JP6168811B2 - Vehicle control section learning system, method and program - Google Patents

Description

  The present invention relates to a vehicle control section learning system, method, and program for learning a vehicle control section.

  2. Description of the Related Art Conventionally, a technique for learning about a situation where driving assistance should be performed in order to perform driving assistance is known. For example, in Patent Document 1, learning should be performed by associating a deceleration operation with a road environment, and should be executed when approaching a point where the deceleration operation is performed according to the road environment (for example, a temporary stop intersection, a signalized intersection, etc.) A technique for determining a driving assistance mode is disclosed.

JP 2011-227833 A

In the above-described prior art, there is a case where normal driving assistance cannot be performed when an intersection exists before the point where the deceleration operation is performed. In other words, driving assistance may be performed as long as it passes through an existing intersection before reaching the point where the deceleration operation is performed and the point where the deceleration operation is performed, but before reaching the point where the deceleration operation is performed Driving assistance should not be performed as long as the vehicle travels in a different direction from the point where the deceleration operation is performed after passing through the intersection. Therefore, even if it associates with the point where deceleration operation is performed and the deceleration operation as in the prior art, accurate control cannot be performed unless the route for performing the deceleration operation is learned. However, if a point where driving assistance should be performed and a route are associated with each other, the amount of information to be learned becomes extremely large.
The present invention has been made in view of the above problems, and provides a technique capable of suppressing the amount of information to be learned while learning information that can accurately determine whether or not to execute vehicle control. With the goal.

  In order to achieve the above object, in the present invention, vehicle control end point acquisition means for acquiring an end point of a section in which vehicle control should be executed based on the travel history of the vehicle, and the vehicle before reaching the end point, A feature intersection acquisition means for acquiring the intersection as a feature intersection when the vehicle does not travel on a continuous road section defined as a road before and after the intersection on a road that has traveled, and a route including the feature intersection A vehicle control section learning system comprising vehicle control section learning means for learning as a section in which vehicle control is to be executed.

  In other words, if the end point of the section in which the vehicle control is to be performed is learned, the vehicle control is executed from the start point to the end point with a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle. It is possible to control the vehicle by regarding it as a section to be performed. In this configuration, according to the present invention, the route is learned if the vehicle does not pass through the intersection before reaching the end point of the section where the vehicle control is to be performed, and the route is acquired when the vehicle passes through the intersection. Do not learn. Here, a continuous road section defined as a road is a road section that passes when traveling along a road. For example, a continuous road section that is closest to a straight road or a continuous road section that has the same road number. A road is defined by a continuous road section having the same attribute.

  Therefore, the route when the vehicle travels along the road can be identified by clarifying the route along the road from the map information indicating the road section without learning the route itself. For this reason, if it is set as the structure which learns a path | route when the vehicle has traveled in the past which is not a road, when the vehicle travels on the learned route during vehicle control, the vehicle is based on the learning result. Can be estimated, and vehicle control can be performed in a section in which vehicle control is to be executed based on the estimation. Further, if the vehicle does not travel along the learned route during vehicle control, it can be assumed that the vehicle travels along the road, and the vehicle operation can be estimated. It can be performed. Furthermore, in the above configuration, since it is not necessary to learn the route when traveling along the road, the amount of information to be learned can be suppressed. Therefore, it is possible to suppress the amount of information to be learned while learning information that can accurately determine whether or not to execute vehicle control.

  Here, the vehicle control end point acquisition means only needs to be able to acquire the end point of the section where the vehicle control is to be executed based on the travel history of the vehicle. That is, the end point at which the vehicle control is to be ended is acquired as information for specifying the section in which the vehicle control is to be executed. If the end point is specified, it is possible to define a section where vehicle control should be executed starting from a point at a predetermined distance from the end point to the opposite side of the vehicle traveling direction, so at least the end point is learned. By doing so, it is sufficient that the section in which the vehicle control is to be executed is specified in combination with the specified distance, but of course at a point of a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle. A certain start point and end point may be learned as a section in which vehicle control is to be executed. In addition, in the configuration in which the section is defined by the start point of the section where the vehicle control is to be executed and the distance between the sections, the end point of the section where the vehicle control is to be executed is specified by the start point and the distance. It is substantially equivalent to the learned configuration.

  Moreover, the end point should just be acquired based on the driving | running | working log | history of a vehicle. That is, since the vehicle control is performed in order to provide support for improving the past operation of the vehicle, the section that should have room for improvement in the past operation of the vehicle is clarified based on the travel history of the vehicle, and at least What is necessary is just to specify the end point of the area which should perform vehicle control. Various types of control can be assumed for vehicle control, and examples include deceleration control and acceleration control. If it is the former, what is necessary is just to set it as the structure which acquires the deceleration end point of a vehicle as an end point of the area which should perform vehicle control, for example. Moreover, if it is the latter, what is necessary is just to set it as the structure which acquires the acceleration end point of a vehicle as an end point of the area which should perform vehicle control, for example.

  The characteristic intersection acquisition means determines that the intersection is a characteristic intersection when the vehicle has traveled before reaching the end point and the vehicle does not travel on a continuous road section defined as a road before and after the intersection. It only has to be acquired. That is, if there are multiple road sections connected to the intersection, and it is possible to enter the intersection from one road section and exit to the other road section, both road sections are continuous and continuous In a situation where there is an intersection defined as a road section that is a road in the road section to be performed, it is only necessary to determine whether or not the vehicle has traveled along the road from the driving history of the vehicle and acquire a characteristic intersection .

  The vehicle control section learning means only needs to be able to learn a route including a characteristic intersection as a section in which vehicle control is to be executed. That is, when one or more characteristic intersections are included in the road that the vehicle traveled before reaching the end point, vehicle control should be executed by learning a route that includes the characteristic intersection and ends at the end point What is necessary is just to be able to learn a section. The number of feature intersections is not limited, but the section where vehicle control is to be executed is usually not an excessively long section, and an upper limit can be set for the distance. good.

  Note that various methods can be adopted as a route definition method. For example, the route indicated by the vehicle control section learning means and the road section on which the vehicle traveled after passing through the feature intersection It is good also as a structure learned as an area which should perform control. That is, if the road section that has traveled after the characteristic intersection is identified, at least the route that the vehicle has traveled after the characteristic intersection is identified. Therefore, such a definition exists on the opposite side of the traveling direction of the vehicle from the end point. If the feature intersection is adopted, a route passing through the feature intersection to the end point can be specified for a plurality of feature intersections.

