JP4326992B2 - Peripheral vehicle identification method and driving support system using the same - Google Patents

Description

  The present invention relates to a driving support system for an automobile, and in particular, a surrounding vehicle identification method for identifying other vehicles based on data from an input unit such as a camera provided in the host vehicle and a position specifying unit such as a navigation device, and the like The present invention relates to a driving support system that performs driving support to a driver based on information obtained by a method.

  In recent years, navigation devices have not only a simple route search function but also a driving support function for a driver for comfortable driving by capturing various information in real time and providing information to the driver. Things are being developed. In particular, as a result of advances in sensor technology and downsizing of digital cameras, etc., radar sensors and small cameras are mounted on the vehicle to capture information on the surrounding road conditions and driving environment, etc. It has been proposed to present information. For example, research and development of a system for recognizing a vehicle type of another vehicle or detecting an approach of another vehicle to give a warning is underway.

  As described above, as a device for executing a driving support operation for a driver using a navigation system and a means for inputting information from the outside of the host vehicle as described above, a road condition detection means such as a navigation system is disclosed in Patent Document 1. And a control content determining means for determining a control content of a vehicle operating mechanism based on information from the road condition detecting means and the travel environment detecting means such as a radar cruise system. An apparatus is disclosed.

JP 2000-46169 A

  As shown in FIG. 1, the vehicle control device disclosed in Patent Document 1 detects a traveling environment such as an inter-vehicle distance between the host vehicle and another vehicle ahead by a traveling environment detecting means, that is, a radar cruise system 1, An engine electronic control unit (E-ECU) 4 for controlling the operation of the prime mover 5 based on the determination of the control content determination means from these detection data, for example, by the situation detection means, that is, the navigation system 2, and the shift It is configured to control an operation mechanism 3 including a transmission electronic control unit (T-ECU) 6 and the like for controlling the operation of the machine 7. According to such a configuration, since the control content of the operation mechanism 3 is determined in consideration of the driving environment such as the inter-vehicle distance in addition to the road conditions such as corners and slopes, the driver's suitable for the actual surrounding situation is determined. The intention can be automatically reflected in the control of the operation mechanism to assist the driver.

  However, recently, as a driving support operation for the driver, not only an object such as another vehicle is detected in front of the host vehicle, but also a prediction information for the next optimal driving operation is quickly presented to the driver. There is a need for technology to do this. In particular, operations such as identifying the vehicle type and traveling direction of other vehicles in front of the host vehicle, and giving other vehicle information to the driver or issuing a warning when an avoidance operation is expected, are practical. It is considered as a research subject as an issue that has not yet been realized. The vehicle control device disclosed in Patent Document 1 calculates an appropriate gear ratio based on the inter-vehicle distance information from the radar cruise system 1 and the road information ahead from the navigation system 2, and automatically sets the transmission. By controlling the gear ratio, the driver's smooth and comfortable driving operation is supported. As mentioned above, the traveling direction of other vehicles is predicted with high accuracy and information is presented to the driver. There is no description of the operation such as doing.

  Accordingly, the present invention has been made to solve the above-described problems, and an object of the present invention is to provide a surrounding vehicle identification method that can accurately detect the type and orientation of another vehicle.

  Another object of the present invention is to provide a driving support system that accurately detects the type and orientation of another vehicle and presents driving support information to the driver based on the detected information.

In order to solve the above-described problems, the present invention provides a detection unit that detects an object existing around a host vehicle and an image data generation unit that captures the object and generates image data. And an arithmetic unit that calculates a relative position of the object with respect to the own vehicle, and a current position and the relative position of the own vehicle on map data. By determining whether the vehicle is located in a lane, a prediction unit that predicts the direction of the other vehicle , and a plurality of shapes that include a glass position and a mirror position viewed from different directions for one vehicle type Based on the template data corresponding to the predicted direction of the template data and the image data, a determination unit that determines the vehicle type and direction of the other vehicle, and the determined other vehicle A driving support system characterized by an output unit for outputting information including vehicle type and orientation. Thereby, the vehicle type and direction of the other vehicle can be detected with high accuracy, and the driving support information can be presented to the driver based on the detected information.

The determination unit may refer to a plurality of template data among the template data and also refer to a template data generated by interpolating the plurality of template data.

The said prediction part can be set as the structure which estimates the direction of each said other vehicle about each other vehicle, when several other vehicles exist around the said own vehicle.

