JP5872222B2 - Travel pattern prediction device - Google Patents

Description

The present invention relates to a vehicle pattern prediction apparatus that predicts a traveling pattern of a vehicle.

  2. Description of the Related Art Conventionally, a car navigation device that predicts a travel pattern of a vehicle, predicts a fuel consumption amount of each road link based on the predicted travel pattern, and searches for a minimum fuel consumption route (see Patent Document 1). .

JP 2010-1070459 A

  In the conventional car navigation apparatus described above, the number N of vehicle acceleration / deceleration on each road link is estimated by the formula N = T / C based on the link travel time T and the constant C determined by the road type. It is used to predict the running pattern of the vehicle. However, with such a simple prediction method, it is difficult to accurately predict the traveling pattern of the vehicle because the road conditions and the presence or absence of traffic lights at the intersection are not taken into consideration.

The travel pattern prediction apparatus according to the present invention selects a link selection unit that selects one of a plurality of links constituting a road network as a selected link, and selects one of the intersections existing in front of the selected link as a target intersection. Driving pattern predicting means for predicting a driving pattern according to the road condition of the link and the type of the target intersection as a driving pattern of the vehicle in the selected link , and the driving pattern predicting means has a smooth road condition of the selected link; and If a traffic signal intersection target intersection is provided of the traffic, based on a stop probability at the beginning of the selected link and stopping probability in object intersection, it characterized that you predict travel pattern.

  ADVANTAGE OF THE INVENTION According to this invention, the driving | running | working pattern of a vehicle can be correctly estimated in consideration of the road condition, the presence or absence of traffic lights at an intersection, and the like.

It is a block diagram which shows the structure of the navigation apparatus as an example of application of the traveling pattern prediction apparatus by one Embodiment of this invention. It is a figure which shows an example of the running pattern by the conventional running pattern prediction process. It is a flowchart of the driving | running | working pattern prediction process by one Embodiment of this invention. It is a flowchart of the traffic light traveling pattern prediction process at the time of smooth. It is a flowchart of the temporary stop travel pattern prediction process at the time of smoothness. It is a flowchart of the passing travel pattern prediction process at the time of smooth. It is a flowchart of the traffic light traveling pattern prediction process at the time of traffic jam. It is a flowchart of the temporary stop travel pattern prediction process at the time of traffic jam. It is a figure which shows the example of the selection link in the traffic signal traveling pattern prediction process at the time of smoothness. It is a figure which shows the example of a red signal stop probability table. It is a figure which shows the example of each scenario in the traffic light traveling pattern prediction process at the time of smoothness. It is a table | surface of the content and occurrence probability of each scenario. It is a figure which shows the example of the prediction driving | running | working pattern by the traffic light driving | running | working pattern prediction process at the time of smoothness. It is a figure which shows the example of the selection link in the temporary stop driving | running | working pattern prediction process at the time of smoothness. It is a figure which shows the example of a preceding number table. It is a figure which shows the example of a slow running table. It is a figure which shows the example of the predicted driving pattern of one acceleration / deceleration in the temporary stop driving pattern prediction process at the time of smoothness. It is a figure which shows the example of the prediction driving | running | working pattern by the temporary stop driving | running | working pattern prediction process at the time of smoothness. It is a figure which shows the example of the selection link in the passage running pattern prediction process at the time of smoothness. It is a figure which shows the example of the prediction driving | running | working pattern by the passage driving | running | working pattern prediction process at the time of smoothness in case the selection link starting end is a traffic signal intersection. It is a figure which shows the example of the prediction driving | running | working pattern by the passage driving | running | working pattern prediction process at the time of smoothness in case the selection link starting end is a temporary stop intersection. It is a figure which shows the example of the prediction driving | running | working pattern by the passing driving | running | working pattern prediction process at the time of smoothness when the selected link starting end is neither a traffic light intersection nor a temporary stop intersection. It is a figure for demonstrating the method to estimate the head of a traffic jam. It is a figure which shows the example of the selection link in the traffic signal traveling pattern prediction process at the time of traffic jam. It is a figure which shows the example of a green signal time table. It is a figure which shows the example of the predicted driving pattern of one acceleration / deceleration in the traffic signal driving pattern prediction process at the time of traffic jam. It is a figure which shows the example of the prediction driving pattern by the traffic light driving pattern prediction process at the time of traffic jam. It is a figure which shows the example of the prediction driving | running | working pattern by the temporary stop driving | running | working pattern prediction process at the time of traffic jam. It is a figure which shows the example of the prediction driving | running | working pattern of the whole object link.

  A travel pattern prediction apparatus according to an embodiment of the present invention will be described below. FIG. 1 is a block diagram showing a configuration of a navigation device as an application example of a travel pattern prediction device according to an embodiment of the present invention. The navigation apparatus 1 is used by being mounted on a vehicle. As shown in FIG. 1, the control unit 10, the vibration gyro 11, the vehicle speed sensor 12, the hard disk drive (HDD) 13, and the GPS (Global Positioning System) reception. Unit 14, VICS (Vehicle Information and Communication System) (registered trademark) information receiving unit 15, display monitor 16, speaker 17, and input device 18.

  The control unit 10 includes a microprocessor, various peripheral circuits, a RAM, a ROM, and the like, and executes various processes based on a control program and map data recorded in the HDD 13. For example, the control unit includes a destination search process when setting a destination, a search process for a recommended route to the set destination, a process for detecting the current position of the host vehicle, various image display processes, and a voice output process. 10 is executed.

  The vibration gyro 11 is a sensor for detecting the angular velocity of the host vehicle. The vehicle speed sensor 12 is a sensor for detecting the traveling speed of the host vehicle. By detecting the motion state of the host vehicle at predetermined time intervals using these sensors, the control unit 10 determines the moving direction and the moving amount of the host vehicle.

  The HDD 13 is a non-volatile recording medium, and records a control program, map data, and the like for executing the above processing in the control unit 10. Data recorded in the HDD 13 is read out under the control of the control unit 10 as necessary, and is used for various processes and controls executed by the control unit 10.

  The map data recorded on the HDD 13 includes route calculation data, road data, and background data. The route calculation data is data used when searching for a recommended route to the destination. The road data is data representing the shape and type of the road. In the road data, each road is constituted by connecting a plurality of points called nodes and shape interpolation points as will be described later. The background data is data representing the background of the map. Note that the background of the map is various components other than roads existing on the map. For example, rivers, railways, green zones, various structures, etc. are represented by background data.

  The minimum unit of each road constituting the road network in map data is called a link. That is, a road network is constituted by a plurality of links corresponding to predetermined road sections, and route calculation data and road data are expressed in units of links. In the road data, points called nodes each having coordinate information set are provided at both ends of each link. A point called a shape interpolation point is provided in the middle of each link as necessary. Coordinate information is set for each shape interpolation point in the same manner as the node. By connecting these points in order, the shape of the road is represented in the road data.

  On the other hand, in the route calculation data, for each link corresponding to each road section, a link cost is set according to the time required for passing when the host vehicle travels the road section. Based on this link cost, a search for a recommended route is performed in the navigation device 1 by obtaining a combination of links according to a preset route search condition. For example, if route search conditions are set so that route search is performed with the shortest travel time as the top priority, the link combination that minimizes the time required to pass from the departure point to the destination is determined as the recommended route. It is done.

  In the above description, the map data is recorded on the HDD 13 in the navigation device 1. However, these may be recorded on a recording medium other than the HDD. For example, map data recorded on a CD-ROM, DVD-ROM, memory card, or the like can be used. That is, the navigation apparatus 1 according to the present embodiment may store these data using any recording medium.

  The GPS receiver 14 receives a GPS signal transmitted from a GPS satellite and outputs it to the controller 10. The GPS signal includes information on the position and transmission time of the GPS satellite that transmitted the GPS signal as information for obtaining the current position of the host vehicle. Therefore, by receiving GPS signals from a predetermined number or more of GPS satellites, the current position of the host vehicle can be calculated by the control unit 10 based on these pieces of information. Based on the calculation result of the current position of the host vehicle based on the GPS signal and the calculation result of the moving direction and the moving amount based on the detection results of the vibration gyroscope 11 and the vehicle speed sensor 12, the control unit 10 automatically A position detection process for detecting the current position of the vehicle is executed, and the current position of the host vehicle is detected.

  The VICS information receiving unit 15 receives VICS information transmitted from the VICS center (not shown) to the navigation device 1. When the VICS information receiving unit 15 receives this VICS information, various road traffic information including traffic jam information is acquired in the navigation device 1. In the traffic jam information provided by the VICS information, the road status of each link is represented by any one of three types of road statuses: smooth, congestion, or traffic jam. The VICS information received by the VICS information receiving unit 15 is output to the control unit 10 and used for displaying traffic jam information, searching for a recommended route, and the like.

  The transmission of VICS information from the VICS center to the navigation device 1 is performed mainly by radio wave beacons installed on highways, optical beacons installed mainly on general roads, or FM multiplex broadcasting. . The radio wave beacon and the optical beacon transmit VICS information locally to the vehicle passing near the installation point by radio waves or light (infrared rays). In contrast, FM multiplex broadcasting can transmit VICS information to a relatively wide area.

  The display monitor 16 is a device for displaying various screens in the navigation device 1 and is configured using a liquid crystal display or the like. The display monitor 16 displays a map screen, a recommended route guidance display, and the like. The contents of the screen displayed on the display monitor 16 are determined by screen display control performed by the control unit 10. The display monitor 16 is installed at a position where the user can easily see, for example, on the dashboard of the host vehicle or in the instrument panel.

  The speaker 17 outputs various audio information under the control of the control unit 10. For example, route guidance voice for guiding the host vehicle to the destination according to the recommended route, various warning sounds, and the like are output from the speaker 17.

  The input device 18 is a device for a user to perform various input operations for operating the navigation device 1 and includes various input switches. The user operates the input device 18 to input, for example, the name of a facility or point desired to be set as a destination, set a search condition for a recommended route, or set a destination from registered locations registered in advance. You can select the ground and scroll the map in any direction. The input device 18 can be realized by an operation panel, a remote controller, or the like. Alternatively, the input device 18 may be a touch panel integrated with the display monitor 16.