  In this configuration, when the vehicle travels along the road, the route and the characteristic intersection are not learned (the end point can be learned), so the amount of information necessary for learning when the vehicle travels along the road is small. It is suppressed. In addition, various configurations can be adopted as the configuration for learning the characteristic intersection, and the distance from the end point of the section where the vehicle control should be executed and the distance from the start point of the section where the vehicle control should be executed are determined. You may learn, and you may learn the coordinate of a feature intersection.

  As a learning mode for learning about a case where the vehicle travels along a road while suppressing a necessary amount of information, various methods can be adopted. For example, if the vehicle does not pass through a characteristic intersection, the process ends. It is good also as a structure which learns the start point and end point which are points of predetermined distance on the opposite side of the advancing direction of a vehicle from a point as an area which should perform vehicle control. In this configuration, when the vehicle does not pass through the feature intersection, the feature intersection and the route traveled by the vehicle after the feature intersection are not learned. Therefore, the amount of information can be suppressed as compared with the case where these are learned. .

  Further, when a predetermined number of characteristic intersections are included in a section within a predetermined distance, the vehicle control section learning means learns a route including the predetermined number of characteristic intersections as a section where the vehicle control is to be performed, and is within the predetermined distance. If the section includes more than a predetermined number of characteristic intersections, a route of a section shorter than a predetermined distance including a predetermined number of characteristic intersections may be learned as a section on which vehicle control is to be executed. That is, when there are a large number of characteristic intersections, it is preferable that the same vehicle control is not continuously performed, for example, when there are many intersections that have passed without being on the road and the vehicle does not go straight at the intersections. Therefore, by setting an upper limit of a predetermined number to the number of feature intersections that can exist in the section where the vehicle control is to be executed, a section having an appropriate distance is learned as the section where the vehicle control is to be executed.

  Further, the method of learning the route as a vehicle control section when the vehicle does not travel on a continuous road section defined as a road before and after the intersection as in the present invention can be applied as a program or a method. is there. In addition, the system, program, and method as described above may be realized as a single device, or may be realized by using components shared with each unit provided in the vehicle when realized by a plurality of devices. It can be assumed and includes various aspects. For example, it is possible to provide a navigation system, method, and program including the above-described devices. Further, some changes may be made as appropriate, such as a part of software and a part of hardware. Furthermore, the invention can be realized as a recording medium for a program for controlling the system. Of course, the software recording medium may be a magnetic recording medium, a magneto-optical recording medium, or any recording medium to be developed in the future.

It is a block diagram which shows the navigation system containing a vehicle control area learning system. It is a flowchart which shows a learning process. It is a flowchart which shows a vehicle control process. It is a figure which shows the example of a road.

Here, embodiments of the present invention will be described in the following order.
(1) Configuration of navigation terminal:
(2) Learning process:
(3) Vehicle control processing:
(4) Other embodiments:

(1) Configuration of navigation terminal:
FIG. 1 is a block diagram showing a configuration of a vehicle control section learning system mounted on a vehicle. In the present embodiment, the vehicle control section learning system is realized by the navigation system 10. The navigation system 10 includes a control unit 20 including a CPU, a RAM, a ROM, and the like, and a program stored in the ROM can be executed by the control unit 20. In this embodiment, a navigation program can be executed as one of the programs. The navigation program is a program that causes the control unit 20 to realize a function of displaying a map including the current position of the vehicle on the display unit of the navigation terminal and guiding the driver to the destination. The navigation program includes various programs used in the traveling process. In the present embodiment, the vehicle that causes the control unit 20 to implement vehicle control that applies a braking force to the vehicle in a section where the vehicle has decelerated in the past. A control program 21 is provided.

  Map information 30 a is recorded on the recording medium 30. In addition, the learning information 30b is recorded on the recording medium 30 during the traveling process of the vehicle. The map information 30a includes node data indicating the position of the node corresponding to the end point of the road on which the vehicle travels, shape interpolation point data indicating the position of the shape interpolation point for specifying the shape of the road between the nodes, the nodes The link data indicating the connection of the link is included.

  Furthermore, learning information 30 b is recorded on the recording medium 30 during the operation process of the vehicle control program 21. The learning information 30b is information indicating a section in which the vehicle control is to be executed. In the present embodiment, information indicating at least a start point and an end point in each of the sections in which the vehicle control is to be executed is the learning information 30b. Is defined. In the present embodiment, when the information defined as the learning information 30b differs depending on the travel history of the vehicle, and the travel history of the vehicle indicates that it has passed through the intersection, the start point and the end point Is defined as learning information 30b. On the other hand, when the vehicle travel history indicates that the vehicle has not passed through the intersection, the intersection is regarded as a feature intersection, and in addition to the information indicating the start point and the end point, the feature intersection and the feature intersection Information indicating a route indicated by a road section on which the vehicle has traveled after passing through is included in the learning information 30b. In this embodiment, the information indicating the characteristic intersection is defined by the distance from the start point of the section where the vehicle control is to be executed, and the road section where the vehicle travels after passing the characteristic intersection indicates the road section. Defined by link data (link ID).

  The vehicle according to the present embodiment controls the GPS receiver 41, the vehicle speed sensor 42, the gyro sensor 43, the user I / F unit 44, the friction braking unit 45, the internal combustion engine 49b, the gear mechanism 49c, the output shaft 49d, and the generator 47. A unit 47a, a motor 48, a control unit 48a, and a battery 49a are provided. The vehicle according to the present embodiment is a hybrid vehicle that is driven by the rotational driving force of the internal combustion engine 49b and the motor 48 being transmitted to the output shaft 49d by the gear mechanism 49c. Therefore, an engine brake that transmits the rotational driving force of the output shaft 49d to the internal combustion engine 49b by the gear mechanism 49c and a regenerative power that transmits the rotational driving force of the output shaft 49d to the generator 47 by the gear mechanism 49c and accumulates electric power in the battery 49a. The vehicle can also be braked by a brake. Of course, the energy distribution to the internal combustion engine 49b, the generator 47, and the motor 48 can be adjusted by the control unit 47a and the control unit 48a. In this embodiment, the control unit 20 is either the control unit 47a or the control unit 48a or By outputting control signals to both, the ratio between the regenerative brake and the engine brake can be adjusted.

