JP2018114894A - Automatic driving system for vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic driving system for a vehicle suitable to perform a pedestrian-friendly safety driving.SOLUTION: Image processing means comprises: a detection function for detecting, by image processing, presence of people in an image captured by a camera, presence of a traffic light and an illuminated color of the traffic light, as object information at an intersection; and a detection function for detecting a shape of the intersection in an image of a map around a current position, as feature information at the intersection. Control means comprises: a learning mode for learning automatic driving patterns on the basis of the object information at the intersection and behavior patterns of the vehicle at detection time of the information; and an automatic driving mode for outputting control signals required for automatically driving the vehicle on the basis of the automatic driving patters acquired through the learning and the feature information at the intersection.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a system for substituting driving of a car by a computer (hereinafter referred to as “automatic driving system”), and in particular, enables a self-driving of a safe car friendly to a pedestrian.

  2. Description of the Related Art In recent years, technical development of a system (automatic driving system) that substitutes a computer for driving a car has been actively promoted. Among these technical developments, for example, in Patent Document 1, the front of the traveling direction is photographed with a camera mounted on an automobile, and a road lane (specifically, a white line) is recognized from a photographed image by the camera. Discloses driving assistance and automatic driving of an automobile (see the summary of Patent Document 1).

  However, for example, there may be a pedestrian trying to cross the road at the intersection of the road. For this reason, there is a problem that it is difficult to perform automatic driving of a safe car simply by recognizing a road lane as in the technique disclosed in Patent Document 1.

JP 2000-207563 A

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an automatic vehicle driving system suitable for self-driving a safe vehicle friendly to pedestrians.

  In order to achieve the above object, the present invention processes a camera that captures the periphery of a car, map acquisition means that acquires a map around the current location of the vehicle, and an image captured by the camera and the map around the current location. Image processing means, and control means for controlling the automobile based on the processing result of the image processing means, wherein the image processing means performs image processing of an image taken by the camera as intersection object information. The function of detecting the presence or absence of a human shadow, the presence or absence of a traffic light and the color of the traffic light, and the function of detecting the shape of an intersection from the current location surrounding map image as the intersection feature information, the control means, A learning mode for learning an automatic driving pattern based on intersection object information and the behavior pattern of the vehicle at the time of detecting the intersection object information; Characterized in that it includes a an automatic operation mode for outputting the control information S1 required for automatic operation with respect to control system devices such 41 based on the automatic operation pattern and intersection feature information obtained by learning.

  In the present invention, the learning of the automatic driving pattern is a process of storing a plurality of sets of the intersection object information and the behavior pattern of the car as a set of learning basic data in a data storage unit, and stored in the data storage unit. And a process of extracting learning basic data having the same content from a plurality of sets of learning basic data and obtaining an automatic driving pattern based on the learning basic data having the largest number of extractions.

  In the present invention, the camera may be capable of photographing the front, the left and right sides, and the rear of the automobile by being provided so as to be able to rotate 360 ° around the mounting position.

  In the present invention, a plurality of cameras may be provided corresponding to the respective directions as means for photographing the four directions of front, rear, left and right of the automobile.

  In the present invention, the image captured by the camera may include an image of a door mirror of the automobile.

  In the present invention, the automatic driving pattern obtained by the learning is transmitted to an automatic driving monitoring server via an Internet communication network, and the automatic driving monitoring server stores the automatic driving pattern, the database, An information processing function for processing the automatic driving pattern stored in the database as information, and the information processing function has the same intersection feature information from the automatic driving patterns stored in the database. Extracting the automatic driving pattern, grouping the extracted automatic driving pattern as one intersection common group, and referring to one intersection common group obtained by the grouping process, and within the referenced intersection common group The automatic operation pattern with the same content is extracted from the A process of employing a rolling pattern as a representative automatic driving pattern in the one crossing common group, may be characterized by including.

  In the present invention, as a specific configuration of the automobile automatic driving system, as described above, intersection object information such as the presence or absence of a human figure, the presence or absence of a traffic light, the color of a traffic light, and the like of the automobile at the time when this is detected. Based on the behavior pattern, learn the automatic driving pattern, and output the control signal necessary for the automatic driving of the car based on the automatic driving pattern obtained by learning and the intersection feature information (specifically, the shape of the intersection) The configuration is adopted. For this reason, for example, when a pedestrian appears at an intersection during the execution of a learning mode for learning an automatic driving pattern, an automatic driving pattern corresponding to this situation is learned. An automatic vehicle driving system suitable for driving can be provided.

