6. However, in the case of coincidence, the earlier one of the two times T1 + T2 + T3 + T5 and time T1 + T2 + T6 is adopted. This is because there is a high possibility that the route with the earlier time to enter the target link 6 is selected as the optimum route for route guidance. Here, it is assumed that the time T1 + T2 + T6 is earlier. Therefore, the time T1 + T2 + T6 becomes the entry time to the target link L6.

  Subsequently, the process proceeds to step 135, and since there is no external link entering the target link L6 from the node N3 that is the starting point of the target link L6, it is determined that there is no external link for which the degree of congestion has not been calculated, and the process proceeds to step 145.

In step 145, the predicted congestion degree of the target link L6 (corresponding to an example of the sixth link and link Q ) is calculated. However, the target link L6, also a link L5 via the node N3 links (corresponding to one example of the fifth links and P) L3 (corresponding to an example of the seventh link Contact and links R) It is designed to enter. Therefore, here, the predicted congestion degree of the target link L6 resulting from the congestion degree of the link L5 is calculated as the first predicted congestion degree, and the predicted congestion degree of the target link L6 resulting from the congestion degree of the link L3 is calculated as the second congestion degree. Calculated as the predicted congestion level.

  The calculation process of the 1st prediction congestion degree is as follows. By applying the factor that conforms to the time T1 + T2, the lowest hierarchy, the incoming link corresponding to the link L5, and the predicted congestion level at the time T1 + T2 of the link L5 (calculated as 2 in this example) to the congestion prediction data, Since one congestion degree record R5 corresponding to them is determined and one exit link record R6 of the link L3 included in the congestion degree record R5 is also determined, one predicted congestion degree corresponding to the exit link record R6 is determined.

  FIG. 8 shows a table in which the correspondence relationship between various congestion levels of the entrance link L5 and the congestion levels of the exit link L6 generated as a result of the congestion level at the time T1 + T2, the lower layer, and the node N3 is extracted from the congestion prediction data. As shown in this table, when the congestion level of the entry link L5 is 2 at time T1 + T2, the first predicted congestion level of the exit link L6 resulting from the time T1 + T2 + T6 is 2.

  The calculation process of the second predicted congestion level is as follows. By applying the factor suitable for the time T1 + T2 + T3, the lowest hierarchy, the incoming link corresponding to the link L3, and the predicted congestion level of the link L3 at the time T1 + T2 + T3 (calculated as 3 in this example) to the congestion prediction data, Since one congestion degree record R5 corresponding to them is determined and one exit link record R6 of the link L3 included in the congestion degree record R5 is also determined, one predicted congestion degree corresponding to the exit link record R6 is determined.

  FIG. 9 shows a table in which the correspondence relationship between various congestion levels of the entrance link L3 and congestion levels of the exit link L6 resulting from the congestion level at the time T1 + T2 + T3, the lower layer, and the node N3 is extracted from the congestion prediction data. As shown in this table, when the traffic congestion level of the incoming link L3 is 3 at time T1 + T2 + T3, the second predicted traffic congestion level at the time T1 + T2 + T3 + T5 of the outgoing link L6 resulting therefrom is 3.

  Here, the first predicted congestion level is 2, the second predicted congestion level is 3, and the two are different. However, as described in the predicted congestion of the link L3, the first and second predicted congestion levels are as follows. The degree of traffic is not added like the amount of traffic, but of the first and second predicted congestion levels, the higher congestion level (that is, the larger value) (congestion level 3) is estimated at time T1 + T2 + T6 (estimated at step 130). By adopting it as the predicted congestion degree of the link L3 at the time of the

  After step 145, the process returns to step 120, and it is determined again that there is no link for which the congestion degree is not calculated in the links L2, L3, L5, and L6 in the congestion prediction range, and the process of FIG. .