  The GPS receiver 41 receives a radio wave from a GPS satellite and outputs a signal indicating a signal for calculating the current location of the vehicle via an interface (not shown). The control unit 20 acquires this signal and acquires the current location of the vehicle. The vehicle speed sensor 42 outputs a signal corresponding to the rotational speed of the wheels provided in the vehicle. The control unit 20 acquires this signal via an interface (not shown) and acquires the vehicle speed. The gyro sensor 43 detects angular acceleration about turning in the horizontal plane of the vehicle, and outputs a signal corresponding to the direction of the vehicle. The control unit 20 acquires this signal and acquires the traveling direction of the vehicle. The vehicle speed sensor 42, the gyro sensor 43, and the like are used to specify the travel locus of the vehicle. In the present embodiment, the current location is specified based on the departure location and the travel locus of the vehicle. Based on the output signal of the GPS receiver 41, the current location of the vehicle specified based on the above is corrected.

  The user I / F unit 44 is an interface unit for inputting a driver's instruction and providing various information to the driver. The user I / F unit 44 includes an input unit such as a display unit and a switch (not shown), a switch, a speaker, and the like. Audio output unit. The user I / F unit 44 receives a control signal from the control unit 20 and displays an image for performing various guidance on the touch panel display.

  The friction braking unit 45 includes a device that controls the pressure of a wheel cylinder that adjusts the degree of deceleration by a brake mounted on a vehicle wheel, and the control unit 20 outputs a control signal to the friction braking unit 45. It is possible to adjust the pressure of the wheel cylinder. Therefore, when the control unit 20 outputs a control signal to the friction braking unit 45 to increase the pressure of the wheel cylinder, the braking force by the brake increases and the vehicle is decelerated.

  In the present embodiment, the output signal of the vehicle speed sensor 42 for each vehicle position and the operation state of the friction braking unit 45 are temporarily recorded in the RAM as a travel history (until the navigation system 10 is turned off) It can be referred to. Therefore, the control unit 20 can identify the vehicle speed for each vehicle position and the driver's braking intention using the friction braking unit 45 by referring to the travel history.

  In the present embodiment, the vehicle control unit 21d included in the vehicle control program 21 is a program module that causes the control unit 20 to realize a function of controlling the braking force acting on the vehicle, and controls the vehicle based on the learning information 30b. The section in which to execute is specified, and the braking force is applied to the vehicle. Various modes can be adopted for the vehicle control, and in the present embodiment, the control unit 20 uses a braking force for decelerating the vehicle speed of the past vehicle at the end point of the zone where the vehicle control is to be performed. Act on the vehicle. At this time, the control unit 20 controls the control unit 47a and the control unit 48a so that the braking force is generated by the regenerative braking. When the braking force exceeds the maximum value of the braking force generated by the regenerative braking, the control unit 20 compensates the braking force by outputting a control signal to the friction braking unit 45 to apply the friction braking force to the vehicle.

  In order to execute such vehicle control, the learning information 30b needs to be recorded on the recording medium 30 before the vehicle control is executed, and the vehicle control program 21 in the present embodiment has a capacity of the learning information 30b. The control unit 20 can realize the function of acquiring the learning information 30b while suppressing the above. In order to cause the control unit 20 to realize this function, the vehicle control program 21 includes a vehicle control end point acquisition unit 21a, a characteristic intersection acquisition unit 21b, and a vehicle control section learning unit 21c.

  The vehicle control end point acquisition unit 21a is a program module that causes the control unit 20 to realize a function of acquiring an end point of a section in which vehicle control is to be performed based on the vehicle travel history. In the present embodiment, the control unit 20 sets a deceleration end point (a point at which the vehicle is stopped or almost stopped) at which the vehicle is decelerated to a predetermined vehicle speed or less as an end point of the section where the vehicle control is to be executed. get. For this reason, the control unit 20 refers to the travel history and map information 30a recorded in the RAM, and the vehicle speed is changed from a state where the vehicle speed is greater than or equal to a threshold value to a predetermined vehicle speed or less in a state where the braking force is applied to the vehicle. The section which became is specified. And the control part 20 specifies the endpoint which exists ahead of the advancing direction of a vehicle among the endpoints of the said area as an end point.

  Through the above processing, the control unit 20 acquires an end point at which the vehicle control should be ended as information for specifying a section in which the vehicle control is to be executed. If the end point is specified, it is possible to define a section where vehicle control should be executed starting from a point at a predetermined distance from the end point to the opposite side of the vehicle traveling direction, so at least the end point is learned. By doing so, it is only necessary that the section in which vehicle control is to be executed is specified in combination with the specified distance. In the present embodiment, when the vehicle falls below a predetermined vehicle speed from a state equal to or higher than a threshold, the control unit 20 includes a start point that is a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle. The end point is acquired as a section in which the vehicle control is to be executed, and the learning information 30b is defined (details will be described later).

  The characteristic intersection acquisition unit 21b determines the characteristic intersection when the vehicle has traveled before reaching the end point and the vehicle does not travel on a continuous road section defined as a road before and after the intersection. This is a program module that causes the control unit 20 to realize the function acquired as follows. That is, the control unit 20 refers to the traveling history of the vehicle and the map information 30a, and specifies an intersection existing on the road on which the vehicle has traveled. Furthermore, when the control unit 20 can enter the intersection from one road section and exit to the other road section among the plurality of road sections connected to the specified intersection, both road sections Are considered continuous. Furthermore, the control unit 20 determines that both road sections are road sections that are roads when the difference in the traveling direction of the vehicle before and after passing through the intersection is the smallest among the consecutive road sections (when the vehicle is closest to the straight line). Define.

  Further, the control unit 20 determines that the vehicle has traveled along the road when the vehicle travels on both of the road sections defined as roads at an intersection existing on the road on which the vehicle has traveled. It is determined that the vehicle has not traveled along the road when it has not traveled along at least one of the road sections defined as the road. Then, when it is determined that the vehicle has traveled along an intersection along the road, the control unit 20 acquires the intersection as a characteristic intersection.

  The vehicle control section learning unit 21c is a program module that causes the control unit 20 to realize a function of learning a route including a characteristic intersection as a section in which vehicle control is to be executed. In the present embodiment, the control unit 20 uses, as the learning information 30b, information with different information amounts when the characteristic intersection is included in the section where the vehicle travels while decelerating before reaching the end point and when the characteristic intersection is not included. get. That is, when the vehicle does not pass through the characteristic intersection, the control unit 20 determines the start point and the end point that are a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle (referred to as the rear). It acquires as a section which should perform control, and records it in the recording medium 30 as learning information 30b.

  On the other hand, when one or more characteristic intersections are included in the road on which the vehicle traveled before reaching the end point, the control unit 20 acquires a route including the characteristic intersection and ending at the end point, and learning information 30b. To the recording medium 30. Specifically, the control unit 20 specifies the route on which the vehicle has traveled after the feature intersection by identifying the road section on which the vehicle has traveled after the feature intersection based on the travel history of the vehicle and the map information 30a. . When there are a plurality of characteristic intersections, the control unit 20 specifies a road section where the vehicle travels after the characteristic intersections for each of the plurality of characteristic intersections.