FIG. 1 is a block diagram of an automatic vehicle driving system to which the present invention is applied. FIG. 2 is a configuration diagram of the camera in the automatic driving system of FIG. FIG. 3A is a map around the current location used in the automatic driving system of the vehicle (an example of a crossroad intersection), FIG. 3B is a photographed image (example of a crossroad intersection) used in the system, and FIG. ) Is the current location map image (an example of a three-way intersection), and FIG. 3D is an explanatory diagram of the current location map image (an example of a multi-difference intersection). 4 is an explanatory diagram of a data storage unit in the automobile automatic driving system of FIG. FIG. 5 is an explanatory diagram of a flow of acquiring data used in the program when the learning mode in the automobile automatic driving system of FIG. 1 is configured as software processing by the program. FIG. 6 is a flowchart of a main processing program for realizing a learning mode. FIG. 7 is a flowchart of a sub processing program (data collection processing program) used in the main processing program. FIG. 8 is an explanatory diagram of an information network in the automatic driving system of FIG.

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

  FIG. 1 is a block diagram of an automatic vehicle driving system to which the present invention is applied, FIG. 2 is a configuration diagram of a camera in the automatic driving system of FIG. 1, and FIG. 3B is a map image (an example of a crossroad intersection), FIG. 3B is an image taken by a camera used in the system (an example of a crossroad intersection), FIG. 3C is a map image of the current location surroundings (an example of a three-way intersection), and FIG. (D) is explanatory drawing of the present location periphery map image (example of a multi-difference road intersection).

《Outline of Autonomous Driving System》
The automatic vehicle driving system S of FIG. 1 includes a camera 1 of FIG. 2 that captures the surroundings of a vehicle, a map acquisition unit 2 that acquires a current location map image D1 of the vehicle, an in-vehicle computer 3 that is mounted on the vehicle, It is comprised including various vehicle equipment 4.

  The in-vehicle computer 3 includes a CPU 31, an input / output interface 32, a data storage unit 33 including an SSD, a POM, or a RAM, and hardware such as a communication module 34. 2 functions as image processing means 5 for processing the current location surrounding map image D1 acquired in 2 and control means 6 for controlling the automobile in cooperation with various in-vehicle devices 4.

《Outline of various in-vehicle devices》
The various in-vehicle devices 4 can be broadly classified into a control system device 41 for controlling the vehicle and a measurement device 42 for measuring the state of the vehicle. The control system device 41 includes a known engine control device 411, The brake control device 412, the steering control device 413, and the like are provided, and the measuring device 42 includes a known travel speed measuring device 421, a steering angle measuring device 422, and the like.

  The engine control device 411 receives the control information S1 from the in-vehicle computer 3, and controls the automobile engine based on the received control information S1. Similarly, the brake control device 412 and the steering control device 413 receive the control information S1, and control the vehicle brake device and the vehicle steering device based on the received control information S1.

  The data measured by the measuring instruments 42, that is, the traveling speed of the vehicle measured by the traveling speed measuring instrument 421, the steering angle measured by the steering angle measuring instrument 422, etc., are time-sequentially used as the in-vehicle computer. 3 is output.

<Camera configuration>
As a specific configuration of the camera 1, in the automatic vehicle driving system S of FIG. 1, the camera 1 is attached to the roof of the vehicle as shown in FIG. However, the present invention is not limited to such a camera configuration. Other camera configurations will be described later.

<Description of the configuration of the map acquisition means>
As a specific configuration of the map acquisition unit 2, the automatic vehicle driving system S of FIG. 1 is configured to acquire a current vehicle location map image D <b> 1 from a known car navigation device. Without being limited, the current position surrounding map image D1 may be acquired by other known methods.

<< Configuration of image processing means and control means >>
In order for the in-vehicle computer 3 to specifically function as the image processing means 5, the in-vehicle computer 3 has a captured image D1 output from the camera 1 and a map acquisition means 2 (car navigation device) via the input / output interface 32. The present location surrounding map image D2 output from is input.

  In order to make the in-vehicle computer 3 function specifically as the control means 6, the in-vehicle computer 1 sends control information S <b> 1 (for example, engine control) to the control system devices 41 of various in-vehicle devices 4 via the input / output interface 32. Signal, brake control signal, steering control signal, etc.).

