JP2013083576A - Vehicle location recognition system, vehicle location recognition program, and vehicle location recognition method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a technique which can set the recognition rate of a feature for determining a recognition target in an image recognition process in consideration of speed of a vehicle when imaging the feature.SOLUTION: A vehicle location recognition system comprises: a road information acquisition unit 23b for acquiring road information RD; an estimated speed transition determination unit 27 for determining estimated speed transition indicating estimation of speed transition of a vehicle in a feature disposition section which is a road section in which features are disposed, on the basis of at least the road information RD before an estimated vehicle position EP enters the feature disposition section; a recognition rate determination unit 26 for determining a recognition rate with which it is determined whether image recognition of the feature included in the feature disposition section is successful, on the basis of feature information FT and the estimated speed transition; and a recognition target determination unit 25 for determining the recognition target feature in image recognition processing on the basis of the recognition rate.

Description

  The present invention relates to a host vehicle position recognition system, a host vehicle position recognition program, and a host vehicle position recognition method that recognize a host vehicle position using an image captured by a camera provided in the host vehicle.

  As a conventional technique of the vehicle position recognition system as described above, there is a technique described in, for example, Japanese Unexamined Patent Application Publication No. 2009-58429 (Patent Document 1). Patent Document 1 describes a technique for performing image recognition processing of a feature included in a photographed image by a camera and correcting the estimated vehicle position based on the result of image recognition and position information of the feature. . At this time, the feature to be recognized in the image recognition process is determined based on the recognition rate, which is the rate of successful image recognition of each feature. Specifically, when a plurality of features that can be recognized are arranged at narrow intervals, a feature with a high recognition rate is selected as a feature to be recognized. Thereby, it is possible to increase the recognition rate of the feature per fixed distance, reducing the calculation load concerning image recognition processing.

  By the way, since the camera for photographing the feature is provided in the vehicle, depending on the image recognition technique, the performance of the image recognition device, etc., the recognition rate for the same type of feature may be the own vehicle at the time of photographing the feature. It is assumed that it changes according to the speed of the. However, with regard to the speed of the host vehicle, Patent Document 1 has not been particularly recognized.

JP 2009-58429 A

  Therefore, it is desired to realize a technique capable of determining the recognition rate of a feature for determining a recognition target in the image recognition processing in consideration of the speed of the host vehicle at the time of photographing the feature.

  According to the present invention, an estimated own vehicle position acquisition unit that acquires an estimated own vehicle position, which is an estimated position of the own vehicle, and a captured image captured by a camera provided in the own vehicle are acquired and included in the captured image An image processing unit that performs image recognition processing of a feature to be obtained, a feature information acquisition unit that acquires feature information including position information and type information of the feature, a recognition result of the feature by the image recognition processing, and the feature A feature configuration of a host vehicle position recognition system including an estimated host vehicle position correction unit that corrects the estimated host vehicle position based on position information of features included in the object information is road information acquisition that acquires road information. And the speed of the own vehicle in the feature placement section based on at least the road information before the estimated vehicle position enters the feature placement section that is a road section in which the feature is placed. Determine the predicted speed transition that indicates the transition prediction A measurement rate transition determination unit; and a recognition rate determination unit that determines a recognition rate that is a rate of successful image recognition of the feature included in the feature arrangement section based on the feature information and the predicted speed transition; A recognition target determining unit that determines a feature to be recognized in the image recognition processing based on the recognition rate.

According to the above feature configuration, the feature recognition rate is determined based on the speed of the vehicle at the time of shooting the feature (including acceleration which is a time derivative of the speed) in addition to the feature information. Can do. At this time, in the present invention, focusing on the fact that the speed transition of the host vehicle can be predicted by road information, such as when the vehicle is predicted to stop at an intersection with a temporary stop obligation, the recognition rate is used. A configuration is adopted in which the speed of the vehicle to be obtained is obtained from a predicted speed transition determined based on road information before the estimated vehicle position enters the feature arrangement section. Thereby, it is possible to estimate in advance a recognition rate close to the actual recognition rate according to the speed of the host vehicle at the time of photographing the feature.
As described above, according to the above-described feature configuration, the recognition rate of the feature for determining the recognition target in the image recognition process is appropriately set in advance in consideration of the speed of the host vehicle at the time of shooting the feature. It is possible to decide.

  Here, the road information includes information for specifying the presence / absence of an intersection and the type of the intersection, and the predicted speed transition determination unit passes through the intersection when the feature arrangement section includes an intersection. It is preferable that the predicted speed transition is determined so that the speed of the host vehicle at the time of the vehicle becomes a type-dependent speed corresponding to the type of the intersection.

  Specifically, for example, a traveling road on which the host vehicle travels before entering the intersection included in the feature arrangement section is a priority road with respect to another road that intersects the traveling road at the intersection. The type-dependent speed may be determined to be higher than the type-dependent speed when the other road is a priority road for the traveling road.

  According to such a configuration, it is possible to determine the predicted speed transition in consideration of the behavior of the host vehicle that is expected when passing the intersection, and to improve the reliability of the recognition rate determined based on the predicted speed transition. Can be increased.

  In the vehicle position recognition system of each configuration described above, a correction coefficient related to the recognition rate due to the speed of the vehicle when performing image recognition processing of a feature by the image processing unit based on the predicted speed transition. A correction coefficient determination unit that determines a certain speed factor correction coefficient, wherein the recognition rate determination unit calculates a value obtained by correcting a reference recognition rate that is predetermined according to a feature type by the speed factor correction coefficient; A configuration in which the recognition rate is determined is preferable.

  According to this configuration, since the recognition rate can be derived with a simple configuration, when the recognition rate is calculated and determined as necessary, the load related to the calculation can be reduced. Further, when the recognition rate is calculated in advance and stored in the storage device, the cost for the maintenance can be reduced.

  As described above, in the configuration in which the value obtained by correcting the reference recognition rate with the speed factor correction coefficient is determined as the recognition rate, the correction coefficient determination unit increases the acceleration of the host vehicle indicated by the predicted speed transition. Accordingly, it is preferable that the speed factor correction coefficient is determined so as to lower the recognition rate.

  According to this configuration, it is possible to appropriately determine the recognition rate when using a speed detection system that tends to decrease the speed detection accuracy as the acceleration of the host vehicle increases.

  Further, as described above, in the configuration in which a value obtained by correcting the reference recognition rate with the speed factor correction coefficient is determined as the recognition rate, the correction coefficient determination unit is configured to determine whether the speed of the host vehicle indicated by the predicted speed transition is It is preferable that the speed factor correction coefficient is determined so as to lower the recognition rate as it decreases.

  According to this configuration, for example, when using a sensor such as a vehicle speed pulse sensor that tends to decrease the speed detection accuracy as the speed of the host vehicle decreases, the recognition rate can be appropriately determined.

  Further, as described above, in the configuration in which the value obtained by correcting the reference recognition rate with the speed factor correction coefficient is determined as the recognition rate, the recognition rate determination unit is a feature provided on a road surface, For features having the same shape regardless of the position in the road extending direction throughout the recognition target portion in the image recognition processing, the recognition rate is determined by reducing the correction amount by the speed factor correction coefficient. A configuration is preferable.

  According to this configuration, for a feature whose positional relationship between parts at different positions along the road extending direction in the same feature does not greatly affect the success or failure of the image recognition of the feature, the recognition in accordance with the current situation The rate can be determined.

  In the vehicle position recognition system of each configuration described above, for a group of features that are arranged separately in the extending direction of the road and the interval along the extending direction of adjacent features is a predetermined distance or less. It is preferable that the feature arrangement section is set so as to include all the features constituting the group of features.

  According to this configuration, since the recognition rate of each feature constituting a group of features can be determined based on the predicted speed transition determined by the predicted speed transition determination unit, the configuration for determining the recognition rate is simplified. Can be achieved.

  The technical features of the vehicle position recognition system according to the present invention having the above-described configurations can be applied to a vehicle position recognition program and a vehicle position recognition method, and the present invention also includes such a program and method. Can be subject to rights.

  In this case, the feature configuration of the host vehicle position recognition program includes an estimated host vehicle position acquisition function for acquiring an estimated host vehicle position that is an estimated position of the host vehicle, and a photographed image shot by a camera provided in the host vehicle. An image processing function for performing image recognition processing of a feature included in the captured image, a feature information acquisition function for acquiring feature information including position information and type information of the feature, and the image recognition processing A vehicle position recognition program for realizing an estimated vehicle position correction function that corrects the estimated vehicle position based on the feature recognition result by the feature information and the position information of the feature included in the feature information. The road information acquisition function for acquiring road information and the feature based on at least the road information before the estimated vehicle position enters the feature placement section which is a road section where the feature is placed. Within placement section Based on the predicted speed transition determination function for determining the predicted speed transition indicating the prediction of the speed transition of the host vehicle, and the feature information included in the feature arrangement section based on the feature information and the predicted speed transition A recognition rate determination function for determining a recognition rate, which is a rate of successful recognition, and a recognition target determination function for determining a feature to be recognized in the image recognition process based on the recognition rate. is there.