The present invention also includes a step of detecting an object existing around the host vehicle, a step of capturing the object to generate image data, a step of calculating a relative position of the object with respect to the host vehicle, and map data Predicting the direction of the other vehicle by determining in which lane the other vehicle is positioned based on the current position and the relative position of the host vehicle as the other vehicle. And the template data corresponding to the predicted orientation among the plurality of template data indicating the shape, including the position of the glass and the position of the mirror, as seen from different directions for one vehicle type , and the image data, And determining the vehicle type and direction of the other vehicle based on the above. Thereby, the vehicle type and direction of other vehicles can be detected with high accuracy.

  ADVANTAGE OF THE INVENTION According to this invention, the vehicle type and direction of other vehicles can be detected with a sufficient precision, and the driving assistance information with respect to a driver | operator can be shown based on the detected information.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.

  FIG. 2 is a block diagram illustrating a configuration of the driving support system 100 according to the first embodiment. The driving support system 100 is roughly composed of a control device 101 and a navigation device 102. The control device 101 includes a radar sensor 103 and an in-vehicle camera 104 provided in front of the host vehicle, a control unit 105, a template database 106, and a memory 107. The radar sensor 103 functions as a detection unit that detects an object that exists in the vicinity of the host vehicle. For example, the radar sensor 103 is a sensor that uses a millimeter wave radar that uses radio waves in the 76 GHz band, and has a detection range of 16 ° to 80 ° in front. Have. The in-vehicle camera 104 functions as an image data generation unit that captures an object and generates image data, and is, for example, a high-resolution CCD (charge coupled device) camera. The template database 106 is a database that stores, as a data table, template image data corresponding to the vehicle type and direction of another vehicle to be described later. The memory 107 stores an operation routine, which will be described later, and temporarily stores identification information of other vehicles. The control unit 105 is connected to the radar sensor 103, the in-vehicle camera 104, the template database 106, and the memory 107, and controls the operation of the driving support system 100.

  The navigation device 102 includes a navigation controller 108, a map database 109, an image display unit 110, and an audio output unit 111. The navigation controller 108 is connected to the map database 109, the image display unit 110, and the audio output unit 111, and controls the entire navigation device 102. The navigation controller 108 is connected to the control unit 105 of the control device 101 so as to be able to perform data communication with each other. The map database 109 is composed of, for example, a hard disk, a DVD-RW (DVD Rewritable), etc., and is necessary for route guidance, for example, a map data file storing map information, an intersection data file storing information related to an intersection, a road type, For road data files that store information about roads such as the start and end points of each road, node data files that store east longitude and north latitude coordinates at one point on the road, and usage purposes classified into genres such as gas stations and convenience stores A guide point data file or the like storing the position coordinates of the feature such as the corresponding facility or the guide information related to the feature is stored. The map database 109 stores route search data that corresponds to the hierarchically structured map data and is hierarchized from a lower level with a large amount of road network information to an upper level with a small amount of road network information. . The image display unit 110 is configured by a liquid crystal display or the like, and based on drawing data output from the navigation controller 108, a map image around the host vehicle, a vehicle position mark, a departure point mark, a destination mark, a guidance route, etc. Displays the destination setting screen. The audio output unit 111 is, for example, a speaker, and transmits various audio data from the navigation controller 108 to the driver as audio. The image display unit 110 or the audio output unit 111 functions as an output unit that outputs information based on the vehicle type and direction predicted by the control unit 105 as described later.

  Next, the template image data table stored in the template database 106 shown in FIG. 2 will be described. FIG. 3 shows an example of a data table of template image data corresponding to the vehicle type and direction of another vehicle stored in the template database 106. First, as shown in FIG. 3A, image data obtained by viewing the identified other vehicle 120 from eight viewpoints at an equal angle is considered. At this time, the front of the vehicle is set to 0 °, and eight viewpoints are provided 45 ° in a clockwise direction when viewed from above. For example, when the other vehicle is a normal passenger car, template images viewed from eight viewpoints from 0 ° to 315 ° are defined as shown in the table 121 of FIG. This template image schematically represents the shape of the other vehicle that is visible in front of the host vehicle according to the direction of the other vehicle. For example, the length and width of the vehicle for identifying a more detailed vehicle type, glass Data such as the position and the position of the mirror or lamp may be stored simultaneously. Similarly, template image data from eight viewpoints for various types of vehicles such as trucks, buses, and special vehicles is stored in the table 121 to form a template image data table.