  Further, the control unit 20 specifies a start point that is a predetermined distance backward from the end point, and specifies a distance from the start point to the characteristic intersection based on the map information 30a. The control unit 20 records information indicating the start point, the end point, the distance from the start point to the characteristic intersection, and the road section on which the vehicle has traveled after the characteristic intersection as learning information 30 b and records the information in the recording medium 30. Although the number of feature intersections learned in the above configuration is not limited, the section in which the vehicle control is to be executed is usually not an excessively long section, and an upper limit can be set for the distance. When the number of feature intersections reaches a predetermined number (for example, two), the number of feature intersections exceeding the predetermined number is not learned.

  In the above configuration, since the section where the vehicle control is to be executed by the start point and the end point is learned as the learning information 30b, the vehicle control is performed in the section to the start point and the end point by referring to the learning information 30b. Can be executed. In the present embodiment, further, when the vehicle learns a route if it does not pass through the feature intersection before reaching the end point of the section where the vehicle control is to be executed, and when the vehicle passes through the feature intersection Does not learn the route.

  The route when the vehicle travels along the road can be identified by clarifying the route along the road from the map information indicating the road section without learning the route itself. That is, if the vehicle does not travel along the learned route during vehicle control, it can be assumed that the vehicle travels along the road, and the vehicle operation can be estimated. It can be performed. On the other hand, when the vehicle travels on a learned route during vehicle control, the vehicle operation can be estimated based on the learning result, and the vehicle control is performed in the section where the vehicle control is to be executed based on the estimation. Is possible. Furthermore, in the above configuration, since it is not necessary to learn a route when traveling along a road, the amount of information to be learned can be suppressed. Therefore, it is possible to suppress the amount of information to be learned while learning information that can accurately determine whether or not to execute vehicle control.

(2) Learning process:
Next, the learning process will be described in detail. FIG. 2 is a flowchart of the learning process. Here, the learning process will be described based on the example shown in FIG. In FIG. 4, the road boundary line is indicated by a solid line, and the links L _{1 to} L ₇ (road sections) are indicated by dashed-dotted arrows, and the intersections I _{1 to} I ₃ are in front of the vehicles C ₁ and C _2. An example exists. The learning process can be executed at any timing before the travel history is deleted (for example, when an instruction to turn off the navigation system 10 is given). In the learning process, the control unit 20 acquires an end point of deceleration from the travel history by the process of the vehicle control end point acquisition unit 21a (step S100). That is, the control unit 20 refers to the travel history and the map information 30a, specifies a section where the vehicle speed is equal to or lower than a predetermined vehicle speed from a state where the vehicle speed is equal to or higher than the threshold, and is ahead of the traveling direction of the vehicle among the end points of the section. Acquire existing endpoints as end points.

For example, when a history in which the vehicle C ₁ travels on the links L ₄ , L ₃ , L ₂ , and L ₁ shown in FIG. 4 and stops at the point Ps ₁ is recorded in the travel history, the control unit 20 displays the point Ps. _{Get 1} as the end point. When the vehicle C ₂ travels on the links L ₇ , L ₃ , L ₅ shown in FIG. 4 and stops at the point Ps ₂ , the control unit 20 sets the point Ps ₂ as the end point. get.

  Next, the control part 20 sets an end point as a reference | standard point by the process of the characteristic intersection acquisition part 21b (step S105). That is, in the loop processing from step S110 to S165, the control unit 20 changes the reference point to the rear intersection each time the loop is repeated, and performs analysis on the reference point. For this purpose, the end point is registered as the initial point of the reference point.

  Next, the control unit 20 determines whether or not the distance from the end point to the reference point is greater than or equal to a predetermined distance by the process of the characteristic intersection acquisition unit 21b (step S110). Here, the predetermined distance is a distance determined in advance as an upper limit distance of a section in which vehicle control is to be executed. In step S110, when it is determined that the distance from the end point to the reference point is equal to or greater than the predetermined distance, the control unit 20 skips steps S115 to S165. On the other hand, if it is not determined in step S110 that the distance from the end point to the reference point is greater than or equal to the predetermined distance, the control unit 20 executes step S115. If step S110 is executed immediately after the end point is set as the reference point in step S105, the distance from the end point to the reference point is 0. Therefore, in step S110, the distance from the end point to the reference point is set. The distance is not determined to be greater than or equal to the predetermined distance.

  In step S115, the control unit 20 determines whether or not the reference point is a feature intersection by the process of the feature intersection acquisition unit 21b. That is, the control unit 20 refers to the travel history, determines whether or not the vehicle has traveled on the road section defined as the road before and after the vehicle passes through the intersection that is the reference point, and the road defined as the road. If the vehicle is not traveling in the section, it is determined that the reference point is a characteristic intersection. If it is not determined in step S115 that the reference point is a characteristic intersection, the control unit 20 sets the intersection behind the reference point as a new reference point by the process of the characteristic intersection acquisition unit 21b (step S120), and after step S110. Repeat the process. In addition, when step S115 is performed immediately after the end point is set as the reference point in step S105, the reference point is not determined to be a characteristic intersection in step S115 because the reference point is not an intersection. If it is determined in step S115 that the reference point is a feature intersection, the control unit 20 executes step S125 by the process of the feature intersection acquisition unit 21b.

For example, when the point Ps ₁ is an end point based on the travel history for the vehicle C ₁ shown in FIG. 4 and steps S105 and after are executed, the point Ps ₁ is set as the reference point in step S105. In this case, in step S110, it is not determined that the distance from the end point to the reference point is greater than or equal to the predetermined distance. Therefore, although the determination of step S115 is performed, the reference point for the reference point is the end point Ps ₁ is not determined to be characteristic intersection. For this reason, in step S120, the reference point is set to I ₁ which is a rear intersection.

In this case, if the distance from the end point Ps ₁ to the intersection I ₁ is not greater than or equal to the predetermined distance, step S115 is executed again. Since the road section closest to the straight line before and after passing at the intersection I ₁ is the link L ₁ and the link L ₆ , traveling from the link L ₆ to the link L ₁ is a road, and the link L ₂ to the link L _1. Driving towards is not the road. Therefore, the intersection I ₁ is a feature intersection. In this case, step S125 is executed after the determination in step S115.