  In addition, the characteristic configuration of the vehicle position recognition method includes an estimated vehicle position acquisition step for acquiring an estimated vehicle position that is an estimated position of the vehicle, and a captured image captured by a camera provided in the vehicle. An image processing step for performing image recognition processing of the feature included in the photographed image, a feature information acquisition step for acquiring feature information including position information and type information of the feature, and a ground by the image recognition processing. A vehicle position recognition method comprising: an estimated vehicle position correction step for correcting the estimated vehicle position based on a recognition result of an object and position information of a feature included in the feature information, wherein the road information A road information acquisition step for acquiring the vehicle information, and before the estimated vehicle position enters the feature placement section that is a road section in which the feature is placed, at least within the feature placement section based on the road information The vehicle A rate of success in image recognition of a feature included in the feature placement section based on a predicted speed transition determination step for determining a predicted speed transition indicating prediction of a speed transition, and the feature information and the predicted speed transition. A recognition rate determining step for determining the recognition rate, and a recognition target determining step for determining a feature to be recognized in the image recognition processing based on the recognition rate.

  Naturally, these vehicle position recognition programs and vehicle position recognition methods can also obtain the effects of the vehicle position recognition system described above, and some additional examples given as examples of suitable configurations thereof. It is possible to incorporate technology.

It is a mimetic diagram showing a schematic structure of a navigation device concerning an embodiment of the present invention. It is a schematic diagram which shows an example of the running condition of the vehicle which concerns on embodiment of this invention, (a) shows the road network which a node and a link represent, (b) shows the road corresponding to (a), and the said road The provided features are shown, and (c) shows the predicted speed transition in the feature arrangement section shown in (b). It is a schematic diagram which shows another example of the driving | running | working condition of the vehicle which concerns on embodiment of this invention, (a) shows the road network which a node and a link represent, (b) shows the road corresponding to (a), and the said The features provided on the road are shown, and (c) shows the predicted speed transition in the feature arrangement section shown in (b). It is a figure which shows an example of the speed predicted value setting rule which concerns on embodiment of this invention. It is a figure which shows an example of the acceleration estimated value setting rule which concerns on embodiment of this invention. It is a conceptual diagram which shows an example of the information memorize | stored in the reference | standard recognition rate memory | storage part which concerns on embodiment of this invention. It is a figure which shows an example of the speed deduction point setting rule which concerns on embodiment of this invention. It is a figure which shows an example of the acceleration deduction point setting rule which concerns on embodiment of this invention. It is a flowchart which shows an example of the whole procedure of the own vehicle position recognition process which concerns on embodiment of this invention. It is a flowchart which shows an example of the procedure of the estimation own vehicle position correction process which concerns on embodiment of this invention.

  Embodiments of the present invention will be described with reference to the drawings. Here, the case where the vehicle position recognition system according to the present invention is applied to an in-vehicle navigation device will be described as an example. As shown in FIG. 1, the navigation device 1 according to the present embodiment includes a host vehicle position recognition unit 2, and a host vehicle position recognition system is built around the host vehicle position recognition unit 2.

  In the following description, “traveling direction front D1” and “traveling direction rear D2” (see FIGS. 2 and 3) indicate that the own vehicle 3 (hereinafter sometimes referred to as “vehicle 3”) extends in the road extending direction L. It is defined with reference to the state of traveling forward along (the road length direction). Therefore, “traveling direction front D1” indicates a direction toward the front of the vehicle 3, and “traveling direction rear D2” indicates a direction toward the rear of the vehicle 3.

1. Schematic Configuration of Navigation Device As shown in FIG. 1, the navigation device 1 includes a host vehicle position recognition unit 2, a control unit 80, an estimated host vehicle position determination unit 94, a map information storage unit 10, and a reference recognition rate storage. Part 11. The own vehicle position recognition unit 2 and the control unit 80 include a plurality of functional units.

  Each functional unit included in the navigation device 1 is a hardware or software (functional or functional unit for performing various processes on input data using a common arithmetic processing unit such as a CPU or the like as a core member. Program) or both. These functional units are configured to be able to exchange information with each other via a communication line, and are configured to be able to extract data from the storage units 10 and 11. Here, when each functional unit is configured by software (program), the software is stored in a storage device that can be referred to by the arithmetic processing device.

  The map information storage unit 10 and the reference recognition rate storage unit 11 are databases configured by a storage device. This storage device includes, as a hardware configuration, a recording medium capable of storing and rewriting information, such as a hard disk drive and a flash memory. Each of the storage units 10 and 11 may have independent hardware, or may be provided in common hardware.

  The control unit 80 includes a route setting unit 81, a route search unit 82, and a route guide unit 83. The route setting unit 81 is a functional unit that sets, for example, a departure place such as the vehicle position, an input destination, a passing point, and traveling conditions (whether or not an expressway is used). The route search unit 82 is a functional unit that performs calculation processing (route search processing) for searching for a guide route from the departure point to the destination based on the conditions set by the route setting unit 81. The route guidance unit 83 operates according to the guidance from the display screen of a monitor (not shown) or voice guidance from a speaker (not shown) according to the route from the departure point to the destination searched by the route search unit 82. It is a functional part which performs the calculation process (route guidance process) for performing suitable route guidance with respect to a person.

  The map information storage unit 10 stores (stores) map information (road map data) that is referred to when the above route search processing or route guidance processing is executed, or when map display processing on a monitor is executed. Has been. Specifically, the map information includes road information RD representing a road network and feature information FT associated with the road information RD. That is, the map information storage unit 10 is integrally formed with a road information storage unit 10b that stores road information RD and a feature information storage unit 10a that stores feature information FT.

  As conceptually shown in FIG. 2 (a), the road information RD is obtained by connecting information of a large number of nodes N having position information on a map expressed by latitude and longitude and two nodes N to form roads. Information of a number of links K constituting a corresponding route. The node N is set, for example, at a road intersection 70 (see FIGS. 2 and 3), a turning point, a branch point, or the like. The intersection 70 includes a junction and a branch point. The road information RD includes node information related to the attribute of the node N and link information related to the attribute of the link K. For the node N set at the intersection 70, the node type information of the node N includes the intersection type information and the traffic signal presence / absence information. Thus, the road information RD includes information for specifying the presence / absence of the intersection 70 and the type of the intersection 70.

  In this example, the intersection type includes a priority type based on the priority relationship of the intersection 70. By referring to the information of the priority type, traffic between roads connected to the intersection 70 without a traffic signal is obtained. It is possible to obtain information on the priority order for the. The link information includes a shape interpolation point, a road width, a road type (highway, toll road, national road, prefectural road, etc.) arranged between two nodes N (that is, on the link K) and representing the detailed shape of the link K. Type), speed limit (maximum speed), link length, and the like.

  The feature information FT is information on various features 50 provided on or around the road. The feature 50 in which the feature information FT is stored in the map information storage unit 10 includes a road marking, and in this embodiment, a three-dimensional structure (for example, a sign) whose relative position with respect to the road is fixed. , Traffic lights, etc.). Here, the road marking is a symbol or character for regulation or instruction provided on the road surface by painting or the like. For example, as shown in FIG. 6, the road marking includes a sign indicating a traveling direction (traveling direction sign 51), a sign indicating a stop line (first stop line sign 52 provided on the own lane, an adjacent lane (oncoming lane) Etc.), a second stop line sign 53), a sign indicating the shape of the intersection 70 (a cross-shaped first intersection shape sign 54, a T-shaped second intersection shape sign 55), and a pedestrian crossing. The indication to show (pedestrian crossing indication 56) etc. are included.

  The feature information FT includes position information and type information of the feature 50. Here, the position information is information representing the position (latitude and longitude) of the feature 50 on the map, and in this example, the position information is information based on the node N and the shape interpolation point. This position information is basically information representing the position of the representative point of the feature 50 (for example, the center in the length direction and the width direction of the feature 50). The type information is information indicating the type of the feature 50, and in this example, the feature 50 having the same shape including the shape, color, and arrangement in the road width direction is defined as one type. Specifically, as shown in FIG. 6, the features 50 having different shapes are classified into different types, and the first stop line marking 52 and the second stop line marking 53 have the same shape. Even if it exists, the thing from which the said form differs (specifically what is arrange | positioned in the lane different from the advancing direction lane) is classified into the different classification.