  Next, FIG. 4 shows a flowchart of a driving support operation routine executed by the control unit 105 of the control device 101 shown in FIG. Here, the driving support operation routine is an operation executed when the host vehicle travels using the navigation device 102, and the host vehicle is stopped, for example, stopped on the road shoulder or waiting for a signal at an intersection or the like. Does not work during. In the driving support operation routine, the control unit 105 first detects the presence or absence of an object in the visual field range in front of the host vehicle using the radar sensor 103 (step S1). Subsequently, it is determined whether or not the radar sensor 103 has detected an object within the visual field range (step S2). If no object is detected in the field of view in step S2, the routine is terminated as it is, assuming that there is no object that can be an obstacle such as another vehicle ahead. On the other hand, when an object is detected within the visual field range in step S2, the control unit 105 outputs a command for photographing the visual field range of the radar sensor 103 by the in-vehicle camera 104 and captures image data (step S3). Next, the number of objects in the field of view, that is, the number of other vehicles is specified from the captured image data (step S4). At this time, for example, the other vehicle can be specified by using a method in which, for example, a region that can be identified as a vehicle based on the color or shading of captured image data is set as one vehicle. Subsequently, the control unit 105 determines whether or not the number of other vehicles specified in step S4 is one (step S5). When it is determined in step S5 that the number of other vehicles in the visual field range is one, the process proceeds to a vehicle type / direction determination subroutine for further determining the vehicle type and direction of the determined other vehicle (step S6). This vehicle type / direction discrimination subroutine is a subroutine for determining the relative position of the other vehicle with respect to the host vehicle, the vehicle type and direction of the other vehicle, and the operation thereof will be described later. When the operation of the subroutine is finished in step S6, the other vehicle information specified in the subroutine is temporarily stored in the memory 107 (step S7). Here, the other vehicle information is information including the relative position, the vehicle type, and the direction of the other vehicle specified in the above-described vehicle type / direction determination routine, as well as the position of the own vehicle from the navigation device 102 and the road condition ahead. Yes, drawing data and audio data are generated by the control unit 105 and stored in the memory. Further, in the other vehicle information, for example, when the control unit 105 can execute the above-described operation while traveling at a short sampling time, the relative position of the other vehicle moved within the time and the current speed of the host vehicle. It is also possible to estimate and add the traveling speed of other vehicles. Subsequently, the control unit 105 sends drawing data of the other vehicle information to the navigation controller 108 of the navigation device 102 and issues an output command to display it on the image display unit 110 (step S8). Next, the voice data of the other vehicle information is sent to the navigation controller 108 of the navigation device 102 and a command is issued to output from the voice output unit 111 (step S9), and the routine is terminated.

  On the other hand, when it is determined in step S5 that the number of other vehicles in the visual field range is plural, the process proceeds to a vehicle type / direction determination subroutine for further determining the vehicle type and direction of the determined other vehicle (step S10). This vehicle type / direction discrimination subroutine is the same as that executed in step S6, and its operation will be described later. When the operation of the subroutine is finished in step S10, the other vehicle information specified in the subroutine is temporarily stored in the memory 107 (step S11). The other vehicle information stored at this time is information on one of the detected other vehicles. Subsequently, the control unit 105 determines whether or not all other vehicle information has been specified for the number of detected other vehicles, that is, whether or not the determination has been completed for all detected objects (step S12). If it is determined in step S12 that all the determinations have not been completed, the process returns to step S10 to repeat the specification of other vehicle information for the remaining other vehicles. At this time, the other vehicle information stored in step S11 is added and updated every time the number of other vehicles increases. If it is determined in step S12 that all determinations have been completed, the control unit 105 moves to step S8 and sends drawing data of the other vehicle information to the navigation controller 108 of the navigation device 102 and displays it on the image display unit 110. Next, the voice data of the other vehicle information is sent to the navigation controller 108 of the navigation device 102 and issued from the voice output unit 111 (step S9), and the routine is terminated. To do. At this time, in step S8 and step S9, the control unit 105 preferentially displays and outputs, for example, the vehicle of the other vehicle closest to the host vehicle among the other vehicle information including data of a plurality of other vehicles, or the host vehicle. It is also possible to select, for example, to preferentially display the output of another vehicle approaching the vehicle.