  When the user operates the input device 18 to set the destination, the navigation device 1 uses a predetermined algorithm based on the above-described route calculation data with the current position of the host vehicle detected as described above as the departure point. Performs route search processing by calculation. This process searches for a recommended route from the departure point to the destination. Furthermore, the navigation apparatus 1 displays the recommended route searched in a form that can be distinguished from other roads on the map displayed on the display monitor 16 by a method such as changing the color. Then, by outputting predetermined image information and audio information from the display monitor 16 and the speaker 17 according to the recommended route, the host vehicle is guided to the destination.

  In addition, the navigation device 1 can extract a target link from the map data and perform a travel pattern prediction process for predicting a travel pattern of the host vehicle on the target link as necessary. By performing this traveling pattern prediction process, the navigation apparatus 1 can operate as a traveling pattern predicting apparatus that accurately predicts an actual vehicle traveling pattern in consideration of road conditions, traffic lights at intersections, and the like.

  Hereinafter, the travel pattern prediction process of the present invention executed in the navigation device 1 will be described in detail. First, in order to compare with the traveling pattern prediction process of the present invention, the conventional traveling pattern prediction process will be described. FIG. 2 shows an example of a travel pattern predicted by a conventional travel pattern prediction process.

  As shown in FIG. 2, in the conventional travel pattern prediction process, a travel pattern is predicted in which the vehicle repeats acceleration / deceleration at a predetermined acceleration / deceleration at each link regardless of road conditions, types of intersections, and the like. However, it is difficult to say that such a traveling pattern correctly reflects the actual traveling state of the vehicle. Thus, when the conventional travel pattern prediction process is used, it is difficult to accurately predict the actual vehicle travel pattern.

  On the other hand, in the travel pattern prediction process of the present invention, the process shown in the flowchart of FIG. 3 is performed, so that one of the intersections existing in front of the link that predicts the travel pattern is set as the target intersection, and the road condition and target of the link. A travel pattern corresponding to the type of intersection is predicted as the travel pattern of the vehicle on the link. As a result, an actual vehicle traveling pattern is accurately predicted in consideration of road conditions and the presence or absence of traffic lights at intersections.

  The flowchart of FIG. 3 will be described below. The travel pattern prediction process shown in this flowchart is executed by the control unit 10 in the navigation device 1 in accordance with a user instruction, a predetermined process start condition, or the like.

  In step S <b> 10, the control unit 10 extracts a target link as a travel pattern prediction target from the map data recorded in the HDD 13. Here, for example, all links included in the area designated by the user are extracted as target links. Further, for example, by using the travel pattern predicted by the travel pattern prediction process shown in the flowchart of FIG. 3, it is possible to search for the minimum fuel consumption route from the departure place to the destination. In such a case, an extraction range including the starting point and the destination is set by a predetermined setting method, and a link existing in the extraction range is extracted as a target link. In addition to this, any link can be extracted as the target link.

  In step S20, the control unit 10 selects one of the target links extracted in step S10. Hereinafter, this link is referred to as a selection link. As a result, any one of the links constituting the road network in the map data is selected as the selected link.

  In step S30, the control unit 10 acquires traffic jam information for the selected link selected in step S20. Here, the congestion information of the selected link is acquired from the VICS information received by the VICS information receiving unit 15.

  In step S40, the control unit 10 determines whether or not the road condition of the selected link is smooth based on the traffic jam information acquired in step S30. If the road condition of the selected link is smooth, the process proceeds to step S50. In this case, by the processing of steps S50 to S90 described below, the intersection corresponding to the end of the selected link is set as the target intersection, and a traveling pattern according to the type of the target intersection is predicted. On the other hand, if the road condition of the selected link is not smooth, that is, if it is congested or congested, the process proceeds to step S100. In this case, by the processing of steps S100 to S150 described later, a traveling pattern corresponding to the type of the target intersection is predicted with the intersection corresponding to the head of the congestion or traffic jam as the target intersection. In the following description, the case where the selected link is congested and the case where the selected link is congested may be collectively referred to simply as “congested”.

  In step S50, the control unit 10 determines whether or not the selected link terminal that is the target intersection is an intersection where a traffic signal is installed (hereinafter referred to as a traffic signal intersection). If the selected link terminal is a traffic signal intersection, the process proceeds to step S60. If the selected link terminal is not a traffic signal intersection, the process proceeds to step S70.

  In step S60, the control unit 10 executes a traffic light traveling pattern prediction process during smooth operation. By this process, the traveling pattern in the case where the road condition of the selected link is smooth and the selected link terminal that is the target intersection is a traffic signal intersection is predicted. The specific contents of the traffic light traveling pattern prediction process at the time of smoothing will be described in detail later using the flowchart of FIG. If step S60 is performed, the control part 10 will progress to step S160.

  In step S <b> 70, the control unit 10 determines whether or not the selected link terminal that is the target intersection is a temporary stop intersection. If the selected link end is a temporary stop intersection, the process proceeds to step S80, and if not, the process proceeds to step S90.

  In step S80, the control unit 10 executes a temporary stop travel pattern prediction process during normal operation. By this process, the traveling pattern is predicted when the road condition of the selected link is smooth and the selected link terminal that is the target intersection is a temporarily stopped intersection. In addition, the specific content of the temporary stop travel pattern prediction process at the time of smoothing will be described in detail later using the flowchart of FIG. If step S80 is performed, the control part 10 will progress to step S160.

  In step S90, the control unit 10 executes a passing travel pattern prediction process during normal operation. This process predicts a traveling pattern when the road condition of the selected link is smooth and the selected link terminal that is the target intersection is neither a traffic signal intersection nor a temporary stop intersection. The specific contents of the passing travel pattern prediction process during smooth operation will be described in detail later with reference to the flowchart of FIG. If step S90 is performed, the control part 10 will progress to step S160.

  In step S100, the control unit 10 estimates the head of the traffic jam. Here, by using the method described later with reference to FIG. 23, it is possible to trace which of the intersections is the head of the traffic jam by following the link that is connected to the selected link until a link having a smooth road condition is found. It is estimated whether it is. As a result, the intersection as the target intersection corresponding to the head of the congestion or traffic jam including the selected link is estimated.

  In step S110, the control unit 10 determines whether or not the head of the traffic jam estimated as the target intersection by the process of step S100 is a traffic signal intersection. When the head of the traffic jam is a traffic light intersection, the process proceeds to step S120, and when it is not a traffic signal intersection, the process proceeds to step S130.

  In step S120, the control unit 10 executes a traffic light traveling pattern prediction process during a traffic jam. By this process, the traveling pattern is predicted when the road condition of the selected link is congested or congested and the head of the congested traffic at the target intersection is a traffic signal intersection. The specific contents of the traffic light traveling pattern prediction process during a traffic jam will be described in detail later using the flowchart of FIG. If step S120 is performed, the control part 10 will progress to step S160.

  In step S130, the control unit 10 determines whether or not the head of the traffic jam estimated as the target intersection by the process of step S100 is a temporarily stopped intersection. When the head of the traffic jam is a temporary stop intersection, the process proceeds to step S140, and when it is not a temporary stop intersection, the process proceeds to step S150.

  In step S140, the control unit 10 executes a temporary stop travel pattern prediction process during a traffic jam. By this process, the traveling pattern is predicted when the road condition of the selected link is congested or congested, and the head of the congested traffic that is the target intersection is a temporarily stopped intersection. The specific contents of the temporary stop travel pattern prediction process during a traffic jam will be described in detail later with reference to the flowchart of FIG. If step S140 is performed, the control part 10 will progress to step S160.

  In step S150, the control unit 10 applies a preset default travel pattern as the travel pattern of the selected link. For example, a travel pattern according to the conventional prediction method illustrated in FIG. 2 can be applied to the selected link as a default travel pattern. As a result, the traveling pattern is predicted when the road condition of the selected link is congested or congested, and the cause of the traffic jam is unknown because the head of the traffic jam that is the target intersection is neither a traffic light intersection nor a temporary stop intersection. If step S150 is performed, the control part 10 will progress to step S160.

  For example, places where traffic jams occur frequently, such as uphills and railroad crossings on expressways, even though they are not traffic lights or temporary stop intersections, are registered in advance as traffic jams. If it is the head, a default travel pattern corresponding to the head may be applied in step S150. The default running pattern applied at this time can be set in advance based on statistical data obtained by statistically processing past collected data. Further, when the cause of the traffic jam is obvious, such as an accident traffic jam, a default driving pattern corresponding to the traffic jam cause may be applied in step S150. In addition to this, various default travel patterns can be applied to the selected link.

  In step S160, the control unit 10 determines whether all of the target links extracted in step S10 have been selected as selected links in step S20. When all the target links have been selected as the selected links, that is, when the travel pattern has been predicted for all the target links, the control unit 10 ends the travel pattern prediction process shown in the flowchart of FIG. On the other hand, when there is an unselected target link, the control unit 10 returns to step S20. In this case, after selecting any one of the unselected target links as the selected link in step S20, the control unit 10 executes the processing after step S30 on the selected link. By repeating this for all target links, the travel pattern of all target links is predicted.

  The navigation apparatus 1 predicts a vehicle travel pattern in an arbitrary target link by executing the travel pattern prediction process as described above in the control unit 10. The vehicle running pattern predicted in this way can be used for various processes in the navigation device 1. For example, when searching for the minimum fuel consumption route from the departure point to the destination as described above, the energy consumption value of each target link is estimated based on the traveling pattern obtained by the traveling pattern prediction process of FIG. The minimum fuel consumption route can be searched for by obtaining a link combination that minimizes the sum of the energy consumption values from the starting point to the destination.

Next, specific contents of the traffic light traveling pattern prediction process at the time of smoothing executed in step S60 of FIG. 3 will be described. FIG. 4 is a flowchart of traffic light traveling pattern prediction processing during smooth operation. In the following, each process of the flowchart of FIG. 4 will be described using the selection link of the start node N _a and the end node N _b as shown in FIG. 9 as an example. In FIG. 9, the terminal node _Nb is a traffic light intersection.

In step S200, the control unit 10 refers to the pre-stored red stop probability table in HDD 13, to determine the stop probability P _b of the vehicle in the terminal node N _b of the selection links hitting the target intersection. The red signal stop probability table is a table of the probabilities that the host vehicle will stop at a red signal at various traffic signal intersections according to the road type. In other words, the HDD 13 stores look-up table information representing a stop probability corresponding to a combination of road types of roads intersecting each other at an intersection as a red signal stop probability table.