  The map information having the road information RD and the feature information FT as described above is made into a database and stored in a form according to the specifications of the map information storage unit 10. For example, the map information may be divided into a plurality of layers associated with each other and stored in the map information storage unit 10. Specifically, it includes at least a road network layer indicating connection information between roads, a road shape layer indicating road shapes, and a feature layer indicating information on various features 50 provided on or around the road. The map information can be divided into three layers. In this case, the road information RD is composed of information stored in the road network layer and the road shape layer, and the feature information FT is composed of information stored in the feature layer.

  The estimated own vehicle position determination unit 94 determines the estimated own vehicle position EP that is the estimated position of the own vehicle 3 based on the outputs of the GPS measurement unit 91, the distance sensor 92, and the direction sensor 93 provided in the vehicle 3. Part. Specifically, the estimated host vehicle position determination unit 94 derives the current position (latitude and longitude) of the host vehicle 3 by analyzing a GPS (Global Positioning System) signal acquired from the GPS measurement unit 91 and determines the GPS. Get position data. Further, the estimated vehicle position determination unit 94 derives a dead reckoning position based on the moving distance information and the azimuth information acquired from the distance sensor 92 and the azimuth sensor 93 to obtain dead reckoning position data. Then, the estimated own vehicle position determination unit 94 performs a calculation for deriving the estimated own vehicle position EP that is the estimated position of the own vehicle 3 from the GPS position data and the dead reckoning position data by a known method, and estimates the own vehicle position EP And the information of the determined estimated own vehicle position EP is output to the own vehicle position recognition unit 2.

  The estimated host vehicle position EP determined as described above is information including measurement errors, and may be off the road depending on the case. In this embodiment, in such a case, based on the road map data stored in the map information storage unit 10, the estimated vehicle position determining unit 94 adjusts the estimated vehicle position EP on the road indicated on the road map. (Map matching process) is performed, and information on the estimated vehicle position EP after the correction is output to the vehicle position recognition unit 2.

  The own vehicle position recognition unit 2 includes functional units 22 to 27 for correcting the estimated own vehicle position EP acquired from the estimated own vehicle position determination unit 94. The configuration of the host vehicle position recognition unit 2 will be described in detail in the section “2. Configuration of the host vehicle position recognition unit” later. The host vehicle position recognition unit 2 is a camera 90 ( By correcting the estimated vehicle position EP using the captured image PG according to FIGS. 1 to 3), the vehicle position can be determined with higher accuracy than the estimated vehicle position EP determined by the estimated vehicle position determining unit 94. It is possible to recognize.

  In the present embodiment, as shown in FIGS. 2B and 3B, the camera 90 is a back camera (rear camera) that captures the rear of the vehicle 3. The road surface behind (in the advancing direction rear D2 side) is included. Specifically, in this embodiment, the camera 90 is installed in the vehicle 3 with a depression angle (for example, about 30 degrees) on the optical axis so that the road surface is mainly photographed, and the estimated vehicle position EP is The feature 50 as a target in the image recognition for correction is a road marking provided on the road surface.

2. Next, the configuration of the vehicle position recognition unit 2, which is a main part of the present invention, will be described. As shown in FIG. 1, the vehicle position recognition unit 2 includes an estimated vehicle position acquisition unit 21, an image processing unit 22, a map information acquisition unit 23, and an estimated vehicle position correction unit 24. A determination unit 27, a recognition rate determination unit 26, and a recognition target determination unit 25 are provided.

2-1. Configuration of Estimated Own Vehicle Position Acquisition Unit The estimated own vehicle position acquisition unit 21 is a functional unit that acquires an estimated own vehicle position EP that is an estimated position of the own vehicle 3. In the present embodiment, the estimated vehicle position EP determined by the estimated vehicle position determination unit 94 is input to the estimated vehicle position acquisition unit 21 so that the estimated vehicle position acquisition unit 21 acquires the estimated vehicle position EP. To do.

2-2. Configuration of Image Processing Unit The image processing unit 22 is a functional unit that acquires a captured image PG captured by the camera 90 provided in the host vehicle 3 and performs image recognition processing of the feature 50 included in the captured image PG. is there. That is, the image processing unit 22 has an image acquisition function and an image recognition function. The image processing by the image processing unit 22 is executed when the estimated vehicle position EP is located within an image recognition section set along the road extending direction L as a range in which image recognition processing is performed. In the present embodiment, this image recognition section basically matches a feature arrangement section A described later.

  The feature 50 to be recognized in the image recognition processing by the image processing unit 22 can specify at least the position in the road extending direction L when the image is recognized (that is, when the image recognition is successful). The feature 50 is used. Therefore, the feature 50 provided continuously or intermittently along the road extending direction L such as a lane marking (not shown) basically does not become the feature 50 to be recognized. FIG. 6 shows an example of a feature 50 that can be a recognition target. The feature 50 shown in FIG. 6 is composed of a single road marking, but the feature 50 composed of a plurality of (for example, two) road markings is to be recognized in the image recognition processing. Is also possible. The feature 50 to be recognized in the image recognition processing is determined by the recognition target determination unit 25 described later.

  Captured images PG (for example, captured images PG of 15 to 30 frames per second) at predetermined time intervals by the camera 90 are temporarily stored in a buffer memory (not shown), and the image processing unit 22 captures captured images PG from the buffer memory. To get. In the present embodiment, the image processing unit 22 performs image recognition of the feature 50 by combining a plurality of frames in which the feature 50 to be recognized is captured. At this time, the image processing unit 22 combines frames based on the vehicle speed information acquired from the distance sensor 92 according to the vehicle speed (speed of the host vehicle 3) at the time of photographing the feature 50 to be recognized. In this embodiment, the image recognition processing by the image processing unit 22 is performed by pattern matching (template matching) based on contour information acquired from the captured image PG and shape information of the recognition target feature 50 included in the feature information FT. ) Method is used.

  As described above, the vehicle speed used in the image recognition processing by the image processing unit 22 is based on the detection result of the distance sensor 92. In this embodiment, the distance sensor 92 is a sensor that uses a vehicle speed pulse sensor that outputs a pulse signal each time a drive shaft, a wheel, or the like rotates by a certain amount. Therefore, the detection accuracy (resolution) of the vehicle speed by the distance sensor 92 decreases as the vehicle speed decreases as a whole. Further, there is a difference between the vehicle speed acquired by the image processing unit 22 and the actual vehicle speed at the time of the acquisition due to a control delay or a detection delay. The acceleration generally increases as the positive direction side (acceleration side) or the negative direction side (deceleration side) increases (that is, the absolute value increases).

  Therefore, in this embodiment, the maximum value of the error that can be included in the vehicle speed acquired by the image processing unit 22 tends to increase as the vehicle speed decreases and increases as the acceleration increases. When the vehicle speed includes an error, the shape of the feature 50 obtained by combining a plurality of frames is a shape obtained by lengthening or shortening the shape of the actual feature 50 along the road extending direction L. Become. For this reason, the recognition rate R, which is the rate of successful image recognition of the feature 50, tends to decrease as the error included in the vehicle speed increases. That is, the recognition rate R of the feature 50 decreases as the vehicle speed at the time of shooting the feature 50 decreases, and the recognition rate of the feature 50 increases as the acceleration of the host vehicle 3 at the time of shooting the feature 50 increases. There is a tendency for R to decrease. In the present embodiment, a configuration is adopted in which the recognition rate determination unit 26 determines the recognition rate R of each feature 50 in consideration of the change tendency of the recognition rate R according to the speed and acceleration of the host vehicle 3. By doing this, it is possible to appropriately determine the recognition rate R for each feature 50. The configuration of the recognition rate determination unit 26 will be described later.

2-3. Configuration of Map Information Acquisition Unit The map information acquisition unit 23 is a functional unit that acquires the feature information FT and the road information RD. In the present embodiment, the map information acquisition unit 23 refers to the map information storage unit 10 and acquires the feature information FT and the road information RD. Specifically, the map information acquisition unit 23 includes a road information acquisition unit 23b that acquires road information RD with reference to the map information storage unit 10 (road information storage unit 10b), and the road information acquisition unit 23b acquires the information. The road information RD is used in various processes executed by the vehicle position recognition unit 2 including the predicted speed transition determination process by the predicted speed transition determination unit 27. Moreover, the map information acquisition part 23 is provided with the feature information acquisition part 23a which acquires the feature information FT with reference to the map information storage part 10 (feature information storage part 10a), and the feature information acquisition part 23a acquires. The feature information FT thus obtained includes an image recognition process performed by the image processing unit 22, an estimated vehicle position correction process performed by the estimated vehicle position correction unit 24, and a recognition rate determination process performed by the recognition rate determination unit 26. It is used in various processes executed in

2-4. Configuration of Estimated Own Vehicle Position Correction Unit The estimated own vehicle position correction unit 24 is based on the recognition result of the feature 50 by the image recognition process of the image processing unit 22 and the position information of the feature 50 included in the feature information FT. This is a functional unit that corrects the estimated vehicle position EP. That is, when the image recognition of the feature 50 by the image processing unit 22 is successful, the estimated vehicle position correction unit 24 corrects the estimated vehicle position EP using the position information of the feature 50. In this example, the estimated host vehicle position correcting unit 24 corrects the estimated host vehicle position EP at least in the road extending direction L.