  The image processing means 5 performs a function of detecting the presence / absence of a human shadow, the presence / absence of a traffic light and the color of the traffic light from the photographed image D1 taken by the camera 1 as the intersection object information, and the current location map as the intersection feature information. A function of detecting the shape of an intersection (see FIGS. 3A, 3C, or 3D) from the image D2 is provided.

  In the image processing means 5, image processing for detecting the presence or absence of a human shadow from the photographed image D 1 by the camera 1 is performed by, for example, registering a human (human) reference image in a predetermined area of the data storage unit 33 in advance. A method of checking whether or not an image that matches the registered reference image is included in the captured image D1 may be adopted. The same applies to image processing for detecting the presence / absence of a traffic light and the color of the traffic light from the photographed image D1.

  In the image processing means 5, as the image processing for detecting the shape of the intersection from the current location surrounding map image D2, for example, a known contour extraction processing or contour enhancement processing is performed on the entire current location surrounding map image D2. By performing the processing so that the contour shape of the intersection included in the current location surrounding map image D2 is raised, the shape of the intersection may be detected.

  The image processing described above can be realized as software processing by a program. In this case, the program is stored in a predetermined area of the data storage unit 33 of the in-vehicle computer 3 and is executed by the CPU 31 of the in-vehicle computer 3. Can be configured.

  FIG. 4 is an explanatory diagram of a data storage unit in the automobile automatic driving system of FIG.

  Referring to FIG. 4, the data storage unit 33 of the in-vehicle computer 3 has a plurality of data storage areas A1 and A2 in addition to a predetermined area for storing a human (human) reference image and a program. In the data storage areas A1 and A2, the presence / absence of a human shadow detected by the image processing means 5 as the first data PM1 (intersection object information) and the presence / absence of the traffic light detected by the image processing means 5 as the second data PM2 (intersection) Object information) and the traffic light color (intersection object information) detected by the image processing means 5 are stored as third data PM3, and in addition to this, fourth data PM4 is stored in a main processing program described later. The behavior pattern of the automobile specified by the determination process and the shape of the intersection (intersection feature information) detected by the image processing means 5 are stored as the fifth data PM5. Note that the behavior pattern of the car stored as the fourth data PM4 and the shape of the intersection (intersection feature information) stored as the fifth data PM5 are the intersection object information such as the presence / absence of a human shadow by the image processing means 5. It corresponds to the time of detection.

  FIG. 5 is an explanatory diagram of a flow of acquiring data used in the program when the learning mode in the automobile automatic driving system of FIG. 1 is configured as software processing by the program.

Referring to FIG. 5, for example, immediately before an intersection, the image processing means 5 detects the shape of the intersection and a traffic light, detects red as the color of the traffic light, and further detects a human figure. In the example, the first data PM1 is “with a traffic light”, the second data PM2 is “red”, the third data PM3 is “with a human figure”,
The fourth data PM4 is information (value) “stop”, and the fifth data PM5 is information specifying the shape of the intersection.

  In addition to “stop”, “speed maintenance”, “deceleration / deceleration”, and “departure” can be adopted as the behavior pattern of the automobile, but the behavior pattern is not limited to these.

  There are various methods for identifying the behavior pattern of an automobile. For example, as described above, of the measurement information D3 (information such as travel speed and rudder angle) output from the measurement device 42 to the in-vehicle computer 3, what kind of vehicle is immediately before the intersection from the travel speed measurement information. The behavior pattern of the car may be specified whether the behavior is taken.

  The control means 5 is the intersection object information (specifically, the first to third data PM1 to PM3 described above) and the behavior pattern of the automobile at the time of detecting this (specifically, described above). A learning mode for learning an automatic driving pattern based on the fourth data PM4), an automatic driving pattern obtained by the learning, and intersection feature information (specifically, the fifth data PM5 described above); And an automatic operation mode in which control information S1 necessary for automatic operation is output to the control system devices 41.

  As described above, the learning of the automatic driving pattern in the learning mode is a process of storing the first to fifth data PM1 to PM5 in the data storage area of the data storage unit 33, that is, (1) intersection object information and car action pattern. And a process of storing a plurality of sets of intersection feature information as a set of learning basic data in the data storage unit 33, and (2) the same among a plurality of sets of learning basic data stored in the data recording unit 33 as described above A process of extracting learning basic data of contents and obtaining an automatic driving pattern based on the learning basic data having the largest number of extractions.

