Calculation device and method for producing database

Abstract

The invention provides an arithmetic device and a database production method. A computing device is provided with: a server communication unit that receives, from an in-vehicle device mounted on a first vehicle, information on an interruption position, which is a position at which automatic driving is interrupted, and that receives, from an in-vehicle device mounted on a second vehicle, manual driving information, which is information on manual driving of the second vehicle in a region of the interruption position; an extraction unit that extracts, from manual driving information in the region of the interruption position, a control point that is a point at which the direction of travel is changed; and a creation unit that creates knowledge information including travel information of the vehicle on a trajectory passing through the control point and capable of generating a trajectory in which the direction of travel of the vehicle is changed. The server communication unit further transmits at least one of the knowledge information and information indicating a trajectory of the vehicle generated using the knowledge information to an on-board device of the vehicle.

Description

[0001] This application was submitted on August 26, 2020. The Chinese patent application number is 202080004840.9 (international application number is PCT / JP2020 / 032218). The invention name is a divisional application of the patent application "Computing device, vehicle-mounted device and autonomous driving system". Technical Field

[0002] The present invention relates to a computing device, a vehicle-mounted device and an automatic driving system. Background Art

[0003] In recent years, autonomous driving technology using map data has been put into practical use. In addition to sensor information, current autonomous driving technology also uses map data to correct the vehicle's position and pre-read the route and road information. This map data requires higher accuracy than the previous map data used for car navigation. However, even if a map with sufficient accuracy is prepared, there are cases where autonomous driving is interrupted due to various reasons. Patent document 1 discloses a vehicle control system, characterized in that it comprises: a data processing device that generates and distributes reference information to be referred to by an autonomous driving vehicle in order to perform autonomous driving; and an autonomous driving vehicle that performs autonomous driving with reference to the reference information obtained from the above-mentioned data processing device; the above-mentioned data processing device comprises: a driving history information acquisition unit that obtains driving history information from multiple vehicles; a reference information generation unit that generates, as the above-mentioned reference information, vector information used to represent the paths traveled by multiple vehicles and attribute information about the paths represented by the above-mentioned vector information to establish corresponding reference information; and a reference information distribution unit that distributes the reference information generated by the above-mentioned reference information generation unit to the above-mentioned autonomous driving vehicle; the above-mentioned autonomous driving vehicle comprises: a reference information acquisition unit that obtains the above-mentioned reference information from the above-mentioned data processing device; and a control unit that performs autonomous driving along the path represented by the above-mentioned reference information.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2018-155894 Summary of the invention

[0007] Problems to be solved by the invention

[0008] The invention described in Patent Document 1 does not adequately address locations where autonomous driving is difficult. For example, since appropriate routes are not considered for locations where autonomous driving is often canceled due to the positional relationship with the road or surrounding vehicles, it may be difficult to continue autonomous driving.

[0009] Means used to solve problems

[0010] A computing device according to a technical solution of the present invention comprises: a server communication unit, which receives information of an interruption position as a position where automatic driving is interrupted from a vehicle-mounted device mounted on a first vehicle, and receives manual driving information as information of manual driving of the second vehicle in the area of ​​the interruption position from a vehicle-mounted device mounted on a second vehicle; an extraction unit, which extracts a control point as a location for changing the direction of travel from the manual driving information in the area of ​​the interruption position; and a production unit, which produces knowledge information including driving information of the vehicle on a trajectory passing through the control point and capable of generating a track for changing the direction of travel of the vehicle; the server communication unit also sends the knowledge information and at least one of the information indicating the trajectory of the vehicle generated using the knowledge information to the vehicle-mounted device.

[0011] The computing device of the first technical solution of the present invention comprises: a server communication unit, which receives information of an interruption position as a position where automatic driving is interrupted from a vehicle-mounted device mounted on a first vehicle, and receives manual driving information as information of manual driving of the second vehicle in the area of ​​the interruption position from a vehicle-mounted device mounted on a second vehicle which is a vehicle different from the first vehicle; an extraction unit, which extracts a control point as a location for changing the direction of travel from the manual driving information in the area of ​​the interruption position; and a production unit, which produces knowledge information including driving information of the speed or steering angle of the vehicle on a trajectory passing through the control point and capable of generating a trajectory for changing the direction of travel of the vehicle; the server communication unit also sends the knowledge information and at least one of the information representing the trajectory of the vehicle generated using the knowledge information to the vehicle-mounted device.

[0012] The vehicle-mounted device of the second technical solution of the present invention comprises: a recording unit, which records information of an interruption position as a position where the automatic driving is interrupted, and manual driving information as information of manual driving after the automatic driving is interrupted; an extraction unit, which uses the above-mentioned manual driving information to extract control points as locations regarding a change in the direction of travel; a production unit, which produces knowledge information capable of generating a track for changing the direction of travel of the vehicle based on the above-mentioned control points; and a track generation unit, which uses the above-mentioned knowledge information to produce a track for the vehicle.

[0013] The automatic driving system of the third technical solution of the present invention is an automatic driving system including a vehicle-mounted device installed in each of a plurality of vehicles and a server communicating with the above-mentioned vehicle-mounted device, wherein the above-mentioned vehicle-mounted device has an on-vehicle communication unit that sends information of an interruption position as a position where the automatic driving is interrupted and manual driving information as information of manual driving after the automatic driving is interrupted in a certain vehicle to the above-mentioned server; the above-mentioned server has: a server communication unit that receives the information of the interruption position and the above-mentioned manual driving information from the above-mentioned vehicle-mounted device; an extraction unit that extracts a control point as a location regarding a change in the direction of travel from the above-mentioned manual driving information; a production unit that produces knowledge information that can generate a track for changing the direction of travel of the vehicle based on the above-mentioned control point; and a sending unit that sends the above-mentioned knowledge information and at least one of the information representing the trajectory of the vehicle generated using the above-mentioned knowledge information to the above-mentioned vehicle-mounted device of a certain vehicle.