  Specifically, the estimated own vehicle position correcting unit 24 determines the position of the feature 50 that has been image-recognized based on the installation form of the camera 90 on the vehicle 3 and the position of the own vehicle 3 at the time of photographing the feature 50. The positional relationship between is calculated. At this time, the position and size of the feature 50 in the captured image PG are taken into consideration. Then, the estimated vehicle position correction unit 24 calculates the feature information FT by calculation based on the calculation result of the positional relationship and the feature information FT (specifically, position information) acquired by the map information acquisition unit 23. The high-accuracy own vehicle position (feature-dependent own vehicle position) based on is acquired. It is also possible to use a predetermined fixed positional relationship without performing the calculation of the positional relationship. For example, the position of the host vehicle 3 at the time of photographing the feature 50 is set as the position of the feature 50, or a predetermined distance (fixed value) along the road extending direction L with respect to the position of the feature 50. ).

  The estimated vehicle position correction unit 24 corrects the estimated vehicle position EP by replacing the acquired feature-dependent vehicle position with the estimated vehicle position EP. Instead of simply replacing the position, it is also possible to adopt a configuration in which the position of the vehicle is adjusted in consideration of the travel distance (capable of being acquired from the distance sensor 92) from when the feature 50 is captured.

2-5. Configuration of Predicted Speed Transition Determination Unit The predicted speed transition determination unit 27 predicts the speed transition of the host vehicle 3 in the feature placement section A, which is the road section in which the feature 50 is placed, based on the road information RD. It is a function part which determines the prediction speed transition shown. The determination of the predicted speed transition is executed before the estimated vehicle position EP enters the feature arrangement section A for which the predicted speed transition is to be determined. That is, the feature arrangement section A is set to the forward direction D1 in the traveling direction with respect to the vehicle 3. The predicted speed transition determined by the predicted speed transition determination unit 27 is used for determination of a recognition rate R by a recognition rate determination unit 26 described later, and image recognition executed by the image processing unit 22 based on the recognition rate R. The feature 50 to be recognized in the process is determined by the recognition target determining unit 25. Therefore, the feature arrangement section A is set so as to include a feature 50 (a road marking as shown in FIG. 6 in this example) that can be a recognition target in the image recognition process.

  The predicted speed transition determination unit 27 determines the predicted speed transition in the feature arrangement section A based on the road information RD of the feature arrangement section A acquired by the map information acquisition unit 23. In the present embodiment, the predicted speed transition determining unit 27 determines a predicted speed transition based on a predicted speed transition setting rule including a speed predicted value setting rule and an acceleration predicted value setting rule. Here, the speed predicted value setting rule is a rule that defines a speed predicted value in each road part (section or point) on the road, and the acceleration predicted value setting rule is a connection part between road parts having different speed predicted values. This is a rule for setting a predicted acceleration value at.

  FIG. 4 is a diagram illustrating an example of a speed prediction value setting rule. In this example, the speed prediction value includes four “straight section”, “curve section”, “non-priority intersection”, and “priority intersection”. It is defined for road parts (road form). Here, the straight section is a section where the radius of curvature of the road is equal to or greater than a predetermined reference value (including infinity), and the curved section is a section where the curvature of the road is smaller than the reference value. This reference value can be a value set based on the speed limit. For example, when the vehicle 3 travels (including turning) at the speed limit, the lateral acceleration acting on the vehicle 3 is a predetermined value (for example, 0. 0). The radius of curvature at which 2 [G] or the like, where G is the acceleration of gravity), can be used as the reference value. The priority intersection is defined as another road 61 (see FIGS. 2 and 3) in which the traveling road 60 (see FIGS. 2 and 3) on which the vehicle 3 travels before entering the intersection intersects the traveling road 60 at the intersection. The non-priority intersection is an intersection where the other road 61 is a priority road with respect to the traveling road 60.

  The first predicted speed Va, which is the speed predicted value corresponding to the straight section, can be a value based on the speed limit, for example, the same value as the speed limit, a value obtained by multiplying the speed limit by a predetermined value or addition Can be a value. Note that the predetermined value (multiplication value or addition value) in this case can be a learning value. The information regarding the speed limit may be configured to be acquired based on road information RD stored in the map information storage unit 10 or based on information obtained by road-to-vehicle communication with a communication device installed on the road side. It can also be set as the structure acquired. The second predicted speed Vb, which is the predicted speed value corresponding to the curve section, can be a recommended vehicle speed based on the curvature radius of the road. The recommended vehicle speed can be set to a speed at which the lateral acceleration acting on the vehicle 3 becomes a predetermined value (for example, 0.2 [G], for example). In this case, the second predicted speed Vb as the radius of curvature decreases. Is set to a low value.

  The third predicted speed Vc, which is the speed predicted value corresponding to the non-priority intersection, is basically set to a value smaller than the fourth predicted speed Vd, which is the speed predicted value corresponding to the priority intersection. Specifically, the third predicted speed Vc is determined when the non-priority intersection to be set is an intersection that is obliged to temporarily stop (for example, the intersection 70 provided with the first stop line marking 52 shown in FIG. 2). , Set to zero. For example, the fourth predicted speed Vd is set to the same value as the predicted speed value of the road section (straight section or curve section) where the priority intersection to be set is provided, or a predetermined value is subtracted from the predicted speed value. Or a value obtained by multiplying a predetermined value of 1 or less. The predetermined value (subtraction value or multiplication value) in this case can be a learning value. In the present embodiment, the third predicted speed Vc and the fourth predicted speed Vd correspond to the “type-dependent speed” in the present invention.

  Note that the speed predicted value defined in the speed predicted value setting rule may be a learning value instead of a fixed value. Moreover, it can also be set as the structure by which the speed predicted value prescribed | regulated to a speed predicted value setting rule is set individually for every individual and every road part (by section or spot). In such a case, the third predicted speed Vc for a certain non-priority intersection is set to the same value as the fourth predicted speed Vd for a certain priority intersection, or is set to a value larger than the fourth predicted speed Vd. You can also

  FIG. 5 is a diagram illustrating an example of an acceleration predicted value setting rule. In this example, the acceleration predicted value is defined for two traveling conditions of “deceleration” and “acceleration”. The first predicted acceleration αa (<0), which is an acceleration predicted value corresponding to deceleration, and the second predicted acceleration αb (> 0), which is an acceleration predicted value corresponding to acceleration, are set based on the general tendency of the driver. For example, the acceleration may be an acceleration having an absolute value of 0.1 [G], 0.2 [G], or the like. As with the speed prediction value, the acceleration prediction value may be a learning value instead of a fixed value, or may be configured to be set individually for each individual or for each road part (for each section or each point). is there.

  In order to provide the above-described configuration, when the intersection 70 is included in the feature arrangement section A, the speed of the host vehicle 3 when passing through the intersection 70 depends on the type-dependent speed according to the type of the intersection 70 ( In the example of FIG. 4, the predicted speed transition is determined so as to be the third predicted speed Vc or the fourth predicted speed Vd). The type in which the traveling road 60 on which the host vehicle 3 travels before entering the intersection 70 included in the feature arrangement section A is a priority road for another road 61 that intersects the traveling road 60 at the intersection 70. The fourth predicted speed Vd that is the dependent speed is determined to be higher than the third predicted speed Vc that is the type-dependent speed when the other road 61 is the priority road for the traveling road 60.

  A specific example of the predicted speed transition determined by the predicted speed transition determination unit 27 will be described with reference to FIGS. FIG. 2B is a diagram showing a situation in which there is an intersection 70 on the front direction D1 side of the host vehicle 3 and the intersection 70 is a non-priority intersection. FIG. 2A is a diagram illustrating FIG. The road network information corresponding to is shown. Both FIG. 2A and FIG. 2B show road information RD recognized on the vehicle 3 side based on the estimated host vehicle position EP. Therefore, although illustration is omitted, the actual position of the feature 50 with respect to the vehicle 3 (the position in the extending direction L) depends on the error included in the estimated own vehicle position EP from the position shown in FIG. The position is shifted along the extending direction L.

  In the example shown in FIG. 2B, all of the traveling direction sign 51, the first stop line sign 52, the first intersection shape sign 54, and the second stop line sign 53 on the front direction D 1 side of the vehicle 3 are included. As described above, the feature arrangement section A is set. These four features 50 (travel direction sign 51, first stop line sign 52, first intersection shape sign 54, and second stop line sign 53) are arranged separately in the extending direction L of the road, An interval S along the extending direction L between adjacent features 50 is equal to or less than a predetermined distance T. That is, in the present embodiment, the feature arrangement section A is arranged separately in the road extending direction L, and the interval S along the extending direction L of adjacent features 50 is equal to or less than the predetermined distance T. A certain group of features is set as a continuous section extending in the extending direction L so as to include all the features 50 constituting the group of features.