On the other hand, if step S105 and subsequent is executed point Ps ₂ becomes and the end point based on the travel history of a vehicle C ₂ shown in FIG. 4, in step S105, the point Ps ₂ is set as a reference point. In this case, steps S110, S115, and S120 are executed, and the reference point is set to I ₂ that is a rear intersection. If the distance from the end point Ps ₂ to the intersection I ₂ is not the predetermined distance or more, step S115 is performed through determination of the step S110, since the travel towards the link L ₃ to the link L ₅ represents is made road again Step S120 is executed and the reference point becomes the intersection I ₃ . Steps S110, S115, and S120 are repeated as long as the vehicle is traveling on a road section defined as a road when passing through an intersection. As long as the vehicle travels on a road section that is defined as a road when passing through an intersection, the distance from the end point to the reference point will eventually be a predetermined distance or more, and step S170 and subsequent steps are executed. Become.

  The description returns to the case where step S125 is executed after the determination of step S115. In this case, the control unit 20 determines whether or not three or more feature intersections have been detected by the process of the feature intersection acquisition unit 21b (step S125). That is, it is determined whether or not three characteristic intersections have been detected in the process of changing the reference point to the rear intersection and performing the loop process.

  If it is not determined in step S125 that three or more characteristic intersections have been detected, the control unit 20 determines whether the intersection determined to be the characteristic intersection in step S115 is the first intersection by the process of the characteristic intersection acquisition unit 21b. It is determined whether or not (step S130). When it is determined in step S130 that the feature intersection is the first place, the control unit 20 holds the road section ahead of the reference point by the processing of the vehicle control section learning unit 21c (step S135). That is, the link data (link ID) indicating the road section ahead of the traveling direction of the vehicle connected to the characteristic intersection set as the reference point is held in the RAM. Next, the control unit 20 acquires a point at a predetermined distance behind the end point as a start point by the process of the vehicle control section learning unit 21c (step S140), and calculates the distance from the start point to the reference point as the first point. Is stored in the RAM as the distance to the feature intersection (step S145). After step S145 is executed, the processing after step S120 is executed.

For example, when the point Ps ₁ shown in FIG. 4 is the end point and step S105 and subsequent steps are executed, and the characteristic intersection I ₁ is set as the reference point after the loop processing, there are three or more characteristic intersections in step S125. Is not determined. In this case, in step S130, it is determined that the feature intersection I ₁ is the first location. In step S135, the ID of the link L ₁ indicating the road section ahead of the reference point (characteristic intersection I ₁ ) is held in the RAM. In Figure 4, shows a predetermined distance _{(D 1 + D 2 + D} 3 + D 4) behind the position as the position Pc ₁ from the end point Ps ₁ along the travel path of the vehicle C _1, in step S140 in this case The position Pc ₁ is acquired as the start point, and the distance (D ₂ + D ₃ + D ₄ shown in FIG. ₄ ) from the start point Pc ₁ to the feature intersection I _{1 as the} reference point is the first feature point in step S145. It is held in RAM as a distance.

On the other hand, when it is not determined in step S130 that the feature intersection is the first location (when the feature intersection is the second location), the control unit 20 performs steps S150, S155 and S155 by the processing of the vehicle control section learning unit 21c. S160 is executed. These processes are the same as steps S135, S140, and S145 except that the feature intersection is the second place. For example, if the determination in step S130 is performed in a state where the point Ps ₁ shown in FIG. 4 is the end point and the characteristic intersection I ₂ is set as the reference point after the loop processing, the characteristic intersection I ₂ is the first point. Not determined to be. In this case, in step S150, the ID of the link L ₂ indicating the road section ahead of the reference point (characteristic intersection I ₂ ) is held in the RAM. In step S155, is obtained as the end a predetermined distance from the point Ps ₁ along the travel path of the vehicle _{C 1 (D 1 + D 2} + D 3 + D 4) is the starting point position behind the reference from the starting point Pc ₁ in step S160 The distance (D ₃ + D ₄ shown in FIG. ₄ ) to the feature intersection I ₂ that is a point is held in the RAM as the distance to the second feature point. When step S160 is executed, the processing after step S120 is repeated.

When it is determined in step S125 that three or more characteristic intersections have been detected, the control unit 20 sets the reference point as the start point and processes from the start point to two characteristic intersections by the processing of the vehicle control section learning unit 21c. A new distance is held (step S165). That is, the third feature intersection as the reference point is reset as the start point, and the distance from the start point to the first feature intersection and the distance from the start point to the second feature intersection are stored in the RAM. Hold. For example, in the example shown in FIG. 4, when step S110 is executed in a state where the intersection I ₃ is the reference point, the distance from the end point to the intersection I ₃ that is the reference point is a predetermined distance (D ₁ + D ₂ + D ₃ + D ₄ ) or higher. Furthermore, the intersection I ₃ that is the reference point is determined as a feature intersection in step S115, and it is determined in step S125 that there are three or more feature intersections. In this case, the position of the characteristic intersection I ₃ that is the reference point is the start point, and the distance D ₃ from the start point to the intersection I ₂ and the distance D ₂ + D ₃ from the start point to the intersection I ₁ are acquired, and the RAM Retained.

  The reason why step S165 is executed in this way is that it is determined that three characteristic intersections have been detected in step S125 without determining that the distance from the end point to the reference point is greater than or equal to the predetermined distance in step S110. This is the case. On the other hand, without determining that three characteristic intersections have been detected, the situation where it is determined in step S110 that the distance from the end point to the reference point is greater than or equal to the predetermined distance is the third characteristic intersection and the end point. This is a case where the distance is equal to or greater than a predetermined distance.

Therefore, in this embodiment, when the distance between the third feature intersection and the end point is shorter than the predetermined distance, the start point is moved forward (to the position of the third feature intersection). ) The section in which the vehicle control is to be executed is defined, and the upper limit of the number of characteristic intersections that can exist in the section in which the vehicle control is to be executed is limited to two. That is, when there are many characteristic intersections in the section where the vehicle control is to be executed, there are many intersections (intersections that have not been straightened) that have passed without being on the road. Therefore, it is preferable that the same vehicle control is not continued. For example, when traveling on links L ₄ , L ₃ , L ₂ , and L ₁ , it frequently turns at intersections, so it is not preferable to perform deceleration control over the entire area of links L ₄ , L ₃ , L ₂ , and L _1. . Therefore, in the present embodiment, by setting an upper limit of two (predetermined number) on the number of characteristic intersections that can exist in the section where the vehicle control is to be executed, an appropriate distance is set as the section where the vehicle control is to be executed. Learn the interval.