  Next, FIG. 5 shows a flowchart of a vehicle type / direction determination subroutine executed in the driving support operation routine shown in FIG. In the vehicle type / direction discrimination subroutine, the control unit 105 first performs image processing by extracting only one area from the image of the other vehicle captured and specified in steps S3 and S4 of the driving support operation routine. Data is converted (step SS1). This image processing includes, for example, screen division processing and edge enhancement processing. Next, the control part 105 reads the relative position from the own vehicle of the other vehicle made into image data from the radar sensor 103 (step SS2). Subsequently, the position information of the own vehicle is read from the map database 109 of the navigation device 102 (step SS3), and the direction of the other vehicle is predicted and temporarily set (step SS4). Here, the position information of the host vehicle read from the map database 109 includes road information in front of the host vehicle as well as the current position of the host vehicle. Therefore, in step SS4, the position of the other vehicle on the map is identified based on the position information of the own vehicle and the road information ahead and the relative position of the other vehicle obtained in step SS2, and the other vehicle thus identified is identified. It becomes possible to narrow down and predict the direction of the other vehicle based on the position of the vehicle and the road information ahead.

  Subsequently, the control unit 105 temporarily sets the vehicle type of the other vehicle (step SS5). For the temporary setting at this time, for example, a process of weighting a normal passenger car that is expected to appear frequently based on data on the number of types of cars in Japan may be added. Next, the control unit 105 reads template data corresponding to the vehicle type and direction temporarily set in Step SS4 and Step SS5 from the template database 106 (Step SS6), and calculates a matching rate with the image data of other vehicles (Step S6). SS7). The matching rate is a comparison between the image data schematically converted into data in step SS1 and the template data read in step SS6, and the degree of matching of the shape without considering the size. Is calculated as follows. Next, it is determined whether or not the calculated match rate is equal to or greater than a predetermined value (step SS8). Here, as an example, the predetermined value is the above-described area overlap rate of 80%. If the match rate is greater than or equal to the predetermined value in step SS8, the model data and direction of the template data currently being compared are set as image data, that is, the model and direction of the other vehicle currently being identified (step SS9), and a subroutine Exit.

  On the other hand, when it is determined in step SS8 that the match rate is smaller than the predetermined value, the control unit 105 adjoins the template data read in step SS6 on the table shown in FIG. 3B. Data is read from the template database 106 (step SS10), and a match rate with the image data of the other vehicle is calculated (step SS11). At this time, the adjacent template data corresponds to data having an angle of 90 ° when the data read in step SS6 has an angle of 45 °, for example. Next, it is determined whether or not the calculated match rate is equal to or greater than a predetermined value (step SS12). At this time, the predetermined value is the same as that used in the determination in step SS8. If the match rate is greater than or equal to the predetermined value in step SS12, the process proceeds to step SS9, where the model type and direction of the template data currently being compared are set as image data, that is, the type and direction of the other vehicle currently being identified. The subroutine is terminated. When it is determined in step SS12 that the match rate is smaller than the predetermined value, the control unit 105 further determines whether or not all angle data has been determined for the template data of the temporarily set vehicle type (step SS13). If it is determined in step SS13 that all angle data has not been determined, the process returns to step SS10 and the subsequent operations are repeated. On the other hand, if it is further determined that all angle data has been determined in step SS13, it is determined that the other vehicle of the image data is not of the temporarily set vehicle type, and the determination has already been completed for all vehicle types. It is further determined whether or not (step SS14). If it is determined in step SS14 that determination has not been completed for all the vehicle types, the process returns to step SS5, and the subsequent steps are repeated from the temporary setting of the vehicle types of other vehicles. On the other hand, if it is further determined in step SS14 that the determination has been completed for all the vehicle types, the image data of the other vehicle to be identified this time does not match any vehicle type stored in the template database. The data is added to the other vehicle information to the effect that the vehicle is an unverified vehicle and a warning is issued, and the routine is terminated.

  In the operation from step SS10 to step SS12, if the match rate in step SS12 exceeds 50% but does not exceed a predetermined value, for example, the template data and the current angle at the angle determined in step SS8 before that The intermediate template data may be interpolated from the template data in step (b) to newly calculate the match rate and determine whether or not the value is equal to or greater than a predetermined value.

  By using the operation routine as described above, the driving support system 100 according to the first embodiment allows the position information of the host vehicle from the map database 109 included in the navigation device 102 and the other vehicle from the radar sensor 103 and the in-vehicle camera 104. Since the vehicle type and direction are determined after predicting the position and direction of the other vehicle on the map based on the relative position and the image data, it is possible to efficiently detect the vehicle type and the traveling direction of the other vehicle. It becomes possible.