FIG. 10 shows an example of a red signal stop probability table. In the red signal stop probability table of FIG. 10, the road types of the own road and the intersection road are classified into national roads, prefectural roads, and other road types. The stop probability is set in percent. In step S200, by referring to the red signal stop probability table, stop probability P _b of terminal nodes N _b of the selected link illustrated in FIG. 9 is determined. For example self road at a selected link is national road, when the road intersecting the selected link in the end node N _b is prefectural road, stop probability P _b of the end node N _b from red stop probability table of Figure 10 is 30% (0.3).

In step S210, the control unit 10 determines the stop probability _{P a} of the vehicle in the start node _{N a} choice link. Here, it determines the stop probability P _a by a method corresponding to the type of start node N _a choice link. For example, if start node N _a is the traffic intersection, as in step S200, it determines the stop probability P _a of the vehicle in the start node N _a by referring to the red light stop probability table. On the other hand, when start node N _a is paused intersection, since the vehicle is always pause at starting node N _a, the stop probability P _a 1 (100%). If the start node N _a is neither _a traffic light intersection nor a temporary stop intersection, the stop probability P _{a is set} to 0 because the host vehicle does not stop at the start node N _a . In this way, it is possible to determine the stop probability P _a in accordance with the type of start node N _a choice link. It is also possible to determine the stop probability P _a by other methods.

For example, the link selection is national road, when start node N _a is the traffic intersection, a road that intersects with the selected link at the start node N _a is assumed to be a road nor prefectural road in national road. In this case, the stop probability P _a of the start node N _a is determined to be 10% (0.1) from the red signal stop probability table of FIG.

In step S220, the control unit 10, step S200, S210 in based on the stop probability _{P a} stop probability _{P b} and start node _{N a} terminal node _{N b} determined respectively the probability of occurrence of each scenario in the selected link, respectively calculate. The scenario represents each traveling pattern predicted when the host vehicle stops and does not stop at each of the start node N _a and the end node N _b of the selected link.

An example of the scenario corresponding to the selected link in FIG. 9 is shown in FIG. Scenario # 1 shown in FIG. 11 (a), the vehicle in both start node N _a and the terminating node N _b represents the running pattern when no stop. According to Scenario # 1, the vehicle from starting node _{N a} to the end node _{N b,} runs only selected link between the constant speed traveling time _{T const} constant speed _{V const.}

Scenario # 2 shown in FIG. 11 (b), the vehicle does not stop at the start node N _a, represents the running pattern of stopping at the end node N _b. According to scenario # 2, the host vehicle after traveling only selected link between the constant speed traveling time _{T const} constant running speed _{V const} from the starting node _{N a,} and deceleration during time decelerated at the deceleration G _{V const} / G stops at the end node _{N b.}

Scenario # 3 shown in FIG. 11 (c), the vehicle is stopped in starting node N _a, represents the running pattern in the case of not stopping at the end node N _b. According to scenario # 3, the vehicle is stopped at the starting node _{N a,} during the acceleration time _{V const} / G to reach from there to a constant running speed _{V const} travels a selected link while accelerating in the acceleration G. When the constant travel speed V _const is reached, the selected link travels at the constant travel speed V _const to the end node N _b for the constant speed travel time T _const .

Scenario # 4 shown in FIG. 11 (d) represents the running pattern when the vehicle is stopped in both the start node N _a and the terminating node N _b. According to Scenario # 4, the vehicle is stopped at the starting node _{N a,} during the acceleration time _{V const} / G to reach from there to a constant running speed _{V const} travels a selected link while accelerating in the acceleration G. After traveling only selected link between the constant speed traveling time _{T const} constant running speed _{V const} reaches a constant running speed _{V const,} and deceleration during time decelerated at the deceleration G _{V const} / G, at the end node _{N b} Stop.

In each scenario shown in FIG. 11 (a) ~ (d) , the time T _a represents the time to start traveling vehicle is starting node N _a pass or from the starting node N _a, the time T _b is the vehicle There represents a time to stop in the pass or end node N _b terminal nodes N _b.

FIG. 12 shows a list of the contents and occurrence probabilities of the scenarios in FIGS. As shown in FIG. 12, when the occurrence probability of the scenario # 1 to # 4 and _P 1 to P _4, the probability _P 1 to P ₄ is the start node _{N a} stop probability _{P a} and the terminating node _{N b} based on the stop probability P _b, respectively can be calculated by the equation (1) below. In the equation (1), the stop probabilities P _a and P _b are not the probability values expressed in percent as shown in the red signal stop probability table of FIG. 10, but 0 ≦ P _a and P _b ≦ 1.

... (1)

In step S230, the control unit 10 calculates the constant traveling speed V _const of the _host vehicle based on the occurrence probabilities P _{1 to} P ₄ of the scenarios calculated in step S220. Here, the constant traveling speed V _const is calculated by the following calculation method.

In order to calculate the constant travel speed V _const , consider the expected value of the travel distance in each scenario. The expected value of the distance that the host vehicle travels on the selected link during acceleration can be expressed by the following equation (2) using the occurrence probabilities P ₃ and P ₄ and the acceleration G of scenarios # 3 and # 4.

... (2)

On the other hand, the expected value of the distance that the host vehicle travels on the selected link while decelerating can be expressed by the following equation (3) using the occurrence probabilities P ₂ and P ₄ and the deceleration G of scenarios # 2 and # 4. it can.

... (3)

The expected value of the distance that the _host vehicle travels on the selected link at a constant travel speed V _const is expressed by the following formula (4) using the occurrence probabilities P _{1 to} P ₄ and the acceleration / deceleration G of scenarios # 1 to # 4. Can do. Here, T represents the link travel time of the selected link. The control unit 10 can determine the link travel time T from the congestion information included in the VICS information received by the map data and the VICS information receiving unit 15.

... (4)

When Expression (4) is transformed using the fact that P ₁ + P ₂ + P ₃ + P ₄ = 1, the following Expression (5) is obtained.

... (5)

  By summing equations (2), (3), and (5), an expected value of the distance that the host vehicle travels on the selected link can be obtained as equation (6).

... (6)

  Here, assuming that the link length of the selected link is L, the expected value of the travel distance represented by Expression (6) matches the link length L. Therefore, the following quadratic equation (7) is obtained.

... (7)

By solving the quadratic equation (7), a constant traveling speed V _const is obtained as in the following equation (8). In the formula (8), P = P ₂ + P ₃ + 2P ₄ is satisfied.

... (8)

  The value of P in the equation (8) can be obtained as in the following equation (9) by using the equation (1).

... (9)

As represented by equation (9), the value of P in formula (8) can be represented by the total value of the stop probability P _b of start node N _a stop probability P _a and the terminating node N _b.

In step S230, the constant traveling speed V _const can be calculated by the calculation method described above.

In step S240, the control unit 10 determines, based on the constant travel speed V _const calculated in step S230, the constant speed travel distance L _const representing the distance and time that the _host vehicle travels the selected link at the constant travel speed V _const. The constant speed travel time T _const is calculated. Here, the constant speed travel distance L _const and the constant speed travel time T _const can be calculated by using the following equations (10) and (11).

(10)

(11)

In step S250, the control unit 10 determines that the vehicle on the selected link is based on the constant travel speed V _const calculated in step S230, the constant speed travel distance L _const and the constant speed travel time T _const calculated in step S240. Determine the predicted travel pattern. Here, each traveling pattern of scenarios # 1 to # 4 specified by the constant traveling speed V _const , the constant speed traveling distance L _const and the constant speed traveling time T _const is expressed at a rate corresponding to the occurrence probability P _{1 to} P _4. By superimposing, the predicted traveling pattern of the host vehicle on the selected link is determined. That is, the overlapping pattern corresponding to the expected value of each scenario is predicted as the traveling pattern of the selected link.

  If step S250 is performed, the control part 10 will complete | finish the flowchart of FIG. 4, and will return to the flowchart of FIG. As described above, the traffic light traveling pattern prediction process during smooth running is executed.

FIG. 13 shows an example of the predicted traveling pattern obtained by the traffic light traveling pattern prediction process during the above-described smooth operation. Portion indicated by the dashed line between times T _b from the time T _a in the estimated travel patterns of Figure 13, the running pattern when accelerating or decelerating the running pattern of the case of a constant running speed V _const is the probability of occurrence of each scenario It represents that they are overlaid according to P _{1 to} P ₄ .

Next, specific contents of the temporary stop travel pattern prediction process at the time of smooth execution executed in step S80 of FIG. 3 will be described. FIG. 5 is a flowchart of the temporary stop travel pattern prediction process during smooth operation. In the following, each process of the flowchart of FIG. 5 will be described using the selection link of the start node N _a and the end node N _b as shown in FIG. 14 as an example. In FIG. 14, the terminal node _Nb is a temporary stop intersection.

  In step S300, the control unit 10 refers to the preceding number table stored in advance in the HDD 13 and determines the number of times of acceleration / deceleration in the selected link. The preceding number table is a table in which the number of preceding vehicles that are predicted to have already been lined up at various intersections where the vehicle has arrived and are temporarily stopped at the intersections according to the road type. That is, the HDD 13 stores look-up table information representing the number of preceding vehicles according to the combination of road types of roads that intersect each other at an intersection as a preceding number table.

FIG. 15 shows an example of the preceding number table. In the preceding number table of FIG. 15, the road types of the own road and the intersection road are classified into national roads, prefectural roads, and other road types, respectively, and when the own vehicle arrives for the combination of the respective road types. Each predicted number of preceding vehicles is set. At step S300, the predicting the number of the preceding vehicle at the terminal node N _b of selected link by referring to the preceding number table to determine that number as acceleration and deceleration times in the selected link. For example the link selection is a self road is national road, when the road intersecting the selected link in the end node N _b is prefectural road, acceleration and deceleration times from the previous number table in FIG. 15 is determined to be 5 times.

In step S310, the control unit 10 determines the stop probability _{P a} of the vehicle in the start node _{N a} choice link. Here, it is possible to determine the stop probability P _a In the same manner as in step S210 of FIG.

In step S320, the control unit 10 refers to the slow travel table stored in advance in the HDD 13 and determines the acceleration G of the host vehicle and the constant travel speed V _const on the selected link. The slow travel table is a table that represents the relationship between the advance distance, acceleration, and travel speed during slow travel of the host vehicle learned by the navigation device 1 so far. That is, the HDD 13 stores look-up table information representing learning acceleration and learning traveling speed according to the forward travel distance during slow traveling based on the past traveling history of the host vehicle as a slow traveling table.