《Detailed explanation of learning mode》
The learning mode can be configured as software processing by a program executed by the CPU of the in-vehicle computer.

  FIG. 6 is a flowchart of a main processing program for realizing the learning mode, and FIG. 7 is a flowchart of a sub processing program (data collection processing program) used in the main processing program.

  The main processing program in FIG. 6 is stored in the data storage unit 33 of the in-vehicle computer 3 in FIG. 1, starts when a learning mode selection button (not shown) is pressed, and is executed by the CPU 31 of the in-vehicle computer 1. In the learning mode, the driver actually drives the vehicle with the handle.

  In the main processing program of FIG. 6, as the initial processing immediately after the start, the data storage area of the data storage unit 33 for storing the first to fifth data PM1 to PM5 and the data collection counter are cleared (ST1). It is determined whether or not it is immediately before the intersection (ST2).

  For example, the map acquisition means 2 (car navigation device) is configured to output the vehicle approach information to the in-vehicle computer 3, and the output is used as a trigger immediately before the intersection. You may determine that there is.

  If it is determined in ST2 that it is not immediately before the intersection, the process returns to ST1 to determine whether it is immediately before the intersection (No in ST2). On the other hand, if it is determined in ST2 that it is immediately before the intersection, the process proceeds to the sub-processing program of FIG. 7 and data collection processing is performed (ST3).

  Referring to FIG. 7, in the data collection process, first, first to fifth data PM1 to PM5 are acquired (ST31). The acquired data PM1 to PM5 are stored in the first data storage area A1 of the data storage unit 33 (see, for example, the example of data No. 1 shown in FIG. 4).

  Here, for example, when a pedestrian crosses an intersection without a traffic light and the car stops immediately before the intersection, the image processing means 5 detects “no traffic light” and “with a human shadow”. , “No signal” as the first data PM1, “null value” with no value as the second data PM2, “with a human figure” as the third data PM3, “stop” as the fourth data PM4, and Then, information specifying the shape of the intersection is acquired and stored as the fifth data PM5 (ST31).

  Next, it is determined whether or not the fourth data PM4 acquired and stored as described above, that is, the behavior pattern of the automobile is “stop” (ST32).

  If it is determined that the operation is not stopped, the process proceeds to ST4 of the main processing program (No in ST32, ST33). On the other hand, if it is determined that the vehicle is stopped, the process returns to ST31 in order to identify the behavior of the stopped vehicle and the behavior pattern of the vehicle (Yes in ST32). The process of ST32 is repeated.

  As a result of repeatedly executing the processes of ST31 and ST32 as described above, if it is finally determined that the process is not stopped, the first to the first acquired in the process of ST31 immediately before the final determination are obtained. 4 data PM4 is stored in the second data storage area A2 of the data storage unit 33 as additional data, and then the process proceeds to ST4 of the main processing program (No in ST32, ST33).

  Thereafter, the main processing program counts up the count value of the data collection counter (ST4), and determines whether or not the count value has reached a predetermined value n (for example, n = 1000) (ST5).

  Here, when the count value is not the predetermined value n, the process returns to ST2 to repeatedly acquire the first to fifth data PM1 to PM5, and the processes from ST2 to ST5 are repeatedly executed. On the other hand, when the count value reaches the predetermined value n, a learning summary process in ST6 described later is performed, and then the process returns to ST2.

  By repeating the processing from ST2 to ST5, the data storage areas A1 and A2 of the data storage unit 33 have the intersection object information (first to third data PM1 to PM3) and the intersection object information detected at the time of detection. A plurality of sets of learning basic data composed of automobile behavior patterns (fourth data PM4) are stored as shown in FIG.

  In the example of FIG. 8, since the predetermined value n of the above-described data collection counter is set to “n = 4”, the data NO. There are four sets of basic learning data from 1 to 4.

  As described above, the learning summary process in ST6 includes a plurality of sets of learning basic data stored in the data storage areas A1 and A2 of the data storage unit 33 (in the example of FIG. 8, four sets of learning basics from data Nos. 1 to 4). This is a process of extracting learning basic data having the same content from the data and finally learning an automatic driving pattern based on the learning basic data having the largest number of extractions.