[0014] Effects of the Invention

[0015] According to the present invention, it is possible to provide information that enables autonomous driving to be performed at a location where autonomous driving was interrupted in the past. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is the hardware structure diagram of the autonomous driving system.

[0017] Figure 2 It is a functional block diagram of the autonomous driving system.

[0018] Figure 3 This is a diagram showing an example of a knowledge DB.

[0019] Figure 4 This is a diagram showing an example of interruption information.

[0020] Figure 5 It is a diagram showing an operation example.

[0021] Figure 6 This is a flowchart showing the processing of the automatic control unit.

[0022] Figure 7 It is a flowchart showing the details of step S602.

[0023] Figure 8 This is a flowchart showing the record creation process of the knowledge DB performed by the server.

[0024] Fig. 9 It is a flowchart showing the details of step S704.

[0025] Fig.10 This is a diagram showing the second operation example.

[0026] Fig.11This is a diagram showing the third operation example.

[0027] Fig.12 This is a functional structure diagram of the automatic driving system of variant example 1.

[0028] Fig.13 This is a functional block diagram of the automatic driving system of variant example 5.

[0029] Fig.14 This is a hardware structure diagram of the automatic driving system of variant example 8.

[0030] Fig.15 It is a functional configuration diagram of the vehicle-mounted device according to the second embodiment.

[0031] Description of symbols

[0032] 10, 10-2, 10-3…In-vehicle device

[0033] 10A…New specification vehicle-mounted device

[0034] 10B…Old specification vehicle-mounted device

[0035] 14…In-vehicle communication department

[0036] 14A…New specification vehicle communication unit

[0037] 14B…Old specification vehicle communication unit

[0038] 20…Server

[0039] 24…Server Communication Department

[0040] 25…Server Storage

[0041] 112…Records Department

[0042] 113…Path generation unit

[0043] 114, 214…Trajectory generation unit

[0044] 116…Interrupt information sending unit

[0045] 117, 217…Extraction department

[0046] 118, 218…Production Department

[0047] 151, 251…Knowledge DB

[0048] 152, 252…High-precision maps

[0049] 153…Interruption information

[0050] 219…Target determination department DETAILED DESCRIPTION

[0051] - First Embodiment -

[0052] Below, refer to Figures 1 to 11 , describing a first embodiment of an automatic driving system including a vehicle-mounted device and a server.

[0053] (Definition of terms)

[0054] In this embodiment, a vehicle equipped with the in-vehicle device of the structure and operation described in the description is referred to as the "own vehicle". A person riding in the own vehicle is referred to as a "user". In this embodiment, a combination of latitude and longitude is referred to as an "absolute coordinate". In addition, in this embodiment, a "path" is path information with a resolution capable of identifying the road on which the vehicle passes, for example, information that arranges the identification codes of nodes and links in the order in which the own vehicle travels.

[0055] In this embodiment, the "automatic driving" refers to a state in which at least one of the driving sources such as the engine or the motor, the braking device and the steering wheel is controlled through calculation processing. In addition, the "manual driving" refers to a state in which the vehicle is in motion, not an automatic driving state. The "manual interruption" of the automatic driving refers to the user's intervention in the area of ​​the automatic driving in the state of automatic driving. For example, when the definition of automatic driving is the control of the driving source and the braking device, the user's behavior of stepping on the accelerator pedal or the brake pedal corresponds to the manual interruption of the automatic driving.

[0056] The "track" in this embodiment is information about positions with sufficient resolution for the realization of autonomous driving, for example, multiple absolute positions with mm-level resolution are arranged in the order in which the vehicle is about to travel. The "track" in this embodiment is information about the absolute positions of the vehicle actually traveling in the order in which it is traveling. That is, the position that is traveled in the past is the "track", and the position that is expected to travel in the future is the "track".

[0057] (Hardware Structure)

[0058] Figure 1 : is a diagram showing the hardware structure of the automatic driving system S of the present embodiment. The automatic driving system S is composed of vehicle-mounted devices 10 respectively mounted on a plurality of vehicles and a server 20 communicating with each vehicle-mounted device 10. Each vehicle-mounted device 10 has a common structure described below. However, the structure of each vehicle-mounted device 10 may be different because each vehicle-mounted device 10 has a structure other than the structure described below. In addition, the following description assumes that the vehicle-mounted device 10 and the server 20 are each one device, but the vehicle-mounted device 10 and the server 20 may each be composed of a plurality of devices.

[0059] The vehicle-mounted device 10 includes a CPU 10 as a central processing unit, a ROM 12 as a dedicated storage device for reading, a RAM 13 as a readable and writable storage device, a vehicle-mounted communication unit 14, a vehicle-mounted storage unit 15, and a sensor 16. The program stored in the ROM 12 is expanded into the RAM 13 by the CPU 11 and executed, thereby realizing multiple functions described later. However, it is also possible to replace the combination of the CPU 11, ROM 12, and RAM 13 with an FPGA (Field Programmable Gate Array) as a rewritable logic circuit or an ASIC (Application Specific Integrated Circuit) as an integrated circuit for a specific purpose. In addition, the vehicle-mounted device 10 may also replace the combination of the CPU 11, ROM 12, and RAM 13 with a combination of different structures, such as a combination of the CPU 11, ROM 12, RAM 13, and FPGA.

[0060] The vehicle-mounted communication unit 14 is a communication module capable of wireless communication for communicating with the server 20. The communication standard corresponding to the vehicle-mounted communication unit 14 is not particularly limited, but the vehicle-mounted communication unit 14 corresponds to, for example, 3G, 4G, and 5G. The vehicle-mounted communication unit 14 can be directly connected to the server 20, or can be connected to the server 20 via a base station provided by a communication manufacturer or the Internet. The vehicle-mounted storage unit 15 is a non-volatile storage device, such as a flash memory. The sensor 16 is one or more sensors that collect information about the surroundings of the vehicle. The sensor 16 is, for example, a camera and a laser rangefinder. The information obtained by the sensor 16 is used for the recognition of road markings and the detection of obstacles.