  By such processing, the predicted traffic congestion degree is calculated for all of the links L2, L3, L5, and L6 within the traffic jam prediction range. Based on the predicted congestion level and the current congestion level of the link L1, a cost corresponding to the congestion level of each link is calculated. Then, using this cost, an optimum route having the lowest total cost from the departure point Po to the destination Pd is calculated (for example, by the Dijkstra method). And, for example, as shown in FIG. 10, even if the congestion level of the link L2 is low at the present time, when the vehicle enters the link L2, there is a high possibility that the congestion level is high as in the above calculation. By avoiding a route passing through the link L2, an optimum route passing through the links L1, L5, L3, and L6 in this order is calculated as a guide route, and route guidance of the guide route is performed. Thereby, congestion in the link L2 can be avoided in advance.

  As described above, in the present embodiment, when calculating the predicted congestion degree of each link, the congestion prediction data is used, and this congestion prediction data is calculated for each factor, for each hierarchy, for each node, and for all of the nodes. For each combination of ingress link and egress link, it includes various predicted congestion levels for the egress link that result from various congestion levels for the egress link.

  The utility that the traffic jam prediction data has information for each factor is as follows. Even with the same entry / exit combination, it is possible to cope with changes in the relationship between the congestion level of the entrance link and the congestion level of the exit link because the traffic flow differs depending on factors (day of the week, time), etc. .

  In addition, this traffic congestion prediction data has data layered in accordance with the road network data hierarchy, and uses the data of the highest hierarchy among the hierarchies that include both the entrance link and the exit link. In step 145, the predicted congestion degree of the exit link is calculated. As a result, for example, the predicted congestion level for long-distance travel using an expressway uses the traffic congestion prediction data that expresses only the traffic flow of the expressway network, which reduces error accumulation and the calculation time. Get smaller.

  Furthermore, the traffic jam prediction data represents the flow of traffic, but is represented not by the traffic volume but by the traffic jam. The traffic volume and the degree of traffic jam do not match, and even with the same traffic volume, traffic may or may not be jammed due to external factors such as gradients and traffic lights. For example, as shown in FIG. 12, even if the traffic volume per unit time is constant, in the case of a traffic volume exceeding a certain threshold, there are portions that are not congested due to conditions such as uphill and traffic lights. Therefore, in order to accurately predict traffic congestion using the passing traffic volume, it is necessary to calculate information such as gradients and traffic lights using external parameters without excess or deficiency. Increases greatly. However, in the present embodiment, the above phenomenon can be expressed succinctly by using the traffic jam prediction data of the degree of traffic jam, and accurate traffic jam prediction can be performed without increasing the calculation time.

  Specifically, the actual congestion degree at time T1 (first time) of link L1 (first link) is acquired from the outside, and the factor, lower layer, and incoming link corresponding to link L1 that conform to time T1 (That is, the approach link L1 itself) and the actual congestion degree at the time T1 of the acquired link L1 are applied to the congestion prediction data.

  Then, due to the structure of the traffic jam prediction data as described above, the exit link L2 that can enter from the approach link L1 via the node N1 (first node) can be specified, and the exit that occurs as a result of various congestion levels of the entrance link L1 Various predicted congestion levels of the link L2 can be specified, and further, the predicted congestion level of the link L2 at the time T1 + T2 (second time point) resulting from the actual congestion level at the time T1 of the link L1 can be acquired.

  In such a method, the actual congestion level at time T1 of the link L1 is acquired from the outside, and the predicted congestion of the link L2 is calculated based on the acquired congestion level. It can reflect the sudden and sudden traffic situation. In addition, instead of the conventional traffic volume simulation, the causal relationship between the traffic congestion degree of the link N1 and the traffic congestion degree of the link L2 in the traffic congestion prediction data is used, and the traffic volume of the link N2 is directly calculated without passing through the traffic volume calculation. The congestion level is calculated. Therefore, the influence other than the traffic volume can be easily reflected in the congestion degree.

  Further, the link L2 (second link) and the link L4 (fourth link) are connected to the link L3 (to the third link) via the same node N2 (second node), and the predicted traffic congestion of the link L3 When the degree of congestion is calculated, if the first predicted congestion degree of the link L3 resulting from the congestion degree of the link L2 is different from the second predicted congestion degree of the link L3 resulting from the congestion degree of the link L4, these are added. Instead of matching, use the one with the highest traffic jam. This is because the congestion degree prediction data of the third link resulting from the congestion degree of the second link is created as data reflecting the influence of inflows from other links to the third link, etc. This is because the congestion degree prediction data of the third link generated as a result of the congestion degree of the link 4 is created as data reflecting the influence of inflow from other links to the third link.