  Here, the predetermined distance T can be set, for example, based on the maximum value of errors that can be included in the estimated host vehicle position EP, and is, for example, within a range of 10 [m] or more and 100 [m] or less. Can be a value. The predetermined distance T can be variably set according to the vehicle speed. In addition, since the maximum value of the error that can be included in the estimated vehicle position EP can change in accordance with the travel distance from the previous correction by the estimated vehicle position correction unit 24, this change is taken into consideration from the previous correction. It is also possible to variably set the predetermined distance T according to the travel distance. Further, in the example shown in FIG. 2, the interval S between the features 50 is the interval between the central portions in the extending direction L of the features 50, but the ends on the sides close to each other in the extending direction L of the features 50. It can also be set as an interval between parts (that is, a road length of a part not occupied by the feature 50).

  Here, among the features 50 included in the feature arrangement section A, the feature 50 (travel direction indicator 51 in the example shown in FIG. 2) located closest to the rearward direction D2 is defined as the “first target feature”. The feature 50 (the second stop line marking 53 in the example shown in FIG. 2) located closest to the forward direction D1 among the features 50 included in the feature arrangement section A is referred to as a “second target feature”. To do. In the example shown in FIG. 2, the end of the feature arrangement section A on the side in the advancing direction rear D2 side is a first predetermined distance on the advancing direction rear D2 side with respect to the end on the advancing direction rear D2 side of the first target feature. The end of the feature arrangement section A on the front side D1 in the traveling direction is set to a second predetermined front side on the front direction D1 with respect to the end on the front side D1 in the traveling direction of the second target feature. It is set at a position separated by a distance. In this embodiment, the first predetermined distance and the second predetermined distance are set in this way, as described above, the feature arrangement section A coincides with the image recognition section that is a range in which the image recognition process is executed. For this reason, the first predetermined distance and the second predetermined distance are values based on the maximum value of errors that can be included in the estimated vehicle position EP (for example, the same value as the maximum value, a value that is twice the maximum value, etc.) ) Can be set.

  FIG. 2C shows a predicted speed transition determined based on the speed predicted value setting rule shown in FIG. 4 and the acceleration predicted value setting rule shown in FIG. 5 in the example shown in FIG. In this example, at the second point P2 where the first stop line sign 52 is installed, the vehicle speed becomes the third predicted speed Vc (zero in this example) corresponding to the non-priority intersection, and the second point P2 On the other hand, the acceleration related to deceleration when approaching from the rear direction D2 side in the traveling direction is “αa”, and the acceleration related to acceleration when moving away from the second point P2 toward the front direction D1 is “αb”. Speed transition has been determined. Since the predicted speed transition is thus determined, in this embodiment, the acceleration / deceleration start point and end point in the predicted speed transition are the position of the intersection 70, the predicted acceleration value (see FIG. 5), and the predicted speed value (see FIG. 4).

  FIG. 3 is a diagram showing a driving situation different from that in FIG. 2, and FIGS. 3 (a), 3 (b), and 3 (c) are respectively shown in FIGS. 2 (a) and 2 (b). This corresponds to FIG. In the example shown in FIG. 3, the intersection 70 on the front direction D1 side of the own vehicle 3 is not a non-priority intersection but a priority intersection, and crosses the second intersection shape mark 55 on the front direction D1 side of the vehicle 3. The feature arrangement section A is set so as to include the sidewalk sign 56.

  By the way, as shown in FIGS. 2 and 3, the feature arrangement section A is basically set so that the vehicle 3 goes straight through the intersection 70, so both sides sandwiching the intersection 70 of the feature arrangement section A The extending directions of these parts are parallel to each other. In addition, when the route guidance part 83 is performing route guidance, it can be set as the structure where the feature arrangement | positioning area A is set along a guidance route. In such a configuration, depending on the guide route, the feature arrangement section A is set to bend at the intersection 70, and in this case, the extending direction of each part on both sides across the intersection 70 of the feature arrangement section A is The directions cross each other.

2-6. Configuration of Recognition Rate Determination Unit The recognition rate determination unit 26 is a functional unit that determines the recognition rate R of the feature 50 included in the feature arrangement section A based on the feature information FT and the predicted speed transition. Here, the recognition rate R is an index (success rate estimation value) indicating the rate of successful image recognition of the feature 50. In order to realize such a recognition rate R determination function, the recognition rate determination unit 26 includes a reference recognition rate acquisition unit 31 and a correction coefficient determination unit 32, and the recognition rate determination unit 26 includes the reference recognition rate acquisition unit 31. Is obtained by determining the value obtained by correcting the reference recognition rate RS obtained by the correction coefficient C determined by the correction coefficient determination unit 32 as the recognition rate R. In the present embodiment, the recognition rate determination unit 26 derives and determines the recognition rate R by an operation of multiplying the reference recognition rate RS by the correction coefficient C. In the present embodiment, the recognition rate determination unit 26 is a feature 50 included in the feature arrangement section A and can be a recognition target in the image recognition processing by the image processing unit 22 (in this example, FIG. 6). The recognition rate R is determined for all of the road markings shown in FIG.

2-6-1. Configuration of Reference Recognition Rate Acquisition Unit The reference recognition rate acquisition unit 31 is a functional unit that acquires the reference recognition rate RS for the feature 50 whose recognition rate R is to be determined. In the present embodiment, the reference recognition rate RS is stored in the reference recognition rate storage unit 11 in association with the type of the feature 50, and the reference recognition rate acquisition unit 31 refers to the reference recognition rate storage unit 11. A necessary reference recognition rate RS is acquired.

  Here, the reference recognition rate RS represents an average success rate (success probability) of image recognition, which is predetermined according to the type of the feature 50, and is set to a value of 0 or more and 1 or less. . In the example shown in FIG. 6, each of the reference recognitions of the traveling direction sign 51, the first stop line sign 52, the second stop line sign 53, the first intersection shape sign 54, the second intersection shape sign 55, and the pedestrian crossing sign 56. The rate RS is set to “RS1”, “RS2”, “RS3”, “RS4”, “RS5”, and “RS6”. For example, RS1 is “0.6”, RS2 is “0.8”, RS3 is “0.7”, RS4 is “0.7”, RS5 is “0.8”, and RS6 is “0.7”. Can be set.

2-6-2. Configuration of Correction Coefficient Determination Unit The correction coefficient determination unit 32 is a functional unit that determines the correction coefficient C based on the predicted speed transition determined by the predicted speed transition determination unit 27. The correction coefficient C is a correction coefficient for correcting the reference recognition rate RS. Specifically, the correction coefficient C is a correction caused by the speed of the host vehicle 3 when performing image recognition processing of the feature 50 by the image processing unit 22. It is a correction coefficient for performing. In the present embodiment, the correction coefficient C corresponds to the “speed factor correction coefficient” in the present invention.

  As described above, in the present embodiment, the recognition rate R of the feature 50 decreases as the vehicle speed at the time of photographing the feature 50 decreases, and the acceleration of the host vehicle 3 at the time of photographing the feature 50. There is a tendency for the value to decrease as the value increases. In view of this tendency, in the present embodiment, the correction coefficient determination unit 32 determines the correction coefficient C so that the recognition rate R decreases as the speed of the host vehicle 3 indicated by the predicted speed transition decreases, and the predicted speed transition The correction coefficient C is determined so as to lower the recognition rate R as the acceleration of the host vehicle 3 indicated by (1) increases. In this embodiment, as described above, the recognition rate R is a value obtained by multiplying the reference recognition rate RS by the correction coefficient C. Therefore, the correction coefficient C is determined to be a value that decreases as the speed of the host vehicle 3 indicated by the predicted speed transition decreases, and is determined to be a value that decreases as the acceleration of the host vehicle 3 indicated by the predicted speed transition increases. The

  The correction coefficient determination unit 32 determines the correction coefficient C based on the correction coefficient setting rule. At this time, in order to determine the correction coefficient C in consideration of the above-mentioned tendency with respect to the vehicle speed and acceleration of the host vehicle 3, in this embodiment, the correction coefficient setting rule is a speed deduction point setting rule and an acceleration deduction point setting rule. It consists of and. Here, in the speed deduction point setting rule, as shown in an example in FIG. 7, a speed deduction point value ΔX is defined in association with the speed V, and a larger speed deduction point value ΔX is defined as the speed V becomes lower. Yes. In the example shown in FIG. 7, the speed V is divided into three speed ranges, and “0” is defined as the speed deduction point value ΔX for the speed V higher than the second reference speed V2, and the second reference speed V2 A positive value “ΔX1” is defined as a speed deduction point value ΔX for a speed V that is equal to or lower than the first reference speed V1 (<V2), and for a speed V that is equal to or lower than the first reference speed V1. “ΔX2” larger than ΔX1 is defined as the speed deduction point value ΔX. For example, V1 can be 10 [km / h], V2 can be 20 [km / h], ΔX1 can be “0.1”, and ΔX2 can be “0.2”.