  Next, a first application example of the driving support system 100 according to the first embodiment will be described. FIG. 6 is a diagram schematically illustrating a case where the host vehicle enters a left curve as a first application example of the driving support system 100. As shown in FIG. 6A, when the other vehicle 131 is traveling ahead when the host vehicle 130 enters the left curve 140, the other vehicle 131 is within the range of the field of view 130a of the radar sensor. When entering, the driving support system 100 detects the presence of the other vehicle 131. At this time, the control unit 105 of the driving support system 100 executes the driving support operation routine shown in FIG. 4 and the vehicle type / direction determination subroutine shown in FIG. 5 to predict the vehicle type and direction of the other vehicle 131. . That is, in the vehicle type / direction determination subroutine, when the vehicle type of the other vehicle 131 is temporarily set as a normal vehicle, the vehicle of the other vehicle 131 detected from the navigation device 102 and information such as the position of the vehicle 130 and the road conditions ahead. Based on the relative position from 130, the position of the other vehicle 131 on the map is specified. As a result, since it can be determined that the other vehicle 131 is positioned on the left lane side of the left curve, it can be predicted that the other vehicle 131 is traveling along the road along the left curve. That is, it can be assumed that the other vehicle 131 appears in the image captured by the in-vehicle camera 104 of the host vehicle 130 so as to substantially match the shape and orientation of the template data as shown in FIG. At this time, when the matching rate between the image data of the other vehicle actually captured and the template data is equal to or greater than a predetermined value, the control unit 105 is the normal vehicle 131 and is directed to the direction shown in FIG. The other vehicle information is output to the navigation controller 108 of the navigation device 102. Based on the other vehicle information and the output command from the control unit 105, the navigation device 102 draws the image display unit 110 indicating the current running state and the guidance data indicating that the other vehicle 131 exists ahead of the driver. 150 is displayed. In addition, the navigation device 102 simultaneously outputs voice information corresponding to the guidance data 150 from the voice output unit 111 to the driver.

  Next, a second application example of the driving support system 100 according to the first embodiment will be described. FIG. 7 is a diagram schematically showing a case where the host vehicle is about to enter an intersection as a second application example of the driving support system 100. As shown in FIG. 7A, when two other vehicles 131a and 131b are traveling ahead when the host vehicle 130 is about to enter the intersection 141, the other vehicles 131a and 131b are viewed by the radar sensor. When entering the range of 130a, the driving support system 100 detects the presence of the other vehicles 131a and 131b. At this time, the control unit 105 of the driving support system 100 executes the driving support operation routine shown in FIG. 4 and the vehicle type / direction determination subroutine shown in FIG. 5 to determine the vehicle types and directions of the other vehicles 131a and 131b. Predict. That is, the prediction operation is first executed for the other vehicle 131a, and then the prediction operation for the other vehicle 131b is executed. As in the case of the first application example, in the vehicle type / direction determination subroutine, first, if the vehicle type of the other vehicle 131a is temporarily set as a normal vehicle, information such as the position of the host vehicle 130 and the road condition ahead of the vehicle from the navigation device 102 is obtained. Based on the detected relative position of the other vehicle 131a from the host vehicle 130, the position of the other vehicle 131a on the map is specified. As a result, since it can be determined that the other vehicle 131a is located in the left lane of the left road of the intersection 141, it can be predicted that the other vehicle 131a is traveling rightward on the left road of the intersection 141. In other words, it is assumed that the other vehicle 131a appears in the image taken by the vehicle-mounted camera 104 of the host vehicle 130 as if it were template data of 90 ° to the right of the ordinary vehicle shown in FIG. However, as shown in FIG. 7 (a), the other vehicle 131a that is actually traveling is a truck, and therefore does not match any template data when temporarily set as a normal vehicle. Therefore, the control unit 105 newly provisionally sets the other vehicle 131a as a truck and executes the same determination operation. As a result, it can be assumed that the image data of the other vehicle 131a has the shape and orientation of the template data as shown in FIG. At this time, if the matching rate between the image data of the other vehicle actually captured and the template data is equal to or greater than a predetermined value, the control unit 105 faces the other vehicle 131a in the direction shown in FIG. Judge that Similarly, the operation of the vehicle type / direction determination subroutine is repeated for the other vehicle 131b, the vehicle type and direction with respect to the other vehicle 131b are specified, and the other vehicle information is updated as data of a plurality of other vehicles. Subsequently, the control unit 105 outputs other vehicle information to the navigation controller 108 of the navigation device 102. On the basis of the other vehicle information and the output command from the control unit 105, the navigation device 102 draws the image display unit 110 on the image data indicating the current traveling state and the other vehicle 131a is about to enter the intersection ahead of the driver from the left side. The guidance data 151 indicating that the user is present is displayed. In addition, the navigation device 102 simultaneously outputs voice information corresponding to the guidance data 151 from the voice output unit 111 to the driver. At this time, the driving support system 100 has other vehicle information about the two other vehicles, but the other vehicle 131b has already entered the intersection 141 and is passing as shown in FIG. As in this application example, it is also possible to preferentially select and present to the driver the information on the other vehicle 131a that is expected to correspond to the host vehicle 130 entering the intersection 141 thereafter.