  Similarly, link L5 (fifth linkAnd link P)And link L3 (seventh link And link R)Is the same node N3 (third nodeAnd node A)Via link L6 (sixth linkAnd link Q)When calculating the predicted congestion degree of the link 63, the first predicted congestion degree of the link L6 resulting from the congestion degree of the link L5 and the second of the link L6 resulting from the congestion degree of the link L3 are calculated. If the predicted congestion level is different, the higher congestion level is used instead of adding them together.

Further, the first approach time (third time point and time point V) to the link L6 calculated based on the transit time of the link L5 and the second approach to the link L6 calculated based on the transit time of the link L3. When the time ( time W) is different, the earlier entry time TX of the first and second entry times is set as the entry time to the link L6.

In this way, FIG. 4 is partly changed, and the link within the traffic jam prediction range is further added to the node N4 (not shown, corresponding to an example of the fourth node and the node B ) at the end of the link L6. When the link of L8 (not shown, corresponding to an example of the eighth link and the link S ) is connected and the end point of the link is the destination Pd, the factor suitable for the above entry time TX, lower level By applying the predicted congestion level (calculated as 3 as described above) of the entry link corresponding to the hierarchy, the node N4, the link L6 and the entry time TX of the link L6 to the congestion prediction data, the link N6 to the node N4 It is possible to acquire the L8 predicted traffic congestion degree of the link at the time of entering the link L8 in the target range via the above. By doing so, it is possible to unify the time of entering the link and advance the calculation of the predicted traffic congestion degree of the node.

  In the above embodiment, the control circuit 17 functions as an example of a predicting unit, and the control circuit 17 functions as an example of a range limiting unit by executing Step 105, and also executes Steps 120 to 145. Thus, it functions as an example of a predicted congestion degree acquisition unit.

(Other embodiments)
As mentioned above, although embodiment of this invention was described, the scope of the present invention is not limited only to the said embodiment, The various form which can implement | achieve the function of each invention specific matter of this invention is included. It is. For example, the following forms are also acceptable.

For example, in the above embodiment, the passing time of each approach link is the one recorded for each degree of congestion in the congestion prediction data, but it does not necessarily have to be this way. For example, the transit time of the approach link for each congestion degree may be estimated from the congestion degree, the link length of the approach link, and the link case. In step 145, the predicted congestion degree of the link L6 is determined. At the time of calculation, data included in the factor record R1 of the factor suitable for the time T1 + T2 + T3 is used as the second predicted traffic congestion degree. In the immediately preceding step 130, the entry time to the link L6 is calculated as T1 + T2 + T6. Therefore, the second congestion degree may be calculated using the data included in the factor record R1 of the factor suitable for the time T1 + T2 + T6-T7 so that the link L3 enters the link L6 at the time T1 + T2 + T6. . Here, time T7 is the transit time of the link L3.

  Here, as the transit time T7 to be used, the transit time in the record of the factor conforming to the time T1 + T2 + T6, the lowest hierarchy, the node N3, the incoming link L3, and the estimated congestion degree of the link L3 is adopted.

  Further, in the above embodiment, the traffic jam prediction data is such that when the traffic jam level is determined for one incoming link, the predicted traffic jam level of the corresponding outgoing link is determined, but this need not be the case. . For example, although the amount of data and the amount of calculation increase, the traffic congestion prediction data may be such that one traffic congestion prediction data for the outgoing link is determined for a combination of the traffic congestion degrees of a plurality of incoming links.

  In the above embodiment, the traffic jam prediction data includes, for each of all the nodes on the road, the various congestion levels of the approach link and the traffic levels of all the combinations of the entrance link and the exit link connected to the node. It includes correspondences with various predicted congestion levels of outgoing links, but in order to further reduce the amount of data, multiple nodes are grouped and representative nodes that represent each group are set. For each group's representative node only, the incoming link connected to that node (ie, only the representative incoming link connected to the other representative node at the other end) and the outgoing link (however, the other representative node at the other end) For all combinations (restricted to the representative exit link to be connected), various congestion levels of the entry link and various predicted congestion levels of the exit link It may be adapted to include a correspondence relationship. In this case, it is only necessary to calculate the predicted congestion degree only for the representative leaving link.