  Further, as shown in FIG. 8, the acceleration deduction point setting rule defines an acceleration deduction point value ΔY in association with the acceleration α, and the acceleration α becomes higher on the acceleration side or the deceleration side (that is, the absolute value). A larger acceleration deduction point value ΔY is defined as In the example shown in FIG. 8, the acceleration α is divided into three acceleration ranges, and an acceleration deduction is made for an acceleration α that is larger than the first reference acceleration α1 (<0) and smaller than the second reference acceleration α2 (> 0). “0” is defined as the value ΔY, a positive value “ΔY1” is defined as the acceleration deduction point value ΔY for the acceleration α below the first reference acceleration α1, and the acceleration deduction point for the acceleration α above the second reference acceleration α2. A positive value “ΔY2” is defined as the value ΔY. For example, both α1 and α2 can be set to 0.1 [G], and both ΔY1 and ΔY2 can be set to “0.1”.

  Then, the correction coefficient determination unit 32 shows the speed deduction point value ΔX and the acceleration deduction point value ΔY corresponding to the speed and acceleration of the host vehicle 3 indicated by the predicted speed transition at the arrangement position of the feature 50 for which the recognition rate R is determined. The correction coefficient C is determined to be a value obtained by subtracting both ΔX and ΔY from “1” (1−ΔX−ΔY).

  The example shown in FIG. 2 will be specifically described. At the first point P1, which is the position where the traveling direction indicator 51 is arranged, the speed indicated by the predicted speed transition is “Va” and the acceleration indicated by the predicted speed transition is “0”. Is. Since Va> V2 in the example shown in FIG. 2, both the speed deduction point value ΔX and the acceleration deduction point value ΔY are “0”, and the correction coefficient C is determined to be “1”. Therefore, the recognition rate R of the traveling direction sign 51 is determined to be the same value as “RS1” that is the reference recognition rate RS. On the other hand, for the first stop line marking 52, at the second point P2 where the first stop line marking 52 is arranged, the speed indicated by the predicted speed transition is “Vc” and whether the predicted speed transition indicates the acceleration. Is “αa”. In the example shown in FIG. 2, since Vc <V1 and αa ≦ α1, the speed deduction point value ΔX is “ΔX2”, and the acceleration deduction point value ΔY is “ΔY1”. Therefore, the recognition rate R of the first stop line marking 52 is determined to be smaller than “RS2” that is the reference recognition rate RS.

  This will be further described using a specific numerical example. When RS1 = 0.6, RS2 = 0.8, ΔX2 = 0.2, and ΔY1 = 0.1, the reference direction recognition 51 is used for the traveling direction indicator 51. Both the rate RS and the recognition rate R are “0.6”. On the other hand, for the first stop line marking 52, the reference recognition rate RS is “0.8”, whereas the correction coefficient C is “0.7”, so the recognition rate R is “0.56”. . Therefore, when the reference recognition rates RS are compared with each other, the value of the first stop line marking 52 becomes larger, whereas when the recognition rates R with the predicted speed transition taken into consideration are compared with each other, the traveling direction indication The value of 51 is larger.

  By the way, in the present embodiment, the feature 50 (hereinafter referred to as “specific feature”) having the same shape regardless of the position in the road extending direction L over the entire recognition target portion in the image recognition processing. The recognition rate determination unit 26 is configured to execute exception processing for determining the recognition rate R by reducing the correction amount by the correction coefficient C. Specifically, in the pedestrian crossing sign 56 shown in FIG. 6, the hatched portion indicated by reference numeral “50a” is a recognition target portion in the image recognition processing, and the portion excluding the recognition target portion 50a (in the road extending direction L). The portion extending in the orthogonal road width direction) is not considered in determining whether image recognition is successful. That is, this pedestrian crossing sign 56 is the above-mentioned specific feature, and when determining the recognition rate R, exception processing for reducing the correction amount by the correction coefficient C is executed. For example, for the pedestrian crossing sign 56 shown in FIG. 3, the correction coefficient C when the exception process is not executed is (1−ΔY2). However, when the exception process is executed, the value of the correction coefficient C is A value greater than (1−ΔY2) and equal to or less than 1 (for example, “1”) is determined.

  The reason for executing such exception processing will be described below. As described above, in the present embodiment, since a plurality of frames in which the recognition target feature 50 is captured are combined according to the vehicle speed at the time of shooting and image recognition is performed, if an error is included in the vehicle speed, the combination is performed. The shape of the feature 50 obtained as described above is a shape obtained by lengthening or shortening the actual shape of the feature along the extending direction L of the road. With regard to this point, in the case of the above-mentioned specific feature, the matching requirement in the pattern matching process for the positional relationship between the portions at different positions along the extending direction L in the same feature Even if it is relaxed compared to 50, image recognition can be performed appropriately. For this reason, the present embodiment adopts a configuration that executes exception processing for the specific feature.

  In the example shown in FIG. 6, the first stop line marking 52 and the second stop line marking 53 can also be the above-mentioned specific features, so that the exception processing is executed in the same manner as the pedestrian crossing marking 56. can do. It is also possible to adopt a configuration in which the exception process is not executed for the specific feature.

2-7. Configuration of Recognition Target Determination Unit The recognition target determination unit 25 determines a feature 50 (hereinafter referred to as “recognition target feature”) as a recognition target in the image recognition processing by the image processing unit 22 based on the recognition rate R. Part. In the present embodiment, when there are a plurality of features 50 whose recognition rate R is determined by the recognition rate determining unit 26 in the feature arrangement section A, the feature 50 having the highest recognition rate R is recognized as the recognition target. Decide on features. In the present embodiment, if there is only one feature 50 having the recognition rate R determined by the recognition rate determination unit 26 in the feature arrangement section A, the feature 50 is recognized as the recognition target feature. To decide.

  Note that when a lower limit value is set as a condition of the recognition rate R for selecting as a recognition target feature, and a feature 50 with the recognition rate R equal to or higher than the lower limit value does not exist in the feature arrangement section A, either The feature 50 may be configured not to be selected as the recognition target feature. In addition, when one feature 50 is image-recognized in the feature arrangement section A, it is possible to further perform image recognition for other features 50 included in the feature arrangement section A. is there.

3. Procedure of Operation Process Next, with reference to FIG. 9 and FIG. 10, a procedure of the vehicle position recognition process executed in the vehicle position recognition unit 2 according to the present embodiment, that is, a vehicle position recognition method will be described. . The procedure of the vehicle position recognition process described below is executed by hardware and / or software (program) or both constituting each functional unit of the navigation device 1 (vehicle position recognition unit 2). When each of the above function units is configured by a program, the arithmetic processing device included in the navigation device 1 operates as a computer that executes the program that configures each of the above function units. FIG. 10 is a flowchart showing the procedure of the estimated host vehicle position correction process in step # 04 of FIG.

3-1. Overall Procedure of Own Vehicle Position Recognition Processing As shown in FIG. 9, when the estimated own vehicle position acquisition unit 21 acquires the estimated own vehicle position EP (step # 01), the map information acquisition unit 23 sets the vehicle 3 in front. The information of the feature 50 (feature information FT) arranged in the predetermined search range is acquired, and the search for the feature 50 that can be recognized in the image recognition processing by the image processing unit 22 within the search range. (Step # 02). If there is a feature 50 that can be recognized within the search range (step # 03: Yes), an estimated vehicle position correction process described later based on FIG. 10 is executed (step # 04). On the other hand, when there is no feature 50 that can be recognized in the search range (step # 03: No), the process ends. This own vehicle position recognition process is repeatedly executed at a predetermined timing.

3-2. Procedure of Estimated Vehicle Position Correction Process Next, the estimated vehicle position correction process in step # 04 of FIG. 9 will be described with reference to FIG. After the feature arrangement section A is set based on the feature information FT acquired in step # 02 of FIG. 9 (step # 11), the road information RD of the feature arrangement section A in front of the vehicle 3 is a map. Obtained by the information obtaining unit 23 (step # 12). Then, based on the road information RD, the predicted speed transition determining unit 27 determines the predicted speed transition in the feature arrangement section A (step # 13). Next, the recognition rate determination unit 26 determines the recognition rate R based on the predicted speed transition determined in step # 13 and the feature information FT acquired in step # 02 of FIG. 9 (step # 14) Based on the determined recognition rate R, the recognition target determining unit 25 determines the recognition target feature that is the feature 50 to be recognized in the image recognition processing by the image processing unit 22 (step # 15). . And the image recognition process about this determined recognition object feature is performed by the process after step # 16.