  With the configuration and operation as described above, the driving support system 100 according to the present invention can detect the position information of the host vehicle from the map database 109 included in the navigation device 102 and the relative position of the other vehicle from the radar sensor 103 and the in-vehicle camera 104. The vehicle type and direction of the other vehicle are efficiently detected and specified in order to determine the vehicle type and direction after predicting the position and direction of the other vehicle on the map based on the position and image data. Appropriate driving support information can be presented to the driver based on the other vehicle information.

  From the above embodiments, according to the present invention, the traveling direction of the other vehicle in front of the host vehicle is predicted based on the position information by the navigation device and the relative position of the other vehicle by the radar sensor, and is actually captured by the in-vehicle camera. Since the control device that compares the image data of the other vehicle extracted from the forward image with the predicted vehicle type and direction data of the other vehicle is used, the surrounding vehicle that can efficiently detect the vehicle type and the traveling direction of the other vehicle An identification method can be provided. It is also possible to provide a driving support system that presents driving support information to the driver based on the information obtained by the above-described surrounding vehicle identification method.

  The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to such specific embodiments, and various modifications can be made within the scope of the gist of the present invention described in the claims.・ Change is possible. The configuration of the driving support system 100 illustrated in FIG. 2 and the template data table illustrated in FIG. 3B are examples, and the present invention is not limited to this configuration.

It is a block diagram which shows the structure of the control apparatus for vehicles in a prior art. 1 is a block diagram illustrating a configuration of a driving support system 100 according to a first embodiment. It is an example of the data table of the template image data according to the vehicle type and direction of the other vehicle memorize | stored in the template database 106 shown by FIG. 3 is a flowchart of a driving support operation routine executed by a control unit 105 of the control device 101 shown in FIG. FIG. 5 is a flowchart of a vehicle type / direction determination subroutine executed in the driving support operation routine shown in FIG. 4. FIG. 1 is a diagram schematically illustrating a first application example of a driving support system 100. FIG. It is the figure which showed the 2nd example of application of the driving assistance system 100 typically.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Driving support system 101 Control apparatus 102 Navigation apparatus 103 Radar sensor 104 Car-mounted camera 105 Control part 106 Template database 108 Navigation controller 110 Image display part 130 Own vehicle 131, 131a, 131b Other vehicle

Claims (4)

A detection unit for detecting an object existing around the host vehicle;
An image data generation unit that images the object and generates image data;
A calculation unit for calculating a relative position of the object with respect to the host vehicle;
Based on the current position of the host vehicle on the map data and the relative position, the direction of the other vehicle is predicted by determining in which lane the other vehicle is positioned with the object as the other vehicle. A prediction unit to
Based on the image data and the template data corresponding to the predicted orientation among a plurality of template data indicating the shape including the position of the glass and the position of the mirror as seen from different directions for one vehicle type And
A determination unit for determining the vehicle type and direction of the other vehicle;
And an output unit that outputs information including the determined vehicle type and direction of the other vehicle.   2. The driving support system according to claim 1, wherein the determination unit refers to a plurality of template data among the template data and also refers to a template data generated by interpolating the plurality of template data.   The prediction unit predicts a direction of each other vehicle with respect to each other vehicle when a plurality of other vehicles exist around the own vehicle. The driving support system according to one item. Detecting an object existing around the vehicle;
Imaging the object to generate image data;
Calculating a relative position of the object with respect to the host vehicle;
Based on the current position of the host vehicle on the map data and the relative position, the direction of the other vehicle is predicted by determining in which lane the other vehicle is positioned with the object as the other vehicle. And steps to
Based on the image data and the template data corresponding to the predicted orientation among a plurality of template data indicating the shape including the position of the glass and the position of the mirror as seen from different directions for one vehicle type Determining the vehicle type and direction of the other vehicle;
The surrounding vehicle identification method characterized by having.

Images