  In the above embodiment, each function realized by the control circuit 17 executing the program is realized by using hardware having those functions (for example, an FPGA capable of programming the circuit configuration). You may come to do.

DESCRIPTION OF SYMBOLS 1 Navigation apparatus 11 Position detector 12 Image display apparatus 13 Operation part 14 Speaker 15 Communication part 16 Map data acquisition part 17 Control circuit 17

Claims (5)

For each factor, for each node, and for each combination of all the incoming links and outgoing links for the node, the traffic congestion prediction data including various predicted traffic congestion levels of the outgoing link resulting from various traffic congestion levels of the incoming link. Acquisition means (15) for acquiring
Prediction means (17) for predicting the degree of congestion of a plurality of links based on the congestion prediction data acquired by the acquisition means (15),
The prediction means (17) includes a range limiting means (105) for limiting a traffic jam prediction range that is a target range of a link for predicting a traffic jam degree,
The actual traffic congestion degree at the first time point of the first link is acquired from the outside, the factor conforming to the first time point, the approach link corresponding to the first link, and the acquired first link By applying the actual traffic congestion degree at the first time point to the traffic congestion prediction data, the second link entering the second link in the traffic congestion prediction range from the first link through the first node. at the time, and the second link predicted congestion degree obtaining means for obtaining a predicted congestion degree (120-145), was closed,
The predicted traffic congestion degree acquisition means (120 to 145) further includes
Factors conforming to the second time point, entering link corresponding to the second link, and the predicted traffic jam degree of the second point of the the acquisition second link, fitting in the traffic jam prediction data Then, the first predicted congestion degree of the third link at the third time point when entering the third link in the congestion prediction range from the second link through the second node ,
In addition, when a fourth link is connected to the second node, the fourth link is a factor that matches a fourth time point that is back by the transit time of the fourth link from the third time point. the appropriate approach link and the predetermined congestion level of said fourth link, by fitting those in the traffic jam prediction data, the fourth said link via said second node a third link Obtaining a second predicted congestion degree of the third link at the third time point to enter ,
Said first prediction congestion degree of the third link, of said second predictive congestion degree of the third link, the higher degree of congestion, shall be the predicted congestion degree of the third link A traffic jam prediction device characterized by that. The predicted traffic congestion degree acquisition means (120 to 145) further includes
Applying a factor that conforms to the first time point, an incoming link corresponding to the first link, and an actual traffic congestion degree at the first time point of the acquired first link to the traffic congestion prediction data. in, in the second time point which enters from the first link to link P (L5) in the traffic jam prediction range via the first node, and obtains the prediction congestion degree of the links P,
In addition, by applying the factor that conforms to the second time point, the incoming link corresponding to the link P , and the predicted congestion level at the second time point of the acquired link P to the congestion prediction data, Obtaining a first predicted congestion degree of the link Q at a time point V (T1 + T2 + T6) entering the link Q (L6) in the congestion prediction range from the link P via the node A (N3) ;
In addition, when the link R (L3) in the traffic jam prediction range is connected to the node A , a factor suitable for a time point W (T1 + T2 + T3) different from the time point V , an incoming link corresponding to the link R , and the congestion degree of the links R in the time W, by applying to the traffic jam prediction data, at point X (T1 + T2 + T3 + T5) when through the node a from the link R enters the link Q, a of the link Q Get the predicted traffic congestion degree of 2,
Said first prediction congestion degree of said link Q, among the second prediction congestion degrees of the link Q, claims a higher congestion level, characterized in that the predicted congestion degrees of the link Q congestion prediction device according to 1. For each factor, for each node, and for each combination of all the incoming links and outgoing links for the node, the traffic congestion prediction data including various predicted traffic congestion levels of the outgoing link resulting from various traffic congestion levels of the incoming link. Acquisition means (15) for acquiring
Prediction means (17) for predicting the degree of congestion of a plurality of links based on the congestion prediction data acquired by the acquisition means (15),
The prediction means (17) includes a range limiting means (105) for limiting a traffic jam prediction range that is a target range of a link for predicting a traffic jam degree,