FIG. 16 shows an example of the slow travel table. In the slow travel table of FIG. 16, the advance distance is classified into 1 to 30 m, 31 to 60 m, 61 to 90 m, 91 to 120 m, 121 m or more. A learning acceleration and a learning traveling speed are set. In step S320, the acceleration G and the constant traveling speed V _const of the _host vehicle in the selected link are determined by referring to the slow traveling table. For example, let L _car be the stop occupation length (length per preceding vehicle including the space between vehicles) when the host vehicle is stopped at a temporary stop intersection. By referring to the slow travel table of FIG. 16 using this L _car as the forward distance, the acceleration G and the constant travel speed V _const of the host vehicle at the selected link can be determined in step S320. For example, if the stop occupation length L _car is 7 m, the acceleration G in the selected link is determined to be 2.5 (km / h) / s and the constant travel speed V _const is determined to be 10 km / s from the slow travel table of FIG. .

  Note that the navigation device 1 determines that the vehicle is traveling slowly when the host vehicle accelerates from a stopped state and travels a predetermined distance or more at a constant traveling speed that is equal to or less than a predetermined value. The traveling speed is stored as learning information. In this way, based on the learning information stored every time the host vehicle travels slowly, the acceleration and traveling speed during slow traveling are classified for each advance distance and statistical processing is performed, whereby a slow traveling table as shown in FIG. Is obtained and stored in the HDD 13.

In step S330, based on the acceleration G determined in step S320, the control unit 10 accelerates from the stop state when the host vehicle travels slowly, decelerates immediately after acceleration, and stops at the acceleration / deceleration. V _max is calculated. FIG. 17A shows an example of a predicted travel pattern for one acceleration / deceleration in this case.

In estimated travel patterns shown in FIG. 17 (a), the acceleration time to the vehicle traveling speed reaches a maximum speed V _max from 0, the deceleration time to reach 0 from the maximum speed V _max are all Can also be expressed by V _max / G. Therefore, the travel distance per acceleration / deceleration can be expressed as (V _max ) ² / G. Since this travel distance is equal to the stop occupation length L _car described above, the following expression (12) is established.

(12)

From the equation (12), the maximum traveling speed V _max at the time of acceleration / deceleration is obtained as the following equation (13).

... (13)

In step S340, the control unit 10 compares the constant traveling speed V _const determined in step S320 with the maximum traveling speed V _max during acceleration / deceleration calculated in step S330, and whether or not V _max ≦ V _const is _satisfied . Determine. If V _max ≦ V _const , the process proceeds to step S350. If V _max ≦ V _const is not satisfied, that is, if V _max > V _const , the process proceeds to step S360.

In step S350, the control unit 10, the running pattern based on the maximum speed V _max of the time calculated deceleration in step S330, i.e. prediction of one of acceleration and deceleration of the running pattern such as shown in FIG. 17 (a) in the selected link Adopt as a running pattern. If step S350 is performed, the control part 10 will progress to step S380.

In step S360, the control unit 10 determines the time from when the host vehicle accelerates to the constant travel speed V _const in one acceleration / deceleration until the vehicle starts to decelerate, to the constant speed travel time T ′ in one acceleration / deceleration. Calculate as _const . FIG. 17B shows an example of a predicted traveling pattern for one acceleration / deceleration in this case.

In the predicted traveling pattern shown in FIG. 17B, the acceleration time until the traveling speed of the _host vehicle reaches 0 to the constant traveling speed V _const and the deceleration time until the traveling speed V _const reaches 0 are expressed as steps. Both can be expressed by V _const / G using the acceleration G determined in S320 and the constant traveling speed V _const . Therefore, the travel distance per acceleration / deceleration can be expressed as T ′ _const × V _const + (V _const ) ² / G. Since this travel distance is equal to the stop occupation length L _car described above, the following equation (14) is established.

(14)

From the equation (14), the constant speed traveling time T ′ _const in one acceleration / deceleration is obtained as the following equation (15).

... (15)

In step S370, the control unit 10 adopts a travel pattern based on the constant travel speed V _const determined in step S320, that is, a travel pattern as shown in FIG. 17B as a predicted travel pattern for one acceleration / deceleration on the selected link. To do. It should be noted that the constant speed travel time T ′ _const in FIG. 17B is obtained by equation (15) in step S360. If step S370 is performed, the control part 10 will progress to step S380.

In step S380, the control unit 10 determines the predicted travel pattern of the host vehicle on the selected link based on the number of times of acceleration / deceleration determined in step S300 and the predicted travel pattern of acceleration / deceleration adopted in step S350 or S370. decide. Here, after the host vehicle travels the selected link at a predetermined running speed, until it reaches the end node N _b of the link selection, only repeated once the predicted travel patterns of acceleration and deceleration acceleration and deceleration times determined in step S300 Is determined as the predicted traveling pattern of the host vehicle in the selected link. That is, when executing step S350, predicts a travel pattern of the vehicle based on the maximum speed V _max of the deceleration time calculated in step S330. On the other hand, when step S370 is executed, the traveling pattern of the host vehicle is predicted based on the constant traveling speed V _const determined in step S320.

Furthermore, based on the host vehicle stop probability P _a at the beginning node N _a selection links determined in step S310, the a travel pattern of when the vehicle is stopped at the start node N _a, if not stopped by the starting node N _a In the same manner as that performed in step S250 of FIG. 4, the driving patterns are superimposed at a rate corresponding to each occurrence probability. Here, the probability of the former running pattern is equal to the stop probability P _a. In addition, when the stop probability P _a is expressed as a probability value that is not _a percentage, that is, a probability value based on 1, the occurrence probability of the latter traveling pattern can be expressed as (1-P _a ). That is, the two running pattern is superimposed at a rate corresponding to the stop probability P _a.

  If step S380 is performed, the control part 10 will complete | finish the flowchart of FIG. 5, and will return to the flowchart of FIG. As described above, the temporary stop travel pattern prediction process during smooth running is executed.

FIG. 18 shows an example of a predicted traveling pattern obtained by the above-described temporary suspension traveling pattern prediction process during smoothing. The estimated travel patterns of FIG. 18, the prior number was deceleration number determined from the table 3, the running pattern based on the maximum speed V _max of the acceleration and deceleration time as shown in FIG. 17 (a) accordingly The example of the prediction driving | running | working pattern which repeated 3 times is shown. The portion indicated by a broken line between times T _b from the time T _a is overlapped with the traveling pattern of when the vehicle does not stop the running pattern of stopping at the start node N _a in response to the stop probability P _a It shows that it is matched.

Next, the specific content of the passing travel pattern prediction process at the time of smooth execution executed in step S90 of FIG. 3 will be described. FIG. 6 is a flowchart of the passing travel pattern prediction process during smooth operation. In the following, each process of the flowchart of FIG. 6 will be described using the selection link of the start node N _a and the end node N _b as shown in FIG. 19 as an example.

In step S400, the control unit 10 determines whether the starting node or N _a is traffic intersection or pause the intersection of the selected link. If start node N _a is the traffic intersection or pause intersection proceeds to step S410, if not a pause intersection in the traffic intersection, the process proceeds to step S460.

In step S410, the control unit 10 determines the stop probability _{P a} of the vehicle in the start node _{N a} choice link. Here, it is possible to determine the stop probability P _a In the same manner as in step S310 in step S210 and 5 of FIG.

In step S420, the control unit 10, based on the stop probability P _a determined in step S410, calculates the probability of occurrence of each scenario in the selected link, respectively. Here, the scenario of FIG. 11 and Scenario # 1 shown in FIG. 11 (a) where the vehicle corresponds to the case where not stop at the start node N _a, the vehicle corresponds to a case of stopping at the start node N _a (c) # 3 and how to calculate each occurrence probabilities based on stop probability P _a. This is obtained as occurrence probabilities P ₁ and P ₃ when the stop probability P _b of the terminal node N _b is set to 0 in the above equation (1). _{That, P 1 = (1-P} a), which is _P 3 = P _a.

In step S430, the control unit 10 calculates the constant traveling speed V _const of the _host vehicle based on the occurrence probabilities P ₁ and P ₃ of each scenario calculated in step S420. Here, as in step S230 of FIG. 4, the constant traveling speed V _const can be calculated using the above-described equation (8). At this time, the value of P in Equation (8) is obtained by setting P ₂ = P ₄ = 0 and P _b = 0 in Equation (9). That is, _{P = P} 3 _{= P a.}

In step S440, based on the constant travel speed V _const calculated in step S430, the control unit 10 determines a constant speed travel distance L _const representing the distance and time that the _host vehicle travels the selected link at the constant travel speed V _const , and The constant speed travel time T _const is calculated. Here, as in step S240 of FIG. 4, the constant speed travel distance L _const and the constant speed travel time T _const can be calculated by using the above-described equations (10) and (11).

In step S450, the control unit 10 determines that the vehicle on the selected link is based on the constant travel speed V _const calculated in step S430, the constant speed travel distance L _const and the constant speed travel time T _const calculated in step S440. Determine the predicted travel pattern. Here, as in step S250 of FIG. 4, each of the travel patterns of scenarios # 1 and # 3 specified by the constant travel speed V _const , the constant speed travel distance L _const and the constant speed travel time T _{const is represented} by the occurrence probability P _1. by superimposing in a ratio corresponding to P _3, to determine the estimated travel patterns of the vehicle in the selected link. That is, the travel pattern of scenario # 1 is a constant running speed from starting node N _a selection links to the end node N _b, and a travel pattern of scenario # 3 becomes constant travel speed V _const after rising in acceleration G , superimposed at a rate corresponding to the stop probability P _a at the beginning node N _a choice link. In this way, the traveling pattern of the selected link is predicted.

In step S460, the control unit 10 calculates a constant traveling speed V _const of the _host vehicle. In this case, the host vehicle without stopping at the start node N _a selection link, running between the starting node N _a to the end node N _b at a constant speed V _const. Accordingly, the constant traveling speed V _{const at} this time is obtained by the following equation (16) using the link length L and the link travel time T of the selected link.

... (16)

In step S470, the control unit 10 determines the predicted travel pattern of the host vehicle on the selected link based on the constant travel speed V _const calculated in step S460. Here, the travel pattern of the constant travel speed V _const from the start node N _a to the end node N _b is determined as the predicted travel pattern of the selected link.

  If step S450 or S470 is performed, the control part 10 will complete | finish the flowchart of FIG. 6, and will return to the flowchart of FIG. As described above, the passing travel pattern prediction process during smooth running is executed.