[0061] The server 20 includes a CPU 21 as a central processing unit, a ROM 22 as a dedicated storage device for reading, a RAM 23 as a readable and writable storage device, a server communication unit 24, and a server storage unit 25. The CPU 21 expands the program stored in the ROM 22 into the RAM 23 and executes it, thereby realizing multiple functions described later. However, it is also possible to replace the combination of the CPU 21, ROM 22, and RAM 23 with an FPGA or an ASIC. In addition, the server 20 may also replace the combination of the CPU 21, ROM 22, and RAM 23 with a combination of different structures, such as a combination of the CPU 21, ROM 22, and RAM 23 with an FPGA.

[0062] The server communication unit 24 communicates with the vehicle communication unit 14. The server communication unit 24 may communicate directly with the vehicle communication unit 14 or communicate via other devices such as a base station or other communication devices. The server storage unit 25 is a non-volatile storage device such as a hard disk drive.

[0063] (Functional Structure)

[0064] Figure 2 is a functional block diagram that represents the functions of the automatic driving system S as blocks. Figure 2 The data is also shown in the figure. The vehicle-mounted device 10 has a drive command unit 111, a recording unit 112, a path generation unit 113, a track generation unit 114, an automatic control unit 115, and an interrupt information sending unit 116 as its functions. The drive command unit 111 drives the driving source, the braking device, and the steering device of the vehicle based on the operation performed by the user and the action command of the automatic control unit 115. The operation performed by the user is the rotation of the steering wheel, the stepping of the accelerator pedal, the stepping of the brake pedal, etc.

[0065] The recording unit 112 records the information of the driving of the vehicle (hereinafter referred to as "driving information") to the RAM 13. The driving information is information that can calculate the trajectory of the vehicle's driving afterwards, such as a list of the absolute positions of the vehicles arranged in the order of the vehicle's driving. In addition, the driving information may also be information on the speed and steering angle of the vehicle at each moment. In addition, the driving information may also be sensor information such as cameras, lasers, LiDAR, etc. The recording unit 112 records the driving information at least when manual driving is performed. In other words, the recording unit 112 may or may not record the driving information when automatic driving is performed. The driving information includes information from the time when the automatic driving is interrupted and manual driving begins to the time when the driving is specified, such as 10 seconds, or a specified distance, such as 100 meters.

[0066] The route generation unit 113 calculates the route that the vehicle is to travel. For example, the route generation unit 113 calculates the route from the current position to the destination specified by the user. The route generation unit 113 may generate multiple routes and present them to the user, and output the information of the route selected by the user to the automatic control unit 115, or may generate only one route and output the information of the route to the automatic control unit 115 without waiting for the user's selection.

[0067] The track generation unit 114 calculates the track on which the vehicle is traveling. The track generation unit 114 calculates the track between the designated nodes with reference to the high-precision map and the knowledge DB 151. Hereinafter, among the two nodes for which the track generation unit 114 calculates the track, the node closer to the departure point is referred to as the "first node", and the node closer to the destination is referred to as the "second node".

[0068] The automatic control unit 115 starts the automatic driving of the host vehicle 9 based on the user's operation. The automatic control unit 115 ends the automatic driving in any of the following situations. That is, the automatic control unit 115 ends the automatic driving when the host vehicle 9 arrives at the destination, when the user intervenes in the automatic driving and manually interrupts the automatic driving, and when the automatic control unit 115 itself spontaneously determines that it is appropriate to end the automatic driving using the information obtained by the sensor 16 (hereinafter referred to as "spontaneous interruption"). Hereinafter, the manual interruption of the automatic driving and the spontaneous interruption of the automatic driving performed by the automatic control unit 115 itself are collectively referred to as "forced interruption" of the automatic driving. The automatic control unit 115 itself determines that it is appropriate to end the automatic driving, for example, when the driving route as the driving plan cannot be maintained, or when the driving route cannot be reached.

[0069] Furthermore, if forced interruption of the automatic driving occurs, the automatic control unit 115 outputs a cause ID indicating the cause of the interruption. By referring to the cause ID, it is possible to perform a post-analysis of not only whether the interruption is spontaneous but also why the spontaneous interruption occurred.

[0070] When the recording of the driving information by the recording unit 112 is completed, the interruption information sending unit 116 sends the absolute coordinates where the forced interruption occurred, the cause ID indicating the cause of the forced interruption, the driving information recorded by the recording unit 112, the vehicle type of the vehicle, and the current date and time to the server 20 as the interruption information 153. Figure 2 For convenience, the interruption information 153 is recorded as being stored in the vehicle-mounted storage unit 15, but the interruption information 153 may be stored in the RAM 13. Furthermore, the interruption information sending unit 116 may fragment the interruption information 153, and send the interruption information 153 to the server 20 by sending the information in order from the information that can be sent, without storing the entire interruption information 153 in the RAM 13.

[0071] In addition, the recording unit also records the driving information in the case where the user himself selects to drive and sets the driving to manual driving (hereinafter referred to as "real manual driving"), in addition to the case where the automatic driving is interrupted. Therefore, the interruption information sending unit 116 also sends the record of the real manual driving to the server 20 as the interruption information 153. However, in the case of the real manual driving, since the automatic driving is not interrupted, the coordinates indicating the occurrence of the forced interruption and the reason ID indicating the reason for the forced interruption constituting the interruption information 153 use a value indicating an empty set, for example.