The actual traffic congestion degree at the first time point of the first link is acquired from the outside, the factor conforming to the first time point, the approach link corresponding to the first link, and the acquired first link By applying the actual traffic congestion degree at the first time point to the traffic congestion prediction data, the second link entering the second link in the traffic congestion prediction range from the first link through the first node. at the time, and the second link predicted congestion degree obtaining means for obtaining a predicted congestion degree (120-145), was closed,
The predicted traffic congestion degree acquisition means (120 to 145) further includes
Factors conforming to the first time point, entering link corresponding to the first link, and the actual degree of congestion of the first point of the the acquisition first link, devoted to the traffic jam prediction data by fitting, in the second time of entering the link P (L5) in the first of the traffic jam prediction range via the first node from the link, and obtains the prediction congestion degree of the links P,
Also, the factor conforming to the second time point, entering link corresponding to the link P, and the predicted traffic jam degree of the second point of the acquired links P, by fitting those in the traffic jam prediction data , Obtaining a first predicted congestion degree of the link Q at a time point V (T1 + T2 + T6) entering the link Q (L6) in the congestion prediction range from the link P via the node A (N3) ,
Further, the node when the link of the congested predicted range A R (L3) is connected, cause fits different times W (T1 + T2 + T3) from the previous SL point V, incoming link that falls under the link R and a congestion degree of a link R in the time W, by applying to the traffic jam prediction data, at point X (T1 + T2 + T3 + T5) when entering the link Q from the link R via the node a, the link Get Q ’s second predicted traffic jam,
Said first prediction congestion degree of the links Q, among the second prediction congestion degree of the links Q, the higher degree of congestion, features and to Rukoto to the predicted congestion degree of the links Q Congestion prediction device characterized by. The predicted traffic congestion degree acquisition means (120 to 145) further includes
Said first prediction congestion degree of the links Q, among the second prediction congestion degree of the links Q, the higher degree of congestion, as well as the predicted congestion degree of the links Q, the point V and the the earlier of the time point X and entry time into the link Q,
Moreover, factors compatible with entry time into the link Q, entering link corresponding to the link Q, and the predicted traffic jam degree of the entry time of the acquired links Q, by applying to the traffic jam prediction data, wherein at the time of entering the link S in the traffic jam prediction range via the link Q a node-B, a traffic jam prediction device according to claim 2 or 3, characterized in that to obtain the predicted congestion degree of the links S. For each factor, for each node, and for each combination of all the incoming links and outgoing links for the node, the traffic congestion prediction data including various predicted traffic congestion levels of the outgoing link resulting from various traffic congestion levels of the incoming link. Acquisition means (15) for acquiring
Prediction means (17) for predicting the degree of congestion of a plurality of links based on the congestion prediction data acquired by the acquisition means (15),
The prediction means (17) includes a range limiting means (105) for limiting a traffic jam prediction range that is a target range of a link for predicting a traffic jam degree,
The actual traffic congestion degree at the first time point of the first link is acquired from the outside, the factor conforming to the first time point, the approach link corresponding to the first link, and the acquired first link By applying the actual traffic congestion degree at the first time point to the traffic congestion prediction data, the second link entering the second link in the traffic congestion prediction range from the first link through the first node. at the time, and the second link predicted congestion degree obtaining means for obtaining a predicted congestion degree (120-145), was closed,
Traffic-congestion prediction data and the acquisition means (15) obtains the hierarchy has a plurality hierarchy links and nodes as would higher are thinned out, each factor, each layer, each node, and, for the node to each combination of all the incoming link and egress links, it includes a variety of predicted congestion degree of the exit link caused the entering link various degree of congestion result of,
The predicted congestion level acquisition means (120 to 145), the factor conforming to the first time point, the first link and the highest hierarchy of the second link layer both contains the previous SL entering link corresponding to the first link, and the actual degree of congestion of the first point of the first link acquired, by applying to the traffic jam prediction data, first from the first link that put the second point which enters the second link in the traffic jam prediction range through the node, traffic jam prediction apparatus characterized that you get a prediction congestion degree of the second link.

Images