Examples of predicted traveling patterns obtained by the above-described smooth traveling pattern prediction process are shown in FIGS. FIG. 20 shows an example of a predicted travel pattern in the case where the selected link start end is a traffic light intersection. In this estimated travel patterns, the time T portion shown by a broken line between _a from the time T _b, the vehicle is running pattern and the stop probability P _a in the case where the running pattern and does not stop when stopping at the start node N _a It is shown that they are overlaid according to. Note that such a predicted traveling pattern is determined in step S450 in FIG.

FIG. 21 shows an example of a predicted travel pattern in the case where the selected link start end is a temporary stop intersection. In this case, since the vehicle is always stopped at the start node N _a is paused intersection, the running speed at the time T _a is 0, continue to accelerate until it reaches from there to a constant running speed V _const. Once you reach a certain speed _{V const,} to maintain the running speed until it is time _{T a.} Such a predicted travel patterns are those stopping probability P _a start end node N _a is determined as the predicted travel pattern when a 1 (100%) at step S450 in FIG. 6.

FIG. 22 shows an example of a predicted travel pattern in the case where the selected link start end is neither a traffic signal intersection nor a temporary stop intersection. In this case, the vehicle is because it does not stop at the start node _{N a,} constant running speed _{V const} from the starting node _{N a} to the end node _{N b} is maintained. Such a predicted traveling pattern is determined in step S470 in FIG.

  Next, a method for estimating the head of the traffic jam in step S100 in FIG. 3 will be described. FIG. 23 is a diagram for explaining a method for estimating the head of a traffic jam.

  In FIG. 23, the link between the node 22 and the node 23 is congested, and the link between the node 23 and the node 24 ahead is smooth. Therefore, when each link between the nodes 21, 22, and 23 is a selected link, it is estimated that the head of the traffic jam is the node 23.

  Next, the link between the node 24 and the node 25 is congested. Here, the branch from the node 25 branches to a link to the node 26, a link to the node 30, and a link to the node 32. In such a case, each link at the branch destination is traced until there is no traffic jam, and the node that smoothly switches from traffic jam or congestion at the branch destination with the longest final traffic jam length is estimated as the head of the traffic jam. That is, when the link between the node 24 and the node 25 is the selected link, it is estimated that the head of the traffic jam is the node 28. Note that even when each link between the nodes 25, 26, 27, and 28 is a selected link, it is estimated that the head of the traffic jam is the node 28.

  On the other hand, when the link from the node 25 to the node 30 is a selected link, the link is congested, and the link between the node 30 and the node 31 is smooth. Therefore, it is estimated that the head of the traffic jam is the node 30. Further, when the link from the node 25 to the node 32 is a selected link, the link and the link between the node 32 and the node 33 are congested, and the link between the node 33 and the node 34 is smooth. is there. Therefore, it is estimated that the head of the traffic jam is the node 33. Similarly, when the link between the node 32 and the node 33 is the selected link, it is estimated that the head of the traffic jam is the node 33. As described above, the traffic jam head node corresponding to the traffic jam head is estimated. In the traffic light traveling pattern prediction process at the time of traffic jam in step S120 in FIG. 3 and the temporarily stopped travel pattern prediction process at the time of traffic jam in step S140, the traffic jam head node estimated in this way is set as the target intersection.

Next, the specific contents of the traffic light traveling pattern prediction process at the time of traffic jam executed in step S120 in FIG. 3 will be described. FIG. 7 is a flowchart of the traffic light traveling pattern prediction process during a traffic jam. In the following, each process of the flowchart of FIG. 7 will be described using the selection link of the start node N _a and the end node N _b as shown in FIG. 24 as an example. Unlike the case of the selection links in traffic driving pattern prediction processing when well illustrated in FIG. 9, an end node N _b in FIG. 24 is not limited to traffic intersections.

In step S500, the control unit 10 refers to a green signal time table stored in advance in the HDD 13 and determines a green signal time T _adv per time when the traffic light becomes a green signal at the traffic jam leading node corresponding to the target intersection. The green light time table is a table of the time per green light at various traffic signal intersections according to the road type. In other words, the HDD 13 stores look-up table information representing a green signal time corresponding to a combination of road types of roads that intersect each other at an intersection as a green signal time table.

FIG. 25 shows an example of the green light time table. In the green light time table of FIG. 25, the road types of the own road and the intersecting road are classified into national roads, prefectural roads, and other road types, and the time per green light for each combination of the road types is shown. Each is set. In step S500, the green light time T _{adv at the} traffic jam head node is determined by referring to the green light time table. At this time, the link having the traffic jam head node as the end node is defined as the own road, and the road intersecting with the traffic jam head node is defined as the intersection road. For example, when the own road is a national road and the intersection road is a prefectural road, the green signal time T _adv of the traffic jam leading node is determined as 100 seconds from the green signal time table of FIG.

In the case of a traffic light whose time per green light changes depending on the traffic volume or time zone of the road, a different green light time table may be referred to corresponding to the change. In this way, an appropriate green signal time T _adv can be obtained.

In step S510, the control unit 10 determines the number of times of acceleration / deceleration in the selected link based on the green signal time T _adv determined from the green signal time table in step S500. Here, the number of times of acceleration / deceleration can be determined as follows.

The distance L _adv that the _host vehicle moves forward during one green light at the time of traffic jam is the above-mentioned green signal time T _adv , the passing number N _car that represents the number of vehicles that pass the green light per unit time, and the aforementioned stop occupation Using the length L _car , it can be expressed by the following equation (17).

... (17)

In the equation (17), a preset value can be used for each of the passing number N _car and the stop occupation length L _car . This value may be a fixed value, or may be changed according to the degree of traffic jams or the road type.

When the acceleration and deceleration times in the selected link and _{N adv,} by the equation (17), the acceleration and deceleration times _{N adv} is calculated by the following equation (18). In Expression (18), L represents the link length of the selected link. Further, the parentheses in the equation (18) are Gauss symbols, and represent the maximum integer not exceeding the numerical value in the parentheses.

... (18)

In step S510, the acceleration / deceleration frequency N _adv on the selected link can be calculated by the calculation method as described above.

In step S520, the control unit 10 refers to the slow travel table stored in advance in the HDD 13, and determines the acceleration G of the host vehicle at the selected link. This slow travel table is the same as that referenced in step S320 of FIG. 5, and as illustrated in FIG. 16, the forward distance and acceleration at the time of slow travel of the host vehicle learned by the navigation device 1 so far. In addition, the relationship with travel speed is tabulated. Here, the acceleration G of the host vehicle in the selected link is determined by referring to the acceleration column corresponding to the forward travel distance L _adv obtained by the above equation (17) in the slow travel table. At this time, it is sufficient to refer only to the acceleration column, and it is not necessary to refer to the traveling speed column.

In subsequent step S530, the control unit 10 determines whether the vehicle on the selected link is constant based on the green signal time T _adv determined in step S500, the acceleration G determined in step S520, the stop occupation length L _car, and the number of vehicles passing N _car. A traveling speed V _const is calculated. Here, the constant traveling speed V _const is calculated by the following calculation method using the predicted traveling pattern of acceleration / deceleration per green signal as exemplified in FIG.

In the predicted acceleration / deceleration traveling pattern shown in FIG. 26, the acceleration time until the traveling speed of the _host vehicle reaches 0 to the constant traveling speed V _const , the deceleration time until the traveling speed V _const reaches 0, Can be expressed by V _const / G using acceleration G. Since the time obtained by subtracting these acceleration time and deceleration time from the green signal time T _{adv is the} time during which the host vehicle travels at a constant travel speed V _const in one green signal, the travel distance per acceleration / deceleration is It can be expressed as (T _adv −V _const / G) × V _const . Since this travel distance is equal to the advance distance L _adv represented by Expression (17), the following Expression (19) is established.

... (19)

  By transforming equation (19), the following quadratic equation (20) is obtained.

... (20)

By solving the quadratic equation (20), the constant traveling speed V _const is obtained as in the following equation (21).

(21)

In step S530, the constant traveling speed V _const can be calculated by the calculation method described above.

In step S540, the control unit 10 travels the selected link on the selected link based on the constant traveling speed V _const calculated in step S530 and the acceleration / deceleration number N _adv calculated in step S510 at the constant traveling speed V _const . A constant speed travel distance L _const and a constant speed travel time T _const representing the distance and time are calculated. Here, the constant speed travel distance L _const and the constant speed travel time T _const can be calculated by using the following equations (22) and (23).

(22)

... (23)

The constant speed travel distance L _const and the constant speed travel time T _const represented by the equations (22) and (23) are the constant travel speed V _const when the host vehicle travels the selected link while repeating acceleration and deceleration. Each total value of the traveling distance and traveling time. That is, by dividing the constant speed travel distance L _const represented by Expression (22) and the constant speed travel time T _const represented by Expression (23) by the acceleration / deceleration number N _adv , respectively, The travel distance and travel time per deceleration can be obtained.

In step S550, the control unit 10 determines that the vehicle on the selected link is based on the constant travel speed V _const calculated in step S530, the constant speed travel distance L _const and the constant speed travel time T _const calculated in step S540. Determine the predicted travel pattern. Here, from the start node N _a selection link until it reaches the terminal node N _b, a single predicted travel patterns of acceleration and deceleration as the vehicle only acceleration and deceleration times N _adv calculated in step S510 Fig. 26 is repeated Is determined as the predicted traveling pattern of the host vehicle in the selected link.

In step S560, the control unit 10 determines the distance from the starting node N _a selection links to jam the top node is the target intersection whether a predetermined threshold value D ₁ or more. The threshold D ₁ is the top represents the distance no longer follow the own vehicle traveling on the selected link to the movement of the acceleration and deceleration of the vehicle at, for example, values such as 300m is preset congestion . When the distance from the starting node N _a selection links to jam the head node is the threshold value D ₁ or more, the process proceeds to step S570.

  In step S570, the control unit 10 performs a waveform rounding process on the predicted travel pattern determined in step S550. In this waveform rounding process, the fluctuation range of the predicted travel pattern determined in step S550 is reduced by a predetermined amount, for example, 20%. That is, when the lower limit value of the travel speed in the predicted travel pattern is 0% and the upper limit value is 100%, the waveform portion included in each of the upper and lower ranges of 10% is cut so that the travel speed does not change any more. Limit to. By doing in this way, when the distance from the own vehicle to the head of the traffic jam is some distance away, the running speed dull caused by the host vehicle not sufficiently following the acceleration / deceleration movement of the preceding vehicle at the head of the traffic jam, That is, the reduction of the fluctuation range is reproduced in the predicted travel pattern. The waveform rounding process in step S570 may be performed as necessary, and may be omitted if unnecessary.