[0072] The server 20 has an extraction unit 217 and a production unit 218 as its functions. The extraction unit 217 uses the driving information received from the vehicle-mounted device 10 to calculate the track that the vehicle has traveled, and extracts the control points that are the intersections of the track and the road markings. Specifically, the extraction unit 217 calculates the absolute coordinates of the control points. The production unit 218 uses the control points extracted by the extraction unit 217 to calculate the automatic driving control points as described later, generates the trajectory of the vehicle, and creates a new record of the knowledge DB 251. In addition, although the details will be described later, the operation of the extraction unit 217 and the production unit 218, that is, the extraction of control points and the calculation of automatic driving control points, has two methods, the first mode and the second mode. Which mode of the first mode and the second mode the extraction unit 217 and the production unit 218 operate in is set in advance by, for example, the operator of the server 20. The operation of the extraction unit 217 and the production unit 218 will be described later together with a specific example. The server communication unit 24 transmits the knowledge DB 251 to all the vehicle-mounted devices 10 at an arbitrary timing, for example, every certain time or every time a new record is added to the knowledge DB 251 .

[0073] (Data Structure)

[0074] The vehicle storage unit 15 stores a knowledge DB 151, a high-precision map 152, and interruption information 153. The server storage unit 25 stores a knowledge DB 251 and a high-precision map 252. Since the knowledge DB 151 and the knowledge DB 251 have the same structure, the structure of the knowledge DB 151 will be described below as a representative of both.

[0075] The knowledge DB 151 is a database created by the server 20 to supplement the automatic driving. The knowledge DB 151 is composed of a plurality of records, and each record constituting the knowledge DB 151 is also referred to as "knowledge information" hereinafter. That is, the knowledge DB 151 is a collection of knowledge information. In addition, the knowledge information may also include information generated based on data obtained by sensors such as cameras or laser rangefinders. In the knowledge DB 151, information used to realize automatic driving at locations where automatic driving was interrupted in the past is stored. Specifically, the knowledge DB 151 stores information about tracks suitable for automatic driving as tracks from the first node to the second node for a plurality of locations. Each record of the knowledge DB 151 may also include restrictions on vehicle types and date and time. In addition, each record of the knowledge DB 151 may also include execution conditions generated by judging surrounding information of a mobile body, etc., based on sensor information such as cameras, lasers, and LiDAR. For example, when there is a vehicle in front, a line of vehicles on the right, and a drivable space on the left, conditions such as driving in the drivable space on the left according to the path. The trajectory generation unit 114 can generate a trajectory in which the autonomous driving is not interrupted even at the point where the autonomous driving is interrupted, by referring to the knowledge DB 151 , using an existing trajectory generation method.

[0076] Figure 3 151 is a diagram showing an example of a knowledge DB 151. The knowledge DB 151 is composed of a plurality of records, each of which has fields for a number 1511, a vehicle type 1512, a date and time 1513, an interruption position 1514, a first node 1515, a second node 1516, a driving coordinate 1517, and an automatic driving control point 1518. In the field for the number 1511, an identification number of each record is stored. Figure 3 In the example shown, a common number is assigned to a series of records, and branch numbers are assigned to distinguish a plurality of records in the series.

[0077] The vehicle type 1512 field stores identification information indicating the vehicle type that sent the information recorded in the record. That is, the vehicle type identification information stored in the vehicle type 1512 field and the vehicle type identification information of the own vehicle are not necessarily the same. The date and time when the interruption information 153 recorded in the record was generated is stored in the date and time 1513 field.

[0078] The absolute coordinates of the automatic control forced interruption are stored in the interruption position field 1514. The identification information of the first node of the vehicle when the interruption information 153 recorded in the record is stored in the first node field 1515. The identification information of the second node of the vehicle when the interruption information 153 recorded in the record is stored in the second node field 1516.

[0079] In the field of the travel coordinates 1517, the absolute coordinates of the vehicle's planned passage, which are created by the server 20 using the travel information included in the interruption information 153, are stored. Figure 3 In the example shown, the vehicle is shown with In the field of the automatic driving control point 1518, the absolute coordinates of the automatic driving control point calculated by the processing described later are stored.

[0080] The high-precision map 152 includes map information of the accuracy required for calculating the path and the track. For example, in order to calculate the path, the high-precision map 152 includes a list of nodes that are points set corresponding to intersections and branches on the road, a list of links that are arranged along the road and connect two road nodes, the absolute coordinates of each node, the length of each link, and information on the speed limit in each link. In addition, for example, in order to calculate the track, the high-precision map 152 includes information such as the position of the road markings in each link, the type of road markings, the curvature of the road markings, the position of the road signs, the slope of the road, etc.

[0081] The high-precision map 252 is substantially the same as the high-precision map 152. The high-precision map 252 is preferably consistent with the high-precision map 152, but as long as they are the same type of information, the accuracy of the two may be different.

[0082] The interruption information 153 includes information on the absolute coordinates where the forced interruption occurred, a cause ID indicating the cause of the forced interruption, the travel information recorded in the recording unit 112 , the vehicle type of the host vehicle, and the current date and time. Figure 4 153 is a diagram showing an example of interruption information 153. Interruption information 153 has fields for an interruption position 1531, a cause ID 1532, a date and time 1533, a vehicle type 1534, and driving information 1535. In the interruption position 1531, the absolute coordinates where the forced interruption occurred are stored. In the cause ID 1532, an ID indicating the cause of the interruption of the autonomous driving is stored. In the date and time 1533, the date and time when the autonomous driving is interrupted is stored. In the vehicle type 1534, an identifier indicating the vehicle type of the vehicle is stored. In the driving information 1535, the coordinates of the vehicle moved by manual driving after the autonomous driving is interrupted are stored in a time series order.