  That is, when the section within a predetermined distance includes two characteristic intersections (predetermined number), the control unit 20 sets a route including two characteristic intersections (predetermined number) as a section on which vehicle control is to be executed. learn. On the other hand, when more than two (predetermined number) of characteristic intersections are included in a section within a predetermined distance, the control unit 20 controls the vehicle in a section shorter than the predetermined distance including two (predetermined number) of characteristic intersections. Is learned as a section to be executed.

  If it is not determined in step S110 that the distance from the end point to the reference point is greater than or equal to the predetermined distance, or if step S165 is executed, the control unit 20 performs processing from the start point by the process of the vehicle control section learning unit 21c. It is determined whether or not the distance to the end point is equal to or greater than a lower limit distance (a distance that is inappropriately short for performing vehicle control) (step S170). When it is determined in step S170 that the distance from the start point to the end point is equal to or greater than the lower limit distance, the control unit 20 skips step S175 and does not perform learning.

  On the other hand, if it is not determined in step S170 that the distance from the start point to the end point is equal to or greater than the lower limit distance, the control unit 20 uses the information stored in the RAM and the start point by the processing of the vehicle control section learning unit 21c. And the end point are learned (step S175). That is, when information indicating the distance from the start point to the characteristic intersection and the road section ahead of the characteristic intersection is stored in the RAM, the control unit 20 associates these information with information indicating the start point and the end point. Then, it is recorded on the recording medium 30 as learning information 30b. When the information indicating the distance from the start point to the characteristic intersection and the road section ahead of the characteristic intersection is not stored in the RAM, the control unit 20 stores information indicating the start point and the end point in the recording medium 30 as learning information 30b. Record.

For example, in the example shown in FIG. 4, based on the travel history of the vehicle C ₁ , in step S165, the distance D ₃ from the intersection I ₃ that is the start point to the intersection I ₂ and the distance from the start point to the intersection I ₁ D ₂ + D ₃ is recorded in the RAM. If the ID of the link L ₁ is held in the RAM in step S135 and the ID of the link L ₂ is held in the RAM in 150, the start point I ₃ and the end point Ps ₁ are stored. And the distances D ₃ and D ₂ + D ₃ and the IDs of the links L ₁ and L ₂ are recorded on the recording medium 30 as learning information 30b.

In the case where a history in which the vehicle C ₂ travels on the links L ₇ , L ₃ , and L ₅ shown in FIG. 4 and stops at the point Ps ₂ is recorded in the travel history, the control unit 20 starts from the end point Ps _2. The learning information 30b indicating the start point Pc ₂ and the end point Ps ₂ is recorded in the recording medium 30 with the predetermined distance Pc ₂ as the start point. If a history in which the vehicle C ₂ travels on the links L ₇ , L ₃ , L ₂ , and L ₁ shown in FIG. 4 and stops at the point Ps ₁ is recorded in the travel history, the control unit 20 determines the start point Pc. ₂ , end point Ps ₁ , distances D ₅ + D ₃ , D ₅ + D ₃ + D _2, and IDs of links L ₁ and L ₂ are recorded on the recording medium 30 as learning information 30b.

(3) Vehicle control processing:
Next, the vehicle control process will be described in detail. FIG. 3 is a flowchart of the vehicle control process. The vehicle control process is executed in the traveling process after the learning process is executed. In the vehicle control process, the control unit 20 acquires the current location of the vehicle by the process of the vehicle control unit 21d (step S200). That is, the control unit 20 acquires the current location of the vehicle based on the output signals of the GPS signal 41, the vehicle speed sensor 42, and the gyro sensor 43.

  Next, the control unit 20 determines whether or not the vehicle has passed the start point based on the acquired current location of the vehicle by the process of the vehicle control unit 21d (step S205), and until it is determined that the vehicle has passed the start point. The processes after step S200 are repeated. In step S205, when it is determined that the vehicle has passed the start point, the control unit 20 performs deceleration control by the process of the vehicle control unit 21d (step S210). That is, the control unit 20 applies a braking force for decelerating the vehicle so that the vehicle speed is equal to or lower than a predetermined vehicle speed at the end point, and covers the braking force with the regenerative brake (the shortage is A control signal is output to the control units 47a and 48a and the friction braking unit 45 so as to be covered by another brake.

  Next, the control unit 20 determines whether or not the vehicle has turned an intersection by the process of the vehicle control unit 21d (step S215). That is, the control unit 20 refers to the map information 30a to determine whether or not the vehicle has reached the intersection. If the vehicle has not reached the intersection, the control unit 20 does not determine that the vehicle has turned around the intersection. When the vehicle has reached the intersection, the control unit 20 acquires node data indicating the end point of the road section connected to the intersection through which the vehicle has passed, and the intersection angle of each road section from the position indicated by each node data. To get. And if the intersection angle of the two road sections where the vehicle has moved before and after the intersection is not the minimum, it is considered that the vehicle has bent the intersection.

  If it is not determined in step S215 that the vehicle has made a turn at the intersection, the control unit 20 determines whether the vehicle has traveled a distance to the characteristic intersection by the process of the vehicle control unit 21d (step S220). That is, after passing through the start point, the control unit 20 acquires the travel distance from the start point based on the output signal output by the vehicle speed sensor 42 so far, and the start point indicated by the learning information 30b and the characteristic intersection (first execution) When the distance from the first characteristic intersection to the second characteristic intersection after passing through the first characteristic intersection, the distance to the characteristic intersection is obtained. It is determined that the vehicle has been driven. When the distance to the feature intersection is not recorded as the learning information 30b, the control unit 20 does not determine that the vehicle has traveled the distance to the feature intersection.

  When it is determined in step S220 that the vehicle has traveled the distance to the characteristic intersection, the control unit 20 determines whether or not the current location matches the characteristic intersection by the process of the vehicle control unit 21d (step S225). . That is, the control unit 20 refers to the learning information 30b, identifies the road section after passing the characteristic intersection, and refers to the map information 30a to identify the position of the intersection where the road section is connected. Then, it is determined whether or not the current location matches the position of the intersection. If it is not determined in step S225 that the current location matches the feature intersection, the control unit 20 regards the current travel route of the vehicle as not a learned route, and terminates the vehicle control process (the vehicle control is also terminated). The same shall apply hereinafter.