If step S570 is performed, the control part 10 will complete | finish the flowchart of FIG. 7, and will return to the flowchart of FIG. The distance from the starting node N _a selection links to jam the top node in step S560 is a case where it is determined to be less than the threshold value D _1, and ends the flow chart of FIG. 7 without executing step S570. As described above, the traffic light traveling pattern prediction process at the time of traffic jam is executed.

FIG. 27 shows an example of a predicted traveling pattern obtained by the traffic light traveling pattern prediction process at the time of traffic jam described above. In the predicted travel pattern of FIG. 27, the number of times of acceleration / deceleration calculated in step S510 is 3, and in accordance with this, a single acceleration / deceleration travel pattern as shown in FIG. 26 is repeated three times. An example is shown. In addition, the time from the time T _a to time T _b, is a link travel time T of the selected link. Therefore, a predicted traveling pattern as shown in FIG. 27 can be obtained by allocating and arranging each traveling pattern of acceleration / deceleration evenly within the range of the link travel time T.

Next, the specific content of the temporary stop travel pattern prediction process at the time of traffic jam executed in step S140 of FIG. 3 will be described. FIG. 8 is a flowchart of the temporary stop travel pattern prediction process during a traffic jam. In the following, each process of the flowchart of FIG. 8 will be described using the selection link of the start node N _a and the end node N _b as shown in FIG. 24 as an example, similar to the traffic light traveling pattern prediction process at the time of traffic jam described above. Unlike the case of the selected link in the pause driving pattern prediction processing when well illustrated in FIG. 14, an end node N _b in FIG. 24 is not limited to pause intersection.

  In step S600, the control unit 10 determines the number of times of acceleration / deceleration on the selected link. Here, unlike the case where the number of acceleration / decelerations is determined in step S300 of FIG. 5 in the temporary stop travel pattern prediction process during smoothing, the number of acceleration / decelerations is determined without referring to the preceding number table. Specifically, the number of times of acceleration / deceleration can be determined as follows.

When there is a traffic jam starting from the paused intersection, there are many vehicles in the range from the paused intersection to the selected link. Each vehicle must be paused before the paused intersection. It is considered to pass through the intersection. Accordingly, if the number of acceleration / decelerations in the selected link is N _adv , the number of accelerations / decelerations N _adv can be obtained as in the following equation (24). In Expression (24), L and L _car represent the link length and stop occupation length of the selected link as described above. Further, the parentheses in the formula (24) are Gaussian symbols, and represent the maximum integer not exceeding the numerical value in the parentheses.

... (24)

In step S600, the acceleration / deceleration frequency N _adv on the selected link can be calculated by the calculation method described above.

In step S610, the control unit 10 refers to the slow travel table stored in advance in the HDD 13 and determines the acceleration G of the host vehicle and the constant travel speed V _const on the selected link. This slow travel table is the same as that referred to in step S320 in FIG. 5 or step S520 in FIG. 7, and as illustrated in FIG. 16, the slow travel of the host vehicle learned so far by the navigation device 1 is performed. The relationship between the forward distance, the acceleration, and the traveling speed at the time is shown as a table. Here, as in the case of step S320 in FIG. 5, the acceleration G and the constant traveling speed V _cons of the _host vehicle in the selected link are determined by referring to the slow traveling table with the stop occupation length L _car as the forward distance.

In step S620, based on the acceleration G determined in step S610, the control unit 10 accelerates from the stopped state when the host vehicle travels slowly, decelerates immediately after acceleration, and stops at the acceleration / deceleration. V _max is calculated. Here, the maximum traveling speed V _max given by the above-described equation (13) is calculated in the predicted traveling pattern as shown in FIG. 17A by the same method as that performed in step S330 of FIG.

In step S630, the control unit 10 compares the constant travel speed V _const determined in step S610 with the maximum travel speed V _max during acceleration / deceleration calculated in step S620, similarly to step S340 in FIG. It is determined whether _max ≦ V _const . If V _max ≦ V _const, the process proceeds to step S640. If V _max ≦ V _const is not satisfied, that is, if V _max > V _const , the process proceeds to step S650.

In step S640, the control unit 10, as in step S350 of FIG. 5, the travel pattern as shown in FIG. 17 based on the maximum speed _{V max} of the deceleration time calculated in step S620 (a), one in the selected link Adopted as a predicted running pattern of acceleration / deceleration times. If step S640 is performed, the control part 10 will progress to step S670.

In step S650, the control unit 10 calculates a constant speed travel time T ′ _const for one acceleration / deceleration based on the acceleration G and the constant travel speed V _const determined in step S610. Here, the constant speed travel time T ′ _const given by the above equation (15) is calculated in the predicted travel pattern as shown in FIG. 17B by the same method as that performed in step S360 of FIG. To do.

In step S660, as in step S370 in FIG. 5, the control unit 10 performs a single acceleration / deceleration on the selected link at the travel pattern shown in FIG. 17B based on the constant travel speed V _const determined in step S610. Adopted as the predicted driving pattern. If step S660 is performed, the control part 10 will progress to step S670.

In step S670, the control unit 10 determines the predicted travel pattern of the host vehicle on the selected link based on the number of times of acceleration / deceleration determined in step S600 and the predicted travel pattern of one acceleration / deceleration adopted in step S640 or S660. decide. Here as in step S380 of FIG. 5, the vehicle is after traveling selected link at a predetermined running speed, until it reaches the end node N _b of selectable link, the nonce acceleration and deceleration times determined in step S600 A travel pattern in which the acceleration / deceleration predicted travel pattern is repeated is determined as the predicted travel pattern of the host vehicle in the selected link. That is, when executing step S640, predicts a travel pattern of the vehicle based on the maximum speed V _max of the deceleration time calculated in step S620. On the other hand, when step S660 is executed, the traveling pattern of the host vehicle is predicted based on the constant traveling speed V _const determined in step S610.

In the above step S670, unlike the case of step S380 in FIG. 5, treated as the vehicle is always stopped at a starting node N _a choice link. That is, without performing the superimposition of the travel pattern corresponding to the stop probability P _a start end node N _a as described above, to determine the estimated travel patterns.

In step S680, the control unit 10, as in step S560 of FIG. 7, whether the distance from the starting node N _a selection links to jam the top node is the target intersection is the predetermined threshold value D ₂ or more Determine. The threshold D ₂ may be the same value as the threshold value D ₁ of the above, the value may be different. When the distance from the starting node N _a selection links to jam the head node is the threshold D ₂ or more, the process proceeds to step S690.

  In step S690, the control unit 10 performs a waveform rounding process similar to that in step S570 in FIG. 7 on the predicted travel pattern determined in step S670. This waveform rounding process may be performed as necessary as in the case of FIG. 7, and may be omitted if unnecessary.

If step S690 is performed, the control part 10 will complete | finish the flowchart of FIG. 8, and will return to the flowchart of FIG. The distance from the starting node N _a selection links to jam the top node in step S680 is a case where it is determined to be less than the threshold value D _2, and ends the flow chart of FIG. 8 without executing step S690. As described above, the temporary stop travel pattern prediction process at the time of traffic jam is executed.

FIG. 28 shows an example of a predicted travel pattern obtained by the above-described temporarily stopped travel pattern prediction process during a traffic jam. In the predicted travel pattern of FIG. 28, the number of times of acceleration / deceleration calculated in step S600 is 5, and according to this, the prediction pattern in which a single acceleration / deceleration travel pattern as shown in FIG. An example of a running pattern is shown. In addition, the time from the time T _a to time T _b, is a link travel time T of the selected link. Therefore, a predicted traveling pattern as shown in FIG. 28 can be obtained by equally allocating and arranging each traveling pattern of acceleration / deceleration within the range of the link travel time T.

In FIG. 28, V _max ≦ V _const is determined in step S630 of FIG. 8, and step S640 is executed accordingly, whereby the maximum traveling speed V _max during acceleration and deceleration as shown in FIG. The example of the prediction driving | running pattern at the time of employ | adopting the driving | running | working pattern based on is shown. However, if it is determined in step S630 that V _max > V _const and step S660 is executed accordingly, a travel pattern based on the constant travel speed V _const as shown in FIG. Thus, a predicted traveling pattern having a shape different from that in FIG. 28 is obtained. That is, as shown in FIG. 27, a predicted traveling pattern in which the speed of the host vehicle periodically changes between 0 and a constant traveling speed V _const is obtained.

  By performing the processing as described above for each link extracted as the target link, and connecting the obtained predicted travel patterns according to the order of the links, the predicted travel pattern of the entire target link can be obtained. can get. An example of the predicted travel pattern of the entire target link obtained in this way is shown in FIG. In FIG. 29, the example of the prediction driving | running | working pattern obtained by making each link between nodes 41-49 into object link is shown.

  In FIG. 29, nodes 42, 43, 45 to 49 are traffic signal intersections, and node 44 is a temporary stop intersection. Moreover, between nodes 41-43 and between nodes 44-46 are favorable, and between nodes 46-47 is congested. There is a traffic jam between the nodes 43 to 44 and the nodes 47 to 49, and the nodes 44 and 49 are estimated to be the heads of the traffic jams. By performing the traveling pattern prediction process as described above under such conditions, for example, a traveling pattern as shown in FIG. 29 is estimated, and based on this, an estimated energy consumption value is obtained for each link.

  According to the embodiment described above, the following operational effects can be obtained.

(1) The navigation device 1 selects one of the plurality of links constituting the road network as a selected link by the process of the control unit 10 (step S20), and selects any of the intersections existing in front of the selected link. As a target intersection, a traffic light traveling pattern prediction process during smoothing (step S60), a temporary traveling traveling pattern prediction process during smoothing (step S80), a passing traveling pattern prediction process during smoothing (step S90), and a traffic light during a traffic jam A travel pattern prediction process (step S120) or a temporarily stopped travel pattern prediction process (step S140) in a traffic jam is executed. Alternatively, a preset default travel pattern is applied as the travel pattern of the selected link (step S150). As a result, the traveling pattern corresponding to the road condition of the selected link and the type of the target intersection is predicted as the traveling pattern of the vehicle on the selected link. Since it did in this way, the driving | running | working pattern of a vehicle can be correctly estimated in consideration of the road condition, the presence or absence of a traffic light at an intersection, and the like.