[0083] (Action Example)

[0084] Figure 5 1 is a diagram showing a situation where the vehicle is traveling from node N950 to node N959 by automatic driving. The trajectory generation unit 114 of the vehicle-mounted device 10 calculates Figure 5 The vehicle is controlled along the track R1 indicated by the solid line in FIG. 1 . However, the user determines that the lane change caused by the automatic driving is too slow, or the automatic control unit 115 determines that it is appropriate to end the automatic driving based on the information obtained by the sensor 16, and the absolute position is At point P1, the user operates the steering wheel to change the lane. Figure 5 The trajectory R2 represented by the single-dot chain line in the figure moves forward and reaches the node N959.

[0085] In this case, the recording unit 112 of the vehicle-mounted device 10 records the absolute coordinates on R2 indicated by the single-dot chain line, for example The interruption information sending unit 116 sends the recorded driving information and the interruption position to the The interruption information 153 is sent to the server 20.

[0086] If the server 20 receives the interruption information 153, the extraction unit 217 calculates the trajectory R2 using the driving information. The extraction unit 217 and the production unit 218 perform different processes in the first mode and the second mode as follows. First, the operation of the extraction unit 217 and the production unit 218 in the first mode is described, and then the operation of the extraction unit 217 and the production unit 218 in the second mode is described. The extraction unit 217 sets the interruption position P1 as the control point P1 in the first mode. In addition, the production unit 218 does not move the control point P1 but sets it as the automatic driving control point P1, and produces a new record having the information including the control point P1 as the knowledge DB 251. The absolute coordinates on the track R2 are used as the driving coordinates and the automatic driving control point is The first node is N950 and the second node is N959.

[0087] The extraction unit 217 in the second mode calculates the position of the road markings using the high-precision map 252, and calculates the control points P21 and P31 which are the intersections of the trajectory R2 and the road markings. Furthermore, the production unit 218 of the second mode calculates the first distance L by a predetermined method, and calculates the automatic driving control points that move the control points the first distance L in front of the direction of travel. That is, the control point P21 moves to the automatic driving control point P22, and the control point P31 moves to the automatic driving control point P32. Furthermore, the production unit 218 of the second mode calculates the trajectory R3 from the node N950 to the node N959 which passes through the automatic driving control point P22 and the automatic driving control point P32. Finally, the production unit 218 of the second mode produces a new record of the knowledge DB 251 having the absolute coordinates on the trajectory R3. As driving coordinates, the first node is N950 and the second node is N959.

[0088] like Figure 5 As shown in the upper part of , in the first mode and the second mode, the control points and the automatic driving control points are different, and the trajectory of the automatic driving using the generated knowledge information for automatic driving is also different. Since the position where the lane change starts in the second mode is moved by the first distance L to the front side of the travel direction compared to the first mode, the automatic driving is not easily released in the second mode even if congestion occurs. In addition, the automatic driving control point P1 in the track R2 is equivalent to the place where the travel direction is changed in the sense of the place where the steering angle is changed, and the control point P21 and the control point P31 are equivalent to the place where the travel direction is changed in the sense of the travel lane.

[0089] (flow chart)

[0090] Figure 61 is a flowchart showing the processing of the automatic control unit 115. The CPU 11 of the vehicle-mounted device 10 is the main body for executing each step described below. The automatic control unit 115 receives the route information calculated by the route generation unit 113 in advance and starts Figure 6 Processing shown.

[0091] The automatic control unit 115 first reads the path information calculated by the path generation unit 113 in step S601. In the next step S602, the automatic control unit 115 generates a track to the next node of the path using the track generation unit 114. Figure 6 Although the description is omitted in the first calculation of step S602, the trajectory is calculated with the departure point set as the first node and only one node close to the destination from the departure point set as the second node. For details of step S602, refer to Figure 7 This will be described in detail later.

[0092] In the next step S603, the automatic control unit 115 determines whether the automatic driving is forcibly interrupted. If the automatic control unit 115 determines that it is forcibly interrupted, that is, a manual interruption or a spontaneous interruption occurs, the automatic control unit 115 proceeds to step S604, and if it determines that it is not forcibly interrupted, the automatic control unit 115 proceeds to step S606.

[0093] In step S604, the automatic control unit 115 uses the recording unit 112 to record the driving information. However, since it takes time to collect the driving information, the automatic control unit 115 waits in step S604 until the vehicle has traveled a predetermined time or a predetermined distance. In the next step S605, the automatic control unit 115 sends the absolute coordinates of the vehicle when the automatic driving was interrupted, the reason for the automatic driving being interrupted, the first node and the second node when the automatic driving was interrupted, and the driving information recorded in step S604 to the server 20, and the process ends. Figure 6 processing.

[0094] In step S606, which is executed if a negative judgment is made in step S603, the automatic control unit 115 controls the vehicle so that the vehicle travels along the track generated in step S602. In addition, the automatic control unit 115 may also control the vehicle using information obtained from the sensor 16 while considering the automatic driving control point 1518 of the knowledge DB 251 as a reference value to avoid collision with other vehicles. In the following step S607, the automatic control unit 115 determines whether the second node has been reached. If the automatic control unit 115 determines that the second node has been reached, the automatic control unit 115 proceeds to step S608. If the automatic control unit 115 determines that the second node has not been reached, that is, it is on the way along the track generated in step S602, it returns to step S603.

[0095] In step S608, the automatic control unit 115 determines whether the destination has been reached. If it is determined that the destination has been reached, the process ends. Figure 6 In the process shown, if it is determined that the destination has not been reached, the process proceeds to step S609. In step S609, the automatic control unit 115 updates the first node and the second node and returns to step S602. Specifically, the automatic control unit 115 moves the first node and the second node by one node each to the side closer to the destination.