  Specifically, the estimated vehicle position EP (step # 16) acquired by the estimated vehicle position acquisition unit 21 is an image recognition section (in this example, the same as the feature arrangement section A) in which image recognition processing is performed. ) (Step # 17: Yes), the image processing unit 22 acquires the captured image PG (step # 18) and executes the image recognition processing of the recognition target feature determined in step # 15. (Step # 19). When the recognition target feature is image-recognized (when the image recognition is successful) (step # 20: Yes), the estimated own vehicle position EP is corrected based on the feature information FT of the recognition target feature. Processing is executed (step # 21). On the other hand, when the recognition target feature is not recognized (when the image recognition fails) (step # 20: No), the estimated own vehicle position EP acquired by the estimated own vehicle position acquisition unit 21 (step # 22). ) Is located in the feature arrangement section A (step # 23: Yes), the processing from step # 18 is repeatedly executed.

4). Other Embodiments Finally, other embodiments of the vehicle position recognition system according to the present invention will be described. Note that the configurations disclosed in the following embodiments can be applied in combination with the configurations disclosed in other embodiments as long as no contradiction arises.

(1) In the example shown in FIG. 4 referred to in the above embodiment, the configuration in which the speed prediction value setting rule defines the speed prediction value for each of the road sections divided into four parts has been described as an example. However, the embodiment of the present invention is not limited to this, and the speed prediction value setting rule defines the speed prediction value for each of the road sections divided into two, three, or five or more. It can also be configured. For example, the speed prediction value setting rule may be configured to define speed prediction values for three road portions of “non-intersection section (cruising section)”, “non-priority intersection”, and “priority intersection”. is there. In this case, the “non-intersection section” corresponds to the combination of the “straight section” and the “curve section” in FIG. 4, and the speed prediction value corresponding to the “non-intersection section” is the above-described “straight section”. It can be set in the same manner as the predicted speed value corresponding to the “curve section”. In addition, for example, the predicted speed values defined for straight sections and curved sections are further subdivided, and the predicted speed values are defined based on the number of lanes, the slope of the road, the presence or absence of a pedestrian crossing, etc. You can also. Furthermore, it can also be set as the structure by which the speed predicted value about specific points (for example, a crossing, a toll booth, etc.) other than the intersection 70 was prescribed | regulated.

(2) In the example shown in FIG. 5 referred to in the above embodiment, a configuration in which predicted acceleration values for two traveling conditions (“deceleration” and “acceleration”) are defined by an acceleration predicted value setting rule is taken as an example. explained. However, the embodiment of the present invention is not limited to this. At least one of “deceleration” and “acceleration” is subdivided, and the acceleration prediction value setting rule sets acceleration prediction values for three or more driving conditions. It is also possible to adopt a defined configuration. For example, when the acceleration prediction value setting rule defines acceleration prediction values corresponding to two of “first deceleration” and “second deceleration” for deceleration, based on the type of intersection, the presence or absence of a pedestrian crossing, etc. Any one of “first deceleration” and “second deceleration” may be selected. The same applies to acceleration.

(3) In the example shown in FIG. 7 referred to in the above embodiment, the configuration in which the speed V is divided into three speed ranges has been described as an example. However, the speed V is divided into two or four or more speed ranges. It is also possible to adopt the configuration described above. Similarly, in the example shown in FIG. 8 referred to in the above embodiment, the configuration in which the acceleration α is divided into three acceleration regions has been described as an example. However, the acceleration α is divided into two or more acceleration regions. It is also possible to adopt the configuration described above.

(4) In the above embodiment, the correction coefficient determination unit 32 determines the correction coefficient C based on the speed deduction point setting rule as shown in FIG. 7 and the acceleration deduction point setting rule as shown in FIG. The configuration has been described as an example. However, the embodiment of the present invention is not limited to this, and the correction coefficient determination unit 32 derives the correction coefficient C based on the following equation (1) and determines the correction coefficient C. You can also.
C = 1−Z1 / V−Z2 × α (1)
Here, “V” represents the speed of the host vehicle 3, and “α” represents the acceleration of the host vehicle 3. “Z1” and “Z2” are weighting coefficients for speed and acceleration, and can be set based on the performance of the navigation device 1, for example. Also, “Z1” and “Z2” can be used as learning values.
Further, a correction coefficient map that defines the relationship between the speed and acceleration of the host vehicle 3 and the correction coefficient C is stored in the storage device, and the correction coefficient determination unit 32 determines the correction coefficient C with reference to the storage device. A configuration is also possible.

(5) In the above embodiment, the configuration in which the correction coefficient determination unit 32 determines the correction coefficient C based on both the speed and acceleration of the host vehicle 3 indicated by the predicted speed transition has been described as an example. However, the embodiment of the present invention is not limited to this, and the correction coefficient determination unit 32 corrects the correction coefficient based on only one of the speed and acceleration of the host vehicle 3 indicated by the predicted speed transition (for example, only the speed). It can also be set as the structure which determines C.

(6) In the above embodiment, the acceleration / deceleration start point and end point in the predicted speed transition are determined based on the position of the intersection 70, the predicted acceleration value (see FIG. 5), and the predicted speed value (see FIG. 4). Was described as an example. However, the embodiment of the present invention is not limited to this, and is provided with a setting rule regarding the start point or the end point of acceleration / deceleration, and the acceleration in the predicted speed transition is not a predetermined rule but the intersection 70 A configuration that is determined based on the position, the start or end point of acceleration / deceleration, and the predicted speed value (see FIG. 4) is also possible.

(7) In the above embodiment, the configuration in which the predicted speed transition determining unit 27 determines the predicted speed transition in the feature arrangement section A based on the road information RD has been described as an example. However, the embodiment of the present invention is not limited to this, and at least if it is based on the road information RD, the predicted speed transition determination unit 27 may further calculate the predicted speed transition based on one or more other information. It is also possible to adopt a configuration to determine. For example, the predicted speed transition determining unit 27 may determine whether the road information RD and the traffic information in the feature arrangement section A are based on both the road information RD and the current traveling state (for example, vehicle speed) of the host vehicle 3. It can be set as the structure which determines prediction speed transition based on both.

(8) In the above embodiment, the configuration in which the reference recognition rate RS is predetermined according to the type of the feature 50 has been described as an example. However, the embodiment of the present invention is not limited to this, and the reference recognition rate RS may be determined based on one or more other information in addition to the type of the feature 50. It is. Such other information includes, for example, information on the fading state of the feature 50, information on the sunshine state on the feature 50 (for example, whether it is included in the sun or in the shade), weather conditions, time, etc. It can be information.

(9) In the above embodiment, when the predicted speed transition determining unit 27 includes the intersection 70 in the feature arrangement section A, the speed of the host vehicle 3 when passing through the intersection 70 is the speed of the intersection 70. The configuration in which the predicted speed transition is determined so that the speed depends on the type is described as an example. However, the embodiment of the present invention is not limited to this, and it is possible to adopt a configuration in which the presence or absence of the intersection 70 is not taken into consideration when determining the predicted speed transition.

(10) In the above-described embodiment, the configuration in which the recognition rate determination unit 26 derives and determines the recognition rate R by an operation of multiplying the reference recognition rate RS by the correction coefficient C has been described as an example. However, the embodiment of the present invention is not limited to this, and a recognition rate map that defines the relationship between the reference recognition rate RS and the correction coefficient C and the recognition rate R is stored in the storage device. The recognition rate determination unit 26 can also be configured to determine the recognition rate R with reference to FIG. In the above-described embodiment, the configuration in which the recognition rate R is a value obtained by multiplying the reference recognition rate RS by the correction coefficient C has been described as an example. However, the recognition rate R sets the correction coefficient C to the reference recognition rate RS. It is also possible to adopt a configuration in which the value is divided or a configuration in which the recognition rate R is a value obtained by adding or subtracting the correction coefficient C to the reference recognition rate RS.

(11) In the above embodiment, the feature arrangement section A is divided and arranged in the road extending direction L, and the interval S along the extending direction L of the adjacent features 50 is the predetermined distance T. As for the following group of features, the configuration set to include all the features 50 constituting the group of features has been described as an example. However, the embodiment of the present invention is not limited to this. In the configuration in which the recognition request range, which is the range in which the image recognition process is performed, is set for each feature 50 along the road extending direction L. The feature arrangement section A is set so as to include all the features 50 constituting the group of features with respect to the group of features whose recognition requirement ranges of the adjacent feature 50 overlap each other. be able to.