(2) When the road condition of the selected link is smooth and the target intersection is a traffic light intersection where a traffic light is installed, the control unit 10 executes a traffic light traveling pattern prediction process during smoothness in step S60. This steadily during traffic driving pattern prediction processing, based on the stop probability P _a at the beginning node N _a stop probability P _b and select links in the terminal node N _b of the selection links striking target intersection, the running pattern of the selected link Predict. Since it did in this way, the driving | running | working pattern of a vehicle when the road condition of a selection link is smooth and a target intersection is a traffic signal intersection in which the traffic signal is installed can be estimated correctly.

(3) The HDD 13 of the navigation device 1 stores in advance a red signal stop probability table representing stop probabilities corresponding to combinations of road types of roads that intersect each other at an intersection. The control unit 10 refers to the red signal stop probability table stored in the HDD 13 in the traffic light traveling pattern prediction process at the time of smoothness, and based on this, the stop probability P _{b at} the terminal node N _b of the selected link that is the target intersection. Is determined (step S200). As a result, the stop probability at the target intersection can be determined to some extent accurately and easily.

(4) In the traffic light traveling pattern prediction process at the time of smoothness, the control unit 10 predicts each traveling pattern predicted when the host vehicle stops and does not stop at each of the start node N _a and the terminal node N _b of the selected link. Are calculated as the probability of occurrence of each scenario in the selected link (step S220). That is, the first travel pattern that does not stop at both the start node N _a and the end node N _b , the second travel pattern that stops at the end node N _b without stopping at the start node N _a , and the start node N _a a third driving pattern does not stop at the end node N _b in stop, for a fourth running pattern stops at both start end node N _a and the terminating node N _b, a stop probability P _b at the terminal node N _b the probability based on the stop probability P _a at the beginning node N _a is calculated. Then, the first traveling pattern, the second traveling pattern, the third traveling pattern, and the fourth traveling pattern are superimposed at a ratio corresponding to each occurrence probability calculated in step S220, thereby traveling the selected link. A pattern is predicted (step S250). Thereby, the traveling pattern of the selected link can be accurately predicted in consideration of the presence or absence of a vehicle stop at each of the start and end of the selected link.

(5) When the road condition of the selected link is good and the target intersection is a temporary stop intersection, the control unit 10 executes a temporary stop travel pattern prediction process during normal operation in step S80. The pause driving pattern prediction processing in steady time, the acceleration and deceleration times determined based on the number of the preceding vehicle to pause at the end node N _b of the selection links striking the target intersection (step S300), the running pattern of the selected link Predict. Since it did in this way, the driving | running | working pattern of a vehicle when the road condition of a selection link is favorable and the object intersection is a temporary stop intersection can be estimated correctly.

(6) The HDD 13 of the navigation device 1 stores in advance a preceding number table that represents the number of preceding vehicles according to the combination of road types of roads that intersect each other at the intersection. Preceding vehicle control unit 10, which in pause driving pattern prediction process steadily at, with reference to the preceding number table stored in the HDD 13, to pause at the end node N _b of selected link is a target intersection based on this The number of accelerations / decelerations is determined by determining the number of (step S300). Thereby, the number of times of acceleration / deceleration can be determined to some extent accurately and easily from the number of preceding vehicles at the target intersection.

(7) The HDD 13 of the navigation device 1 stores in advance a slow travel table in which the relationship between the advance distance, acceleration, and travel speed learned so far in the slow travel is tabulated. The control unit 10 determines the acceleration G of the host vehicle in the selected link based on the slow travel table stored in the HDD 13 in the temporary stop travel pattern prediction process during smooth operation (step S320). In this way, since the acceleration / deceleration in the travel pattern of the selected link is determined based on the past travel history of the host vehicle, the travel pattern is reflected to reflect the travel characteristics of the host vehicle and the driver's driving tendency. Can be predicted.

(8) The control unit 10 determines whether the selected link is based on the acceleration G determined in step S320 and the predetermined stop occupation length _Lcar per vehicle in the temporary stop travel pattern prediction process during smooth running. calculating a maximum speed _{V max} of the acceleration or deceleration in the running pattern (step S330). Then, it is determined whether or not the calculated maximum traveling speed V _max during acceleration / deceleration is equal to or lower than a predetermined constant traveling speed V _const (step S340). As a result, when the maximum traveling speed V _max during acceleration / deceleration is equal to or less than the constant traveling speed V _const , the traveling pattern of the selected link is predicted by adopting a traveling pattern based on the maximum traveling speed V _max during acceleration / deceleration. (Step S350). On the other hand, if the maximum running speed _{V max} of the acceleration and deceleration is larger than a predetermined traveling speed _{V const} predicts the travel pattern of the selected link by adopting the running pattern based on a constant running speed _{V const} (step S370). Since it did in this way, a driving | running pattern is correctly estimated in consideration of the situation where the own vehicle follows the preceding vehicle and repeats the temporary stop and slowing down and travels along the selected link toward the target intersection. be able to.

(9) The control unit 10 refers to the slow travel table stored in the HDD 13 in the temporary stop travel pattern prediction process at the time of smoothing, and determines a constant travel speed V _const based on this (step S320). In this way, since the constant travel speed V _const is determined based on the past travel history of the _host vehicle, the travel characteristics of the host vehicle and the driving tendency of the driver can be reflected in the prediction of the travel pattern. it can.

(10) The control unit 10 causes the temporary stop driving pattern prediction process well during a travel pattern when not stop at the start node N _a selection link, a travel pattern when stopping at the start node N _a selection link and by superimposing in a ratio corresponding to the stop probability P _a at the beginning node N _a selection link, and to predict the travel pattern of the selected link (step S380). As a result, the traveling pattern of the selected link can be accurately predicted in consideration of whether or not the vehicle has stopped at the starting end of the selected link.

(11) When the road condition of the selected link is smooth and the target intersection is neither a traffic light intersection where a traffic light is installed nor a temporary stop intersection, the control unit 10 performs a passing travel pattern prediction process during smoothness in step S90. Run. In passing driving pattern prediction processing in this well at the time of a running pattern of scenario # 1 representing a running pattern with a constant running speed from starting node N _a selection links to the end node N _b, and increases according to a predetermined acceleration G after a running pattern of scenario # 3 representing the traveling pattern by the running speed becomes constant, by superimposing in a ratio corresponding to the stop probability P _a at the beginning node N _a selection link predicts the travel pattern of the selected link (Steps S450 and S470). Since it did in this way, the driving | running | working pattern of a vehicle when the road condition of a selection link is smooth and a target intersection is neither a traffic signal intersection in which a traffic signal is installed nor a temporary stop intersection can be estimated correctly.

(12) The control unit 10 executes the traffic light traveling pattern prediction process at the time of smoothing at Step S60, the case of executing the temporary stop traveling pattern prediction process at the time of smoothing at Step S80, or the passing travel at the time of smoothing at Step S90. when performing pattern prediction processing was set as the target intersection an intersection which corresponds to the end node N _b choice link. Thereby, an intersection suitable for predicting a running pattern when the road condition of the selected link is smooth can be set as the target intersection.

(13) When the road condition of the selected link is congested or congested, and the target intersection is a traffic signal intersection where a traffic signal is installed, the control unit 10 executes traffic light traveling pattern prediction processing at the time of traffic jam in step S120. . In the traffic light traveling pattern prediction process at the time of traffic congestion, the number of accelerations / decelerations N _adv is calculated based on the green signal time T _adv at the head node of the traffic congestion corresponding to the target intersection (step S510), and the traveling pattern of the selected link is predicted. Since it did in this way, the driving | running | working pattern of a vehicle when the road condition of a selection link is congestion or traffic congestion, and a target intersection is a traffic signal intersection where the traffic signal is installed can be estimated correctly.

(14) The HDD 13 of the navigation device 1 stores in advance a green signal time table that indicates a green signal time corresponding to a combination of road types of roads that intersect each other at an intersection. In the traffic light traveling pattern prediction process at the time of traffic jam, the control unit 10 refers to the green signal time table stored in the HDD 13, and based on this, determines the green signal time T _adv at the traffic jam head node that is the target intersection (step S500). ). Thereby, the green light time at the target intersection can be determined to some extent accurately and easily.

(15) In the traffic light traveling pattern prediction process at the time of traffic jam, the control unit 10 refers to the slow traveling table stored in the HDD 13 and determines the acceleration G of the host vehicle on the selected link based on this (step S520). . In this way, since the acceleration / deceleration in the travel pattern of the selected link is determined based on the past travel history of the host vehicle, the travel pattern is reflected to reflect the travel characteristics of the host vehicle and the driver's driving tendency. Can be predicted.

(16) In the traffic light traveling pattern prediction process at the time of traffic jam, the control unit 10 determines the green signal time T _adv determined in step S500, the acceleration G determined in step S520, the stop occupation length L _car per vehicle and the unit time Based on the number of vehicles passing through N _car , a constant traveling speed V _const in the traveling pattern of the selected link is calculated (step S530). As a result, while the traffic light at the traffic congestion leading node is green, the driving pattern is accurately determined in consideration of the situation where the host vehicle follows the preceding vehicle and travels on the selected link while repeatedly stopping and slowing down. Can be predicted.

(17) When the road condition of the selected link is congested or congested and the target intersection is a temporary stop intersection, the control unit 10 performs a temporary stop travel pattern prediction process at the time of the traffic jam in step S140. In this temporary stop travel pattern prediction process in a traffic jam, the number of accelerations / decelerations N _adv is calculated based on a link length L representing the length of the selected link and a predetermined stop occupation length L _car per vehicle ( Step S600), the traveling pattern of the selected link is predicted. Since it did in this way, the driving | running | working pattern of a vehicle when the road condition of a selection link is congestion or traffic congestion and a target intersection is a temporary stop intersection can be estimated correctly.

(18) The control unit 10 refers to the slow travel table stored in the HDD 13 in the temporary stop travel pattern prediction process at the time of traffic jam, and determines the acceleration G of the host vehicle on the selected link based on this (step S610). ). In this way, since the acceleration / deceleration in the travel pattern of the selected link is determined based on the past travel history of the host vehicle, the travel pattern is reflected to reflect the travel characteristics of the host vehicle and the driver's driving tendency. Can be predicted.