  On the other hand, if it is not determined in step S220 that the vehicle has traveled the distance to the characteristic intersection, or if it is determined in step S225 that the current location coincides with the characteristic intersection, the control unit 20 determines whether the vehicle control unit 21d It is determined by the process whether or not the vehicle has traveled a distance to the end point (step S240). That is, the control unit 20 refers to the learning information 30b and acquires the distance from the start point to the end point. And the control part 20 acquires the travel distance from a start point based on the output signal which the vehicle speed sensor 42 output from the passage of a start point to the present, The said distance and the distance from a start point to an end point correspond. If it is determined that the vehicle has traveled the distance to the end point. Of course, the control unit 20 may determine whether or not the current location has passed the end point based on the map information 30a.

  If it is determined in step S240 that the vehicle has traveled the distance to the end point, the control unit 20 regards that the vehicle has traveled through a section in which vehicle control is to be performed, and ends the vehicle control process. On the other hand, when it is not determined in step S240 that the vehicle has traveled the distance to the end point, the control unit 20 repeats the processes after step S200.

  In step S215, when it is determined that the vehicle has made a turn at the intersection, the control unit 20 determines whether the vehicle has traveled a distance to the characteristic intersection by the process of the vehicle control unit 21d (step S230). The determination process in step S230 is the same as that in step S220. However, if it is not determined in step S230 that the vehicle has traveled the distance to the characteristic intersection, the control unit 20 ends the vehicle control process.

  When it is determined in step S230 that the vehicle has traveled the distance to the characteristic intersection, the control unit 20 determines whether or not the current location matches the characteristic intersection by the process of the vehicle control unit 21d (step S235). . The determination process in step S235 is the same as that in step S225. However, if it is not determined in step S235 that the current location matches the feature intersection, the control unit 20 ends the vehicle control process. On the other hand, if it is determined in step S235 that the current location matches the feature intersection, the control unit 20 executes step S240.

Next, the above vehicle control processing will be described with reference to FIG. Here, the vehicle C ₁ shown in FIG. 4 travels on the links L ₄ , L ₃ , L ₂ and L ₁ in the past and stops at the point Ps ₁ , and learning information 30b is recorded based on the travel history of this travel. Assuming that In this case, the start point I ₃ , the end point Ps ₁ , the distance D ₃ , D ₂ + D _3, and the links L ₂ and L ₁ are recorded on the recording medium 30 as learning information 30b. When the vehicle C ₁ travels on the links L ₄ , L ₃ , L ₂ , and L ₁ in this state, vehicle control is not started until the vehicle C ₁ passes through I ₃ . Vehicle C ₁ is determined to have passed the start point in step S205 passes through the I _3, the vehicle control is started in step S210. Before the vehicle C ₁ passes through the intersection I ₂ , it is not determined that the vehicle has made a turn at the intersection in step S215, and it is not determined that the vehicle has traveled the distance (D ₃ ) to the characteristic intersection in step S220. Since it is not determined that the vehicle has traveled the distance (D ₁ + D ₂ + D ₃ ) to the end point, the processes after step S200 are repeated.

On the other hand, when the vehicle C ₁ passes through the intersection I ₂ and reaches the link L ₂ , it is determined in step S215 that the vehicle C ₁ has made a turn, and in step S230, it is determined that the vehicle has traveled the distance (D ₃ ) to the characteristic intersection. . In addition, since the vehicle C ₁ is traveling on the link L ₂ , it is determined in step S235 that the current location matches the feature intersection. In step S240, since it is not determined that the vehicle has traveled the distance (D ₁ + D ₂ + D ₃ ) to the end point, the processes in and after step S200 are repeated. If the vehicle C ₁ passes through the intersection I ₂ and reaches the link L ₅ , it is determined in step S230 that the vehicle has traveled the distance (D ₃ ) to the feature intersection, and in step S235, the current location matches the feature intersection. It is not judged that it did. Accordingly, the vehicle control process is terminated. In other words, vehicle control is performed as long as the vehicle travels on the same route as the learning information 30b is learned, but the vehicle control ends when the vehicle travels on a route different from the route on which the learning information 30b is learned. .

When the vehicle C ₁ passes through the intersection I ₂ and reaches the link L _2, and after step S200 is repeated, when the vehicle C ₁ passes through the intersection I ₁ and reaches the link L ₁ , in step S215 In step S230, it is determined that the vehicle has traveled the distance (D ₂ + D ₃ ) to the characteristic intersection. In step S235, it is determined that the current location matches the feature intersection. In step S240, since it is not determined that the vehicle has traveled the distance (D ₁ + D ₂ + D ₃ ) to the end point, the processes in and after step S200 are repeated. In the process in which the vehicle C ₁ is traveling on the link L ₁ , it is not determined that the vehicle has made a turn at the intersection in step S215, and it is not determined that the vehicle has traveled the distance to the characteristic intersection in step S220. When the vehicle C ₁ passes through the end point Ps ₁ , it is determined in step S240 that the vehicle C ₁ has traveled the distance (D ₁ + D ₂ + D ₃ ) to the end point, and the vehicle control process is ended. Thus, vehicle control is performed based on the learning information 30b.

Next, the vehicle C ₂ shown in FIG. 4 travels on the links L ₇ , L ₃ , L ₅ in the past and stops at the point Ps ₂ , and the learning information 30b is recorded based on the travel history of this travel. Is assumed. In this case, the start point Pc ₂ and the end point Ps ₂ are recorded on the recording medium 30 as the learning information 30b. When the vehicle C ₂ travels on the links L ₇ , L ₃ , and L ₅ in this state, when the vehicle C ₂ passes the start point Pc ₂ , it is determined in step S205 that the start point has been passed, and in step S210, the vehicle control is performed. Be started. Then, before the vehicle C ₂ passes through the intersection I ₃ , it is not determined that the vehicle has turned the intersection in step S215. In step S220, it is not determined that the vehicle has traveled the distance to the feature intersection (because the distance to the feature intersection is not defined in the learning information 30b. The same applies hereinafter), and the distance to the end point in step S240 (D ₅ Since it is not determined that the vehicle has traveled + D ₃ + D ₆ ), the processes after step S200 are repeated.

On the other hand, even if the vehicle C ₂ passes through the intersection I ₃ and reaches the link L ₃ , it is not determined that the vehicle has made a turn at the intersection in step S215, and it is not determined that the vehicle has traveled the distance to the characteristic intersection in step S220. Since it is not determined that the vehicle has traveled the distance to the end point, the processes after step S200 are repeated. Similar processing is performed even when the vehicle C ₂ passes through the intersection I ₂ and reaches the link L _5, and the processing after step S200 is repeated. Even in the process in which the vehicle C ₂ is traveling on the link L ₅ , it is not initially determined that the vehicle has made a turn at the step S215 intersection, and it is not determined that the vehicle has traveled the distance to the characteristic intersection in step S220. Therefore, by the vehicle C ₂ passes the end point Ps ₂ Eventually, it is determined that the vehicle has traveled the distance to the end point _{(D 5 + D 3 + D} 6) in step S240, the vehicle control processing is ended. Thus, vehicle control is performed based on the learning information 30b even when the route is not learned.