(12) In the above embodiment, the configuration in which the image recognition section that defines the range in which the image recognition process is performed matches the feature arrangement section A has been described as an example. However, the embodiment of the present invention is not limited to this, and the image recognition section may be set independently of the feature arrangement section A. In this case, the feature arrangement section A can be set in accordance with the road section occupied by the feature 50 (that is, the first predetermined distance and the second predetermined distance described above are set to zero). In this case, the image recognition section can be set based on the recognition request range, for example.

(13) In the above embodiment, the configuration in which the target feature 50 in the image recognition for correcting the estimated host vehicle position EP is a road marking provided on the road surface has been described as an example. However, the embodiment of the present invention is not limited to this, and a three-dimensional feature 50 such as a traffic light can be used for correcting the estimated host vehicle position EP. In this case, the installation mode of the camera 90 on the vehicle 3 is set so that the image of the three-dimensional feature 50 can be recognized.

(14) In the above embodiment, the configuration in which the camera 90 is a back camera that captures the rear of the vehicle 3 has been described as an example. However, the embodiment of the present invention is not limited to this, and the camera 90 may be configured as a front camera that captures the front of the vehicle 3 or a side camera that captures the side of the vehicle 3. It is.

(15) In the above embodiment, the case where the vehicle position recognition system according to the present invention is applied to the in-vehicle navigation device 1 has been described as an example. However, such a navigation device 1 is fixed to the vehicle 3. The portable navigation device is not limited to the one that is mounted, and may be a portable navigation device that can be arbitrarily moved, and may be brought into the vehicle 3 only at the time of use.

(16) In the above embodiment, the case where the vehicle position recognition system according to the present invention is applied to the navigation device 1 has been described as an example. However, the vehicle position recognition system according to the present invention is not limited to other types provided in a vehicle. It is also possible to apply to a device (running control device or the like).

(17) The assignment of the functional units described in the above embodiment is merely an example, and a plurality of functional units can be combined or one functional unit can be further divided. Further, in the above-described embodiment, each functional unit constituting the own vehicle position recognition unit 2 and each storage unit 10 and 11 have been described as an example of a configuration provided in the vehicle 3, but at least the own vehicle position recognition unit 2 is included. A configuration in which some of the functional units and at least a part of the storage units 10 and 11 are provided outside the vehicle 3 (for example, a server device or the like), and information is transmitted and received via a communication network such as the Internet. You can also

(18) Regarding other configurations as well, the embodiments disclosed herein are illustrative in all respects, and the embodiments of the present invention are not limited thereto. In other words, configurations that are not described in the claims of the present application can be modified as appropriate without departing from the object of the present invention.

  INDUSTRIAL APPLICABILITY The present invention can be suitably used for a host vehicle position recognition system, a host vehicle position recognition program, and a host vehicle position recognition method for recognizing a host vehicle position using an image captured by a camera provided in the host vehicle.

3: Vehicle (own vehicle)
21: Estimated vehicle position acquisition unit 22: Image processing unit 23a: Feature information acquisition unit 23b: Road information acquisition unit 24: Estimated vehicle position correction unit 25: Recognition target determination unit 26: Recognition rate determination unit 27: Predictive speed Transition determination unit 31: reference recognition rate acquisition unit 32: correction coefficient determination unit 50: feature 50a: recognition target portion 60: traveling road 61: other road 70: intersection 90: camera A: feature arrangement section C: correction coefficient (Speed factor correction coefficient)
EP: estimated vehicle position FT: feature information L: extending direction PG: captured image R: recognition rate RD: road information RS: reference recognition rate S: interval T: predetermined distance Vc: third predicted speed (type-dependent speed) )
Vd: Fourth predicted speed (type-dependent speed)

Claims (10)

An estimated host vehicle position acquisition unit that acquires an estimated host vehicle position that is an estimated position of the host vehicle, and a captured image captured by a camera provided in the host vehicle are acquired, and an image of a feature included in the captured image Included in the image processing unit that performs recognition processing, the feature information acquisition unit that acquires feature information including position information and type information of the feature, the recognition result of the feature by the image recognition processing, and the feature information An estimated vehicle position correcting unit that corrects the estimated vehicle position based on position information of a feature, and a vehicle position recognition system comprising:
A road information acquisition unit for acquiring road information;
Prediction of speed transition of the host vehicle in the feature placement section based on at least the road information before the estimated vehicle position enters the feature placement section that is a road section in which the feature is placed. A predicted speed transition determining unit that determines a predicted speed transition indicating
Based on the feature information and the predicted speed transition, a recognition rate determination unit that determines a recognition rate that is a rate of successful image recognition of features included in the feature placement section;
A vehicle position recognition system comprising: a recognition target determining unit that determines a feature to be recognized in the image recognition processing based on the recognition rate. The road information includes information for specifying the presence or absence of an intersection and the type of intersection,
The predicted speed transition determining unit, when an intersection is included in the feature arrangement section, the speed of the host vehicle when passing through the intersection is a type-dependent speed according to the type of the intersection The host vehicle position recognition system according to claim 1, wherein a predicted speed transition is determined.   The type-dependent speed when the traveling road on which the vehicle travels before entering the intersection included in the feature arrangement section is a priority road with respect to another road that intersects the traveling road at the intersection, The host vehicle position recognition system according to claim 2, wherein the vehicle position recognition system is determined to be a value higher than the type-dependent speed when the other road is a priority road for the traveling road. A correction coefficient determination unit that determines a speed factor correction coefficient that is a correction coefficient related to the recognition rate caused by the speed of the host vehicle when performing image recognition processing of a feature by the image processing unit based on the predicted speed transition. Further comprising
The said recognition rate determination part determines the value which correct | amended the reference | standard recognition rate predetermined according to the classification of the feature by the said speed factor correction coefficient to the said recognition rate as described in any one of Claim 1 to 3 The vehicle position recognition system described.   5. The host vehicle position recognition system according to claim 4, wherein the correction coefficient determination unit determines the speed factor correction coefficient so as to lower the recognition rate as the acceleration of the host vehicle indicated by the predicted speed transition increases.   The vehicle position recognition according to claim 4 or 5, wherein the correction coefficient determination unit determines the speed factor correction coefficient so that the recognition rate decreases as the speed of the host vehicle indicated by the predicted speed transition decreases. system.   The recognition rate determination unit is a feature provided on the road surface, and the feature having the same shape regardless of the position in the extending direction of the road throughout the recognition target portion in the image recognition process. The host vehicle position recognition system according to any one of claims 4 to 6, wherein the recognition rate is determined by reducing a correction amount by the speed factor correction coefficient.   For a group of features that are arranged separately in the direction in which the road extends and whose interval along the direction of extension of adjacent features is less than or equal to a predetermined distance, all the features that make up the group of features The vehicle position recognition system according to any one of claims 1 to 7, wherein the feature arrangement section is set so as to include an object. An estimated own vehicle position acquisition function for acquiring an estimated own vehicle position that is an estimated position of the own vehicle, and a photographed image taken by a camera provided in the own vehicle, and an image of a feature included in the photographed image Included in the image processing function for performing the recognition process, the feature information acquisition function for acquiring the feature information including the position information and the type information of the feature, the recognition result of the feature by the image recognition process, and the feature information A vehicle position recognition program for realizing an estimated vehicle position correction function for correcting the estimated vehicle position based on position information of a feature,
A road information acquisition function for acquiring road information;
Prediction of speed transition of the host vehicle in the feature placement section based on at least the road information before the estimated vehicle position enters the feature placement section that is a road section in which the feature is placed. A predicted speed transition determining function for determining a predicted speed transition indicating
Based on the feature information and the predicted speed transition, a recognition rate determination function for determining a recognition rate that is a rate of successful image recognition of the features included in the feature placement section;
A vehicle position recognition program for causing a computer to realize a recognition target determination function for determining a feature to be recognized in the image recognition processing based on the recognition rate. An estimated vehicle position acquisition step for acquiring an estimated vehicle position that is an estimated position of the vehicle, a captured image captured by a camera provided in the vehicle, and an image of a feature included in the captured image Included in the image processing step for performing the recognition process, the feature information acquisition step for acquiring the feature information including the position information and the type information of the feature, the recognition result of the feature by the image recognition process, and the feature information An estimated vehicle position correcting step for correcting the estimated vehicle position based on position information of a feature, and a vehicle position recognition method comprising:
A road information acquisition step for acquiring road information;
Prediction of speed transition of the host vehicle in the feature placement section based on at least the road information before the estimated vehicle position enters the feature placement section that is a road section in which the feature is placed. A predicted speed transition determining step for determining a predicted speed transition indicating
A recognition rate determination step for determining a recognition rate, which is a rate of successful image recognition of features included in the feature placement section, based on the feature information and the predicted speed transition;
A vehicle position recognition method comprising: a recognition target determination step of determining a feature to be recognized in the image recognition process based on the recognition rate.

Images