  In the example of FIG. 8, the data No. 1 stored in the first data storage area A1 of the data storage unit 33 is displayed. 1 to 4 of data NO. The learning basic data of 1, 2 and 4 have the same contents, that is, there are three learning basic data with the same contents. The learning basic data of 1, 2 and 4 are extracted as learning basic data having the same content. In association with this, the data No. 2 stored in the second data storage area A2 of the data storage unit 33 is stored. The basic learning data of 1, 2 and 4 are also extracted together. At this time, the number of extracted learning basic data having the same content is 3, which is the maximum, so in the example of FIG. A predetermined automatic driving pattern is learned based on the learning basic data of 1, 2, and 4.

  The automatic driving pattern obtained by learning in the example of FIG. Since it is based on the basic learning data of 1, 2, and 4, it is a series of operations in which a car is stopped when there is a person at an intersection without a signal and a car is started when there is no person.

  In addition, intersection feature information (specifically, fifth data PM5 included in data Nos. 1, 2 and 4) is added as additional information to the automatic driving pattern obtained by the learning described above. The For example, if the automatic driving pattern obtained by learning in the example of FIG. 8 is based on the learning basic data obtained immediately before the intersection shown in FIG. 3, the information specifying the shape of the intersection in FIG. Data PM5) is added as intersection feature information.

  The automatic driving pattern obtained by the learning described above is stored in a predetermined area of the data storage unit 33 of the in-vehicle computer 3 and is referred to in the automatic driving mode.

<Detailed explanation of automatic operation mode>
The automatic operation mode can also be configured as software processing by a program executed by the CPU 31 of the in-vehicle computer 3.

  In addition to the shape of the intersection, the image processing means 5 described above can detect the shape of the road connected to the intersection from the current location map image by image processing. Control information S1 necessary for automatic driving along the road shape detected by the image processing means 5 is output from the in-vehicle computer 3 to the control system devices 41 of the automobile via the input / output interface 32.

  Also, in the automatic driving mode, as in the learning mode described above, it is determined whether or not it is immediately before the intersection, and when it is determined that it is immediately before the intersection, the automatic driving pattern learned as described above is referred to. Then, the control information S1 necessary for performing the automatic driving in accordance with the referred automatic driving pattern is output from the in-vehicle computer 1 to the control system devices 41 of the automobile via the input / output interface 32.

  Further, in the automatic driving mode, when the vehicle enters a right turn or left turn at the intersection, the automatic driving is performed along the intersection feature information detected by the image processing means 5 immediately before the intersection, that is, the shape of the intersection. The control information S1 necessary for the control is also output from the in-vehicle computer 3 to the control system devices 41 of the automobile via the input / output interface 32.

  By the way, the image processing means 5 may be configured to detect an oncoming vehicle at an intersection from an image captured by the camera 1 by image processing. In the automatic driving mode when such a configuration is adopted, for example, when trying to make a right turn at an intersection, it is not possible to detect a human figure by the image processing means 5 and no oncoming vehicle can be detected. You may output control signal S1 required in order to perform an automatic driving | running along a shape.

<Description of information network>
FIG. 8 is an explanatory diagram of an information network in the automatic vehicle driving system of FIG.

  Referring to FIG. 9, the automatic driving pattern obtained by the learning is transmitted to the automatic driving monitoring server 101 via the Internet communication network 100. At this time, since the intersection feature information (intersection shape) is added to the automatic driving pattern as described above, the intersection feature information is also transmitted to the automatic driving monitoring server 100 as additional information of the automatic driving pattern. Become.

  As a connection method to the Internet communication network, the communication module 34 provided in the in-vehicle computer 3 may be used to connect directly to the Internet communication network 100, or the communication module 34 may connect to the Internet communication network 100 via a smartphone. You may connect to.

  The automatic driving monitoring server 100 has a database that stores automatic driving patterns (including intersection feature information), and an information processing function that processes the automatic driving patterns stored in the database as information.

  The information processing function of the automatic driving monitoring server 100 extracts the automatic driving pattern having the same intersection feature information from the automatic driving patterns stored in the database, and the extracted automatic driving pattern as one intersection common group Refers to the grouping process and one intersection common group obtained by this grouping process, extracts the automatic driving pattern of the same content from the referenced intersection common group, and the automatic driving with the largest number of extraction And a process of adopting the pattern as a representative automatic driving pattern in the one intersection common group. The process of adopting the representative automatic driving pattern can be performed at a predetermined timing such as periodically.