In the case where the vehicle C ₂ has traveled the links L ₇ , L ₃ , L ₅ in the past and stopped at the point Ps ₂ , and the learning information 30b is recorded based on the travel history of this travel, the vehicle C ₂ When the vehicle travels on the links L ₇ , L ₃ , and L ₂ , it is determined that the vehicle C ₂ passes the intersection I ₂ and reaches the link L ₂ , thereby turning the intersection at step 215. In this case, it is not determined in step S230 that the vehicle has traveled the distance to the characteristic intersection, and the vehicle control process is terminated. Therefore, the vehicle control is performed as long as the vehicle travels on the same route as the learning information 30b is learned, but the vehicle control ends when the vehicle travels on a route different from the learning information 30b. .

(4) Other embodiments:
The above embodiment is an example for carrying out the present invention, and as long as the route is learned as a vehicle control section when not traveling on a continuous road section defined as a road before and after an intersection, Various other embodiments can be employed. For example, the navigation system 10 may be fixedly mounted on the vehicle, or may be a mode in which the portable navigation system 10 is brought into the vehicle and used.

  Furthermore, the information for specifying the section where the vehicle control should be executed is not limited to the above-mentioned information, and the section is defined by the start point of the section where the vehicle control should be executed and the distance between the sections. May be. Even this configuration is substantially equivalent to a configuration in which the end point is learned because the end point of the section where the vehicle control is to be executed is specified by the start point and the distance. Various configurations can be adopted as the configuration for learning the feature intersection, and the feature intersection is identified by the distance from the end point of the section where the vehicle control should be performed, and the feature intersection is identified by the coordinates of the feature intersection It may be configured to do so.

  Furthermore, the vehicle control only needs to be executed in order to provide support for improving the past operation of the vehicle, and a section where there should be room for improvement in the past operation of the vehicle is determined based on the travel history of the vehicle. Obviously, any vehicle control that can be executed in the section may be used. Therefore, the vehicle control is not limited to the control at the time of deceleration, and various types of control can be assumed. For example, acceleration control or the like may be used. In this case, what is necessary is just to set it as the structure which acquires the acceleration end point of a vehicle as an end point of the area which should perform vehicle control.

  Furthermore, as a method for defining continuous road sections as roads, various methods can be employed. That is, it is possible to define a road section that passes when traveling along a road as a road. For example, a road having a road by a continuous road section having the same road number, a continuous road section having the same attribute, or the like. An interval may be defined.

  DESCRIPTION OF SYMBOLS 10 ... Navigation system, 20 ... Control part, 21 ... Vehicle control program, 21a ... Vehicle control end point acquisition part, 21b ... Feature intersection acquisition part, 21c ... Vehicle control area learning part, 21d ... Vehicle control part, 30 ... Recording medium , 30a ... map information, 30b ... learning information, 41 ... GPS receiver, 42 ... vehicle speed sensor, 43 ... gyro sensor, 44 ... user I / F unit, 45 ... friction braking unit, 47 ... generator, 47a ... control unit 48a motor 48a controller 49a battery 49b internal combustion engine 49c gear mechanism 49d output shaft

Claims (6)

Vehicle control end point acquisition means for acquiring an end point of a section in which vehicle control is to be performed based on the travel history of the vehicle;
In the road on which the vehicle traveled before reaching the end point, the intersection is acquired as a characteristic intersection when the vehicle does not travel on a continuous road section defined as a road before and after the intersection. An intersection acquisition means;
When the vehicle passes through the characteristic intersection, the route including the characteristic intersection is learned as a section in which the vehicle control is to be executed ,
A continuous road section defined as a road before and after an intersection existing on a route between a start point and a finish point that are a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle from the end point Vehicle control section learning means for learning the end point and the start point as a section in which the vehicle control is to be executed, without learning the route between the start point and the end point .
A vehicle control section learning system comprising: The vehicle control section learning means
Learning a route indicated by the characteristic intersection and a road section on which the vehicle has traveled after passing through the characteristic intersection as a section on which the vehicle control is to be executed;
The vehicle control section learning system according to claim 1.   The vehicle control section learning means
    When a predetermined number of the characteristic intersections are included in a section within the predetermined distance, a route including the predetermined number of the characteristic intersections is learned as a section on which the vehicle control is to be executed,
    When the section within the predetermined distance includes more than the predetermined number of characteristic intersections, a route in a section shorter than the predetermined distance including the predetermined number of characteristic intersections is determined as the section on which the vehicle control is to be executed. learn,
The vehicle control section learning system according to claim 1 or 2.   The vehicle control end point acquisition means acquires a deceleration end point of the vehicle as the end point.
The vehicle control section learning system according to any one of claims 1 to 3.   A step in which a vehicle control end point acquisition unit acquires an end point of a section in which vehicle control is to be performed based on a travel history of the vehicle;
  When the characteristic intersection acquisition unit does not travel on a continuous road section defined as a road before and after the intersection on the road on which the vehicle has traveled before reaching the end point, the intersection is acquired. Obtaining as a feature intersection,
  When the vehicle passes through the characteristic intersection, the vehicle control section learning unit learns a route including the characteristic intersection as a section where the vehicle control is to be executed,
  A continuous road section defined as a road before and after an intersection existing on a route between a start point and a finish point that are a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle from the end point Learning the route between the start point and the end point without learning the route between the start point and the end point as a section in which the vehicle control is to be executed,
A vehicle control section learning method including:   Computer
  A vehicle control end point acquisition unit that acquires an end point of a section in which vehicle control should be executed based on a vehicle travel history;
  In the road on which the vehicle traveled before reaching the end point, the intersection is acquired as a characteristic intersection when the vehicle does not travel on a continuous road section defined as a road before and after the intersection. Intersection acquisition unit,
  When the vehicle passes through the characteristic intersection, the route including the characteristic intersection is learned as a section in which the vehicle control is to be executed,
  A continuous road section defined as a road before and after an intersection existing on a route between a start point and a finish point that are a predetermined distance from the end point to the opposite side of the traveling direction of the vehicle from the end point A vehicle control section learning unit that learns the end point and the start point as a section in which the vehicle control is to be executed, without learning the route between the start point and the end point.
Vehicle control section learning program to function as.

Images