[0096] Figure 7 Yes means Figure 6 The flowchart of step S602 is shown in detail. In step S611, the automatic control unit 115 determines the search conditions and searches for records in the knowledge DB 151. Items of the search conditions are, for example, vehicle type, time, first node and second node. Specifically, the vehicle type of the vehicle is completely consistent with the vehicle type recorded in the knowledge DB 151 or consistent within a prescribed range, the current time is completely consistent with the time recorded in the knowledge DB 151 or consistent within a prescribed range, the currently set first node and second node are consistent with the first node and second node recorded in the knowledge DB 151, etc. In the next step S612, the automatic control unit 115 proceeds to step S613 when it is determined that there are records that meet the search conditions, and proceeds to step S614 when it is determined that there are no matching records.

[0097] In step S613, the automatic control unit 115 generates a track with reference to the matching records of the knowledge DB 151 and the high-precision map. Specifically, the automatic control unit 115 includes the track that moves along the driving coordinates that match the record in the track from the first node to the second node. Among them, when the driving coordinates in the matching records of the knowledge DB 151 are the same as the automatic driving control point, the automatic control unit 115 changes the direction of travel at the coordinates of the automatic driving control point. In step S614, the automatic control unit 115 calculates the track from the first node to the second node with reference to the high-precision map.

[0098] Figure 8 This is a flowchart showing the record generation process of the knowledge DB 251 performed by the server 20. The server 20 executes the process every time it receives the interruption information 153 from the vehicle-mounted device 10. Figure 8 In addition, steps S701 to S703 and step S711 described below are executed by the extraction unit 217, and steps S704 to S706 and step S716 are executed by the creation unit 218.

[0099] In step S700 , the CPU 21 of the server 20 determines whether the first mode or the second mode is set. If the CPU 21 determines that the first mode is set, the process proceeds to step S711 , and if the CPU 21 determines that the second mode is set, the process proceeds to step S701 .

[0100] In step S701, the extraction unit 217 calculates the trajectory of the vehicle based on the manual driving information included in the interruption information 153. Since the manual driving information is, for example, a list of position information, that is, point information, the information of the line that smoothly connects these points, that is, the trajectory that the vehicle has traveled, is calculated in absolute coordinates. In the following step S702, the extraction unit 217 calculates the white lines of the road around the interruption position, that is, the road markings, in absolute coordinates based on the high-precision map. In the following step S703, the extraction unit 217 calculates the intersection of the vehicle trajectory calculated in step S701 and the road markings calculated in step S702, that is, the control point.

[0101] In the following step S704, the production unit 218 produces an automatic driving control point that moves the control point forward by a specified distance relative to the direction of travel. The method for determining the processing distance and the detailed production method of the automatic driving control point will be described later. In the following step S705, the production unit 218 calculates a track from the first node to the second node that passes through the automatic driving control point. This track is generally a track that moves the track calculated based on the manual driving information in parallel by a specified distance. In the following step S706, the production unit 218 uses the track generated in step S705 and the received interruption information 153 to record a new record in the knowledge DB 251, and ends. Figure 8 processing.

[0102] The records created in step S706 are, for example, the following records. In number 1511 of knowledge DB 251, the smallest number that has not been used so far is saved. In vehicle type 1512, information on the vehicle type included in the interruption information 153 is saved. In date and time 1513, information on the date and time included in the interruption information 153 is saved. In interruption position 1514, information on the absolute coordinates where the autonomous driving is interrupted is saved. In first node 1515 and second node 1516, information on the first node and the second node included in the interruption information 153 is saved. In driving coordinates 1416, the absolute coordinates on the track calculated in step S705 are saved at predetermined intervals, for example, 1 m. In autonomous driving control point 1417, the absolute coordinates of the autonomous driving control point are saved.

[0103] In step S711, the extraction unit 217 calculates the trajectory of the vehicle based on the manual driving information included in the interruption information 153, similarly to step S701. In the next step S716, the creation unit 218 records a new record in the knowledge DB 251 with the interruption position included in the interruption information 153 as the automatic driving control point and the absolute coordinates on the trajectory calculated in step S711 as the driving coordinates, and the process ends. Figure 8 The driving coordinates include the absolute coordinates of the interruption position.

[0104] Fig. 9 Yes means Figure 8 The flowchart of the details of step S704 is shown in FIG. Fig. 9 In step S711, the creation unit 218 first reads the calculation method of the first distance. For example, three calculation methods A to C are preset, and which one is used is recorded in the ROM 22 or the server storage unit 25. However, which calculation method to use can also be specified from outside the server 20 each time.

[0105] In the next step S712, the creation unit 218 determines which of A to C the calculation method read in step S711 is. If the creation unit 218 determines that the calculation method is "A", the process proceeds to step S713; if the creation unit 218 determines that the calculation method is "B", the process proceeds to step S714; and if the creation unit 218 determines that the calculation method is "C", the process proceeds to step S715.

[0106] In step S713, the creation unit 218 sets the first distance to a preset fixed value, such as 20 meters, and proceeds to step S716. In step S714, the creation unit 218 sets the first distance to the product of the speed limit in the link where the first node exists and the preset number of seconds, and proceeds to step S716. In step S714, for example, when the speed limit is 60 kilometers per hour and the preset number of seconds is 2 seconds, the first distance is calculated to be 33 meters. In step S714, the product of the speed limit at the control point and the preset number of seconds may also be used.

[0107] In step S715, the creation unit 218 calculates the first distance as the product of the speed limit in the link where the first node exists, the number of lane changes from the first node to the second node, and the preset number of seconds, and proceeds to step S716. In step S715, for example, when the speed limit is 60 km / h, the number of lane changes from the first node to the second node is "3", and the preset number of seconds is 1 second, the first distance is calculated to be 50 m.

[0108] In step S716, the production unit 218 makes Figure 8The control point calculated in step S703 is moved to the front side of the traveling direction by a first distance as the automatic driving control point, and the automatic driving is ended. Fig. 9 In addition, when there are a plurality of control points calculated in step S703, each of them is moved by the first distance in the same manner.