  The representative automatic driving pattern is corrected, for example, by the automatic driving monitoring server 101 via the Internet communication network 100 to the in-vehicle computer 3 of each car, thereby automatically correcting the automatic driving pattern learned by each car. It may be used as data to be used.

<< About other embodiments >>
As a specific configuration of the camera 1, the automatic vehicle driving system S of FIG. 1 has a configuration in which the camera 1 is attached to the roof of the vehicle as described above and the front of the vehicle in the traveling direction is photographed with the camera 1 attached. In addition to this, in addition to this, it may be configured to be able to photograph directions other than the front (for example, both right and left sides of the automobile and the rear).

  In this case, the camera 1 is attached to the roof of the automobile as shown in FIG. 1 via a rotation drive mechanism (not shown), and is provided so as to be able to rotate 360 ° around the attachment position. You may comprise so that the back can be image | photographed.

  In addition, the camera 1 may employ a configuration in which a plurality (for example, four) of cameras are provided corresponding to each direction as means for photographing the front, rear, left, and right directions of the automobile. In this case, the number of cameras can be appropriately increased or decreased according to the shooting direction.

  Further, the camera 1 may be configured such that an image of an automobile door mirror is included in the captured image depending on the setting of the capturing direction and the mounting position.

S Automobile driving system 1 Camera 2 Map acquisition means 3 In-vehicle computer 31 CPU
32 CPU
33 Data storage unit 34 Communication module 4 Various in-vehicle devices 41 Control system devices 411 Engine control device 412 Brake control device 413 Steering control device 42 Measuring devices 421 Traveling speed measuring device 422 Steering angle measuring device 5 Image processing means 6 Control means 100 Internet communication network 101 Automatic operation monitoring server A1, A2 Data storage area D1 Current location map image D2 Captured image D3 Measurement information S1 Control information

Claims (6)

A camera that captures the area around the car,
Map acquisition means for acquiring a map around the current location of the vehicle;
Image processing means for processing an image captured by the camera and the current location map image;
Control means for controlling the automobile based on a processing result in the image processing means,
The image processing means is configured to detect the presence / absence of a human shadow, the presence / absence of a traffic light, and the color of a traffic light from the image captured by the camera as intersection object information by image processing, and the current location map as intersection feature information It has a function to detect the shape of the intersection from the image,
The control means includes a learning mode for learning an automatic driving pattern based on the intersection object information and the behavior pattern of the automobile at the time of detecting the intersection object information, an automatic driving pattern obtained by the learning, and the intersection feature information, And an automatic driving mode for outputting control information necessary for automatic driving of the vehicle based on the vehicle. The learning of the automatic driving pattern is as follows:
A process of storing a plurality of sets of the intersection object information and the behavior pattern of the automobile as a set of learning basic data in a data storage unit;
Processing for extracting learning basic data having the same content from a plurality of sets of learning basic data stored in the data storage unit and obtaining an automatic driving pattern based on the learning basic data having the largest number of extractions. The automatic vehicle driving system according to claim 1. 3. The camera according to claim 1, wherein the camera can be photographed in front, left and right sides, and rear of the vehicle by being provided so as to be able to rotate 360 ° around the mounting position. The automobile automatic driving system according to item 1. 3. The automatic driving of a vehicle according to claim 1, wherein a plurality of the cameras are provided corresponding to each direction as means for photographing four directions of front, rear, left and right of the vehicle. 4. system. The vehicle automatic driving system according to claim 1, wherein an image of the door mirror of the automobile is included in an image captured by the camera. The automatic driving pattern obtained by the learning is transmitted to the automatic driving monitoring server via the Internet communication network,
The automatic operation monitoring server is
A database for accumulating the automatic driving patterns;
An information processing function for processing the automatic driving pattern stored in the database as information, and
The information processing function is
Extracting the automatic driving pattern having the same intersection feature information from the automatic driving pattern stored in the database, and grouping the extracted automatic driving pattern as one intersection common group,
Refers to one intersection common group obtained by the grouping process, extracts an automatic driving pattern having the same contents from the referenced intersection common group, and selects the automatic driving pattern having the largest number of extractions as the common intersection of the one intersection 6. The automatic vehicle driving system according to claim 1, further comprising: a process employed as a group representative automatic driving pattern.

Images