(19) In the temporary stop travel pattern prediction process at the time of traffic jam, the control unit 10 determines the selected link based on the acceleration G determined in step S610 and the predetermined stop occupation length _Lcar per vehicle. calculating a maximum speed _{V max} of the acceleration or deceleration in the running pattern (step S620). Then, it is determined whether or not the calculated maximum traveling speed V _max during acceleration / deceleration is equal to or lower than a predetermined constant traveling speed V _const (step S630). As a result, when the maximum traveling speed V _max during acceleration / deceleration is equal to or less than the constant traveling speed V _const , the traveling pattern of the selected link is predicted by adopting a traveling pattern based on the maximum traveling speed V _max during acceleration / deceleration. (Step S640). On the other hand, if the maximum running speed _{V max} of the acceleration and deceleration is larger than a predetermined traveling speed _{V const} predicts the travel pattern of the selected link by adopting the running pattern based on a constant running speed _{V const} (step S660). Since it did in this way, a driving | running pattern is correctly estimated in consideration of the situation where the own vehicle follows the preceding vehicle and repeats the temporary stop and slowing down and travels along the selected link toward the target intersection. be able to.

(20) The control unit 10 determines the constant traveling speed V _const based on the slow traveling table stored in the HDD 13 in the temporarily stopped traveling pattern prediction process at the time of traffic jam (step S610). In this way, since the constant travel speed V _const is determined based on the past travel history of the _host vehicle, the travel characteristics of the host vehicle and the driving tendency of the driver can be reflected in the prediction of the travel pattern. it can.

(21) When the control unit 10 executes the traffic light traveling pattern prediction process at the time of traffic jam at step S120, or when executing the temporary stop travel pattern prediction process at the time of traffic jam at step S140, the selected link is included. The intersection which hits the head of the existing congestion or traffic jam is estimated (step S100), and the intersection is set as the target intersection. Thereby, an intersection suitable for predicting a traveling pattern when the road condition of the selected link is congested or congested can be set as the target intersection.

(22) control unit 10, traffic driving pattern prediction processing when congestion step S120, or in pause driving pattern prediction processing when congestion step S140, congestion leading node is the target intersection from the starting node N _a selection link distance to it is determined whether the predetermined threshold value _{D 1} or _{D 2} or more (step S560, S680). As a result, when the distance from the starting node _{N a} to jam the top node is the threshold value _{D 1} or _{D 2} or more, performs a rounding process for reducing the fluctuation band of the running pattern (S570, S690). Because of this, when the distance from the host vehicle to the head of the traffic jam is some distance away, the running speed dull that occurs when the host vehicle does not sufficiently follow the acceleration / deceleration movement of the preceding vehicle at the head of the traffic jam is considered. Thus, the traveling pattern can be predicted.

  In the above-described embodiment, the application example in the navigation device 1 that uses the travel pattern predicted by the travel pattern prediction process for searching for the minimum fuel consumption route has been described. However, the present invention is applied to an in-vehicle device other than the navigation device. You may apply. Alternatively, the present invention can be applied to a device other than the in-vehicle device such as a personal computer. The present invention can be applied to any device as long as it can predict the traveling pattern of the vehicle at an arbitrary target link using the traveling pattern prediction process as described above.

  The embodiment and various modifications described above are merely examples, and the present invention is not limited to these contents as long as the features of the invention are not impaired.

1: navigation device, 10: control unit, 11: vibration gyro, 12: vehicle speed sensor,
13: HDD, 14: GPS receiver, 15: VICS information receiver, 16: Display monitor,
17: Speaker, 18: Input device

Claims (21)

A link selection means for selecting one of the links constituting the road network as a selected link;
Travel pattern predicting means for predicting a travel pattern according to the road condition of the selected link and the type of the target intersection as a travel pattern of the vehicle on the selected link, with any of the intersections existing in front of the selected link as a target intersection. It equipped with a door,
When the road condition of the selected link is smooth and the target intersection is a traffic signal intersection where a traffic signal is installed, the travel pattern predicting means determines a stop probability at the target intersection and a stop probability at the start of the selected link. based on the bets, driving pattern prediction apparatus characterized that you predict the travel pattern. In the travel pattern prediction device according to claim 1 ,
A red signal stop probability table representing stop probabilities corresponding to combinations of road types of roads intersecting each other at an intersection is stored in advance,
The travel pattern predicting means determines a stop probability at the target intersection based on the red signal stop probability table. In the travel pattern prediction device according to claim 1 or 2 ,
The travel pattern prediction means includes a first travel pattern that does not stop at both the start end of the selected link and the target intersection, and a second travel pattern that stops at the target intersection without stopping at the start end of the selected link; , Stop at the target intersection for the third travel pattern that stops at the start of the selected link and does not stop at the target intersection and the fourth travel pattern that stops at both the start of the selected link and the target intersection Calculating the occurrence probability based on the probability and the stop probability at the start of the selected link,
Predicting the travel pattern by superimposing the first travel pattern, the second travel pattern, the third travel pattern, and the fourth travel pattern at a rate corresponding to each occurrence probability. A travel pattern predicting device characterized by the above. In the travel pattern prediction device according to any one of claims 1 to 3 ,
The traveling pattern predicting means determines the number of times of acceleration / deceleration based on the number of preceding vehicles temporarily stopped at the target intersection when the road condition of the selected link is smooth and the target intersection is a temporary stop intersection. A travel pattern predicting apparatus for predicting the travel pattern. In the traveling pattern prediction device according to claim 4 ,
A preceding number table representing the number of preceding vehicles according to the combination of road types of roads intersecting each other at an intersection is stored in advance,
The travel pattern predicting means determines the number of preceding vehicles to be temporarily stopped at the target intersection based on the preceding number table. In the traveling pattern prediction device according to claim 4 or 5 ,
The travel pattern predicting unit determines acceleration / deceleration in the travel pattern based on a past travel history of the vehicle. In the traveling pattern prediction device according to claim 6 ,
The travel pattern predicting means calculates a maximum travel speed during acceleration / deceleration in the travel pattern based on the acceleration / deceleration and a predetermined stop occupation length per vehicle,
When the maximum traveling speed at the time of acceleration / deceleration is equal to or less than a predetermined constant traveling speed, the traveling pattern is predicted based on the maximum traveling speed at the time of acceleration / deceleration,
When the maximum traveling speed at the time of acceleration / deceleration is larger than the constant traveling speed, the traveling pattern predicting apparatus predicts the traveling pattern based on the constant traveling speed. In the travel pattern prediction device according to claim 7 ,
The travel pattern predicting means determines the constant travel speed based on a past travel history of the vehicle. In the traveling pattern prediction device according to any one of claims 4 to 8 ,
The travel pattern predicting unit superimposes the travel pattern when not stopping at the start end of the selected link and the travel pattern when stopping at the start end of the selected link at a rate corresponding to the stop probability at the start end of the selected link. A travel pattern predicting apparatus that predicts the travel pattern. In the traveling pattern prediction device according to any one of claims 1 to 9 ,
The travel pattern predicting means is configured to perform a constant travel from the start to the end of the selected link when the road condition of the selected link is smooth and the target intersection is neither a traffic signal intersection nor a temporary stop intersection where a traffic light is installed. The travel pattern is predicted by superimposing the travel pattern based on the speed and the travel pattern based on the travel speed that becomes constant after increasing according to a predetermined acceleration at a rate corresponding to the stop probability at the start of the selected link. A travel pattern predicting apparatus characterized by that. In the travel pattern prediction device according to any one of claims 1 to 10 ,
The traveling pattern predicting device, wherein the traveling pattern predicting means sets the intersection corresponding to the end of the selected link as the target intersection. In the traveling pattern prediction device according to any one of claims 1 to 11 ,
When the road condition of the selected link is congested or congested and the target intersection is a traffic signal intersection where a traffic signal is installed, the running pattern prediction means calculates the number of times of acceleration / deceleration based on a green signal time at the target intersection. A travel pattern prediction apparatus that calculates and predicts the travel pattern. The travel pattern prediction apparatus according to claim 12 ,
A green signal time table representing a green signal time corresponding to a combination of road types of roads intersecting each other at an intersection is stored in advance,
The traveling pattern predicting device, wherein the traveling pattern predicting means determines a green signal time at the target intersection based on the green signal time table. In the traveling pattern prediction device according to claim 12 or 13 ,
The travel pattern predicting unit determines acceleration / deceleration in the travel pattern based on a past travel history of the vehicle. The travel pattern prediction apparatus according to claim 14 ,
The travel pattern predicting means calculates a constant travel speed in the travel pattern based on the green time, the acceleration / deceleration, the stop occupation length per vehicle, and the number of vehicles passing per unit time. Travel pattern prediction device. In the traveling pattern prediction device according to any one of claims 1 to 15 ,
When the road condition of the selected link is congested or congested, and the target intersection is a temporarily stopped intersection, the travel pattern predicting means determines the length of the selected link and a predetermined stop occupation length per vehicle. A travel pattern predicting apparatus that calculates the number of times of acceleration / deceleration based on the above and predicts the travel pattern. The travel pattern prediction apparatus according to claim 16 ,
The travel pattern predicting unit determines acceleration / deceleration in the travel pattern based on a past travel history of the vehicle. The travel pattern prediction apparatus according to claim 17 ,
The travel pattern predicting means calculates a maximum travel speed during acceleration / deceleration in the travel pattern based on the acceleration / deceleration and a predetermined stop occupation length per vehicle,
When the maximum traveling speed at the time of acceleration / deceleration is equal to or less than a predetermined constant traveling speed, the traveling pattern is predicted based on the maximum traveling speed at the time of acceleration / deceleration,
When the maximum traveling speed at the time of acceleration / deceleration is larger than the constant traveling speed, the traveling pattern predicting apparatus predicts the traveling pattern based on the constant traveling speed. The travel pattern prediction apparatus according to claim 18 ,
The travel pattern predicting means determines the constant travel speed based on a past travel history of the vehicle. In the traveling pattern prediction device according to any one of claims 12 to 19 ,
The travel pattern predicting unit is characterized in that an intersection corresponding to the head of a congestion or a traffic jam including the selected link is set as the target intersection. The travel pattern prediction apparatus according to any one of claims 12 to 20 ,
The traveling pattern predicting means performs a rounding process for reducing a variation width of the traveling pattern when a distance from a starting end of the selected link to the target intersection is a predetermined threshold value or more. Travel pattern prediction device.