[0109] (Other action examples)

[0110] Fig.10 FIG. 2 is a diagram showing a second operation example of the first embodiment. Fig.10 In FIG. 1 , the first node is node N950 and the second node is node N959, and the vehicle is traveling from the bottom to the top of the diagram. Fig.10 The left and right roads merge at the central part. The track generation unit 114 calculates the track R1 represented by the solid line and controls the vehicle along the track R1. However, the user ends the automatic driving by himself, or after the automatic driving ends based on sensor information, etc., the user operates the steering wheel to avoid the merging point and advances along the track R2 represented by the single-dot chain line to reach the node N959. The interruption information sending unit 116 sends the interruption information 153 to the server 20, and the server 20 generates knowledge information that can generate the track R3 represented by the double-dot chain line, that is, a new record of the knowledge DB251 based on the interruption information 153.

[0111] Fig.11 FIG. 3 is a diagram showing a third operation example of the first embodiment. Fig.11 , a situation in which the first node is node N950 and the second node is node N959 is shown. Multiple vehicles heading toward node N959 are parked in a line. The track generation unit 114 calculates track R1 represented by a solid line and controls the vehicle along track R1. At this time, after the sensor detects the line of multiple vehicles heading toward node N959 and the automatic driving ends, the user advances along track R2 represented by a single-dot chain line and reaches node N959 to be arranged at the end of the line of parked vehicles. The interruption information sending unit 116 sends the interruption information 153 to the server 20, and the server 20 generates knowledge information capable of generating track R3 represented by a double-dot chain line, i.e., a new record of the knowledge DB 251 in the second mode based on the interruption information 153. In addition, the server 20 generates a new record of the knowledge DB 251 capable of generating track R2 represented by a single-dot chain line in the first mode.

[0112] (Relationship between sending and receiving data)

[0113] Here, the vehicle that sends the interruption information 153 about the generation of a certain record of the knowledge DB 151, that is, a certain knowledge information, is referred to as the "first vehicle". In addition, the vehicle that sends the manual driving information about the generation of the knowledge information is referred to as the "second vehicle", and the vehicle that receives the certain knowledge information from the server 20 is referred to as the "third vehicle". Figure 5 In the operation example shown, since the autonomous driving of the host vehicle is interrupted and the information of the subsequent manual driving of the user is transmitted to create knowledge information, the host vehicle corresponds to both the first vehicle and the second vehicle. Figure 5 In the illustrated operation example, if the host vehicle receives the knowledge information indicating the route R2 or the route R3 from the server 20, the host vehicle also corresponds to the third vehicle. However, if the host vehicle does not receive the knowledge information, the host vehicle is not the third vehicle with respect to the knowledge information.

[0114] In addition, the first vehicle and the second vehicle may not be consistent as follows. For example, it is assumed that a certain vehicle-mounted device 10 sends the first interruption information 153 of automatic driving, and manual driving information in actual manual driving is sent from multiple other vehicles traveling at the same first node to second node. In this case, if the server 20 generates knowledge information using the average value of the received manual driving information except for the deviation value, the manual driving information included in the first interruption information 153 is ignored in the case of a deviation value. Therefore, in this case, the manual driving information of the vehicle that generated the first interruption information 153 is not used for the generation of knowledge information, and although it corresponds to the first vehicle, it does not correspond to the second vehicle. In addition, since whether to correspond to the third vehicle is determined by whether its knowledge information is received, there are cases where it corresponds to the first vehicle and the third vehicle, and there are also cases where it corresponds to only the first vehicle.

[0115] As described above, when focusing on a certain record of the knowledge DB 251, that is, one piece of knowledge information, the first vehicle, the second vehicle, and the third vehicle may all be the same, or only correspond to two or only correspond to one.

[0116] According to the above-mentioned first embodiment, the following effects can be obtained.

[0117] (1) The server 20 as a computing device includes a server communication unit 24, which receives an interruption position 1531 of the interruption information 153, which is information of an interruption position where the automatic driving is interrupted, from the vehicle-mounted device 10 mounted on the first vehicle, and receives manual driving information, which is information of manual driving of the second vehicle in the area of ​​the interruption position, from the vehicle-mounted device mounted on the second vehicle, which is information included in the driving information 1535 of the interruption information 153. The server 20 includes an extraction unit 217 that extracts a control point as a location for changing the traveling direction from the manual driving information in the area of ​​the interruption position, and a creation unit 218 that creates knowledge information capable of generating a track for changing the traveling direction of the vehicle based on the control point. The server communication unit 24 further transmits at least one of the knowledge information and information indicating the track of the vehicle generated using the knowledge information to the vehicle-mounted device 10 of the third vehicle. The third vehicle is the same as one of the first vehicle and the second vehicle, or is different from both the first vehicle and the second vehicle, and the first vehicle is the same as the second vehicle, or is different from the second vehicle. Therefore, the vehicle-mounted device 10 that receives the knowledge DB251 can realize automatic driving at the same location. In other words, the server 20 can provide knowledge information that can realize automatic driving at the location where automatic driving was interrupted in the past. In addition, the vehicle-mounted device 10 that receives the knowledge DB251 from the server 20 is not limited to the vehicle-mounted device 10 that sent the interruption information 153. That is, the third vehicle and the first vehicle may also be different. Therefore, for other vehicles equipped with the vehicle-mounted device 10 that did not send the interruption information 153, automatic driving can be realized even though it is the first time to travel in the place and automatic driving is difficult. In addition, as mentioned above, the second vehicle and the first vehicle may also be different.

[0118] (2) The creation unit 218 can create an autonomous driving control point for autonomous driving control and can create knowledge information that can generate a trajectory for changing the vehicle's traveling direction at the autonomous driving control point. Therefore, it is possible to generate a trajectory using the autonomous driving control point.

[0119] (3) The first distance is determined by one of methods A to C. Method A is a preset fixed value. Method B is the product of the speed limit on the link where the first node exists and the preset number of seconds. Method C is the product of the speed limit on the link where the first node exists, the number of lane changes from the first node to the second node, and the preset number of seconds.

[0120] (4) The control point is a point where the vehicle track intersects the vehicle line. Therefore, it is possible to generate a track based on the position where the vehicle lane is changed.

[0121] (5) The creation unit 218 can operate in at least two modes, namely, the first mode and the second mode. In the first mode, the creation unit 218 uses the control point as it is as the automatic driving control point, and in the second mode, creates the automatic driving control point by moving the control point forward by a first distance relative to the traveling direction. Therefore, in the second mode, the continuity of the automatic driving can be further improved.

[0122] (Variant 1)

[0123] In the first embodiment described above, the server 20 transmits the entire knowledge DB 251 to all connected vehicle-mounted devices 10. However, the server 20 may further include a target determination unit that determines the records of the knowledge DB 251 to be transmitted to each vehicle-mounted device 10, and transmit only a portion of the knowledge DB 251 to each vehicle-mounted device 10.

[0124] Fig.12 FIG. 1 is a functional block diagram of the automatic driving system of variant example 1. Fig.12 In, with Figure 2 In contrast, the server 20 further includes an object determination unit 219. The object determination unit 219 determines a record of the knowledge DB 251 that meets at least one condition of the vehicle type, date and time, and location for each vehicle-mounted device 10, and transmits it using the server communication unit 24. For example, the vehicle type condition is that the value of the vehicle type 1512 field of each record is consistent with the vehicle type of the vehicle on which the vehicle-mounted device 10 is mounted, or that the vehicle type is inconsistent but the difference in the size and weight of the vehicle body is within a specified range.

[0125] The date and time condition is that the date and time stored in the date and time field 1513 of each record and the current date and time are consistent with at least one of the season, month, day of the week and time band. The position condition is that the difference between the position stored in the interruption position field 1514 of each record and the latest position of each vehicle-mounted device 10 is within a specified range. For example, the straight-line distance or the distance along the road can be within a specified distance as a condition, and the condition can be that the absolute position is rewritten as an address expression and the address is the same within a specified range, such as the prefecture level or the city and town level.

[0126] According to this modification example 1, the following effects can be obtained.

[0127] (6) The server communication unit 24 receives information on the vehicle type and date and time from the vehicle-mounted device 10 as part of the interruption information 153. The knowledge information as each record of the knowledge DB 251 includes information on the vehicle type 1512, date and time 1513, and interruption position 1514 of the first vehicle and the second vehicle. The server 20 includes a target determination unit 219 that determines the target vehicle-mounted device to which the knowledge information is to be transmitted based on at least one of comparison of the vehicle type of the third vehicle with the vehicle type included in the knowledge information, comparison of the current date and time with the date and time included in the knowledge information, and comparison of the position of the third vehicle with the interruption position included in the knowledge information. Therefore, only the knowledge information suitable for use in each vehicle-mounted device 10 is transmitted from the server 20 to the vehicle-mounted device 10 as the third vehicle, so that the communication volume is suppressed and the search of the knowledge DB 151 of each vehicle-mounted device 10 is made efficient.

[0128] (Variant 2)

[0129] In the first embodiment described above, each record of the knowledge DB 151 includes the first node and the second node. However, the information of the absolute position may be stored instead of each node, and the direction of travel of the vehicle may also be stored. In addition, the interruption information sending unit 116 may not include the cause ID 1532 in the interruption information 153.

[0130] (Variant 3)

[0131] In the first embodiment described above, the server 20 generates a record of the knowledge DB 251 each time it receives the interruption information 153 from the vehicle-mounted device 10. However, the server 20 may generate a record of the knowledge DB 251 at a time by so-called batch processing. In this case, when there are a plurality of interruption information 153 having the same combination of the first node and the second node, the server 20 can group them using the average value of each driving coordinate.

[0132] (Variant 4)

[0133] The vehicle-mounted device 10 may include supplementary information such as information on the vehicle train ahead and traffic information obtained from the outside in the interruption information 153 transmitted to the server 20. The server 20 that has received the interruption information 153 including the supplementary information further stores the supplementary information by providing a field for storing the supplementary information in the record of the generated knowledge DB 251. In addition, the vehicle-mounted device 10 may also consider the supplementary information included in the knowledge information when generating the track.

[0134] (Variant 5)

[0135] In the first embodiment described above, the server 20 transmits the knowledge DB 251 to the vehicle-mounted device 10 . However, instead of transmitting the knowledge DB 251 to the vehicle-mounted device 10 , the server 20 may calculate the trajectory and transmit information indicating the trajectory to the vehicle-mounted device 10 .

[0136] Fig.13 This is a functional block diagram of the automatic driving system of variant example 5. Fig.13 Compared with the first embodiment Figure 2 The server 20 is different from the server 20 in that it further includes a track generation unit 214. Fig.13 In the embodiment, the knowledge DB 151 is not stored in the vehicle-mounted storage unit 15. The operation of the track generation unit 214 included in the server 20 is the same as that of the track generation unit 114 of the vehicle-mounted device 10 of the first embodiment.

[0137] In this variation, the vehicle-mounted device 10 sends the route calculated by the route generation unit 113 to the server 20. The track generation unit 214 of the server 20 calculates the track by referring to the knowledge DB 251 and the high-precision map 252, and sends the calculated track information to the vehicle-mounted device 10 using the server communication unit 24.

[0138] (Variant 6)

[0139] In the first embodiment described above, the vehicle-mounted device 10 includes the vehicle type and date and time in the interruption information 153 and transmits it to the server 20. However, the vehicle-mounted device 10 may not include at least one of the vehicle type and date and time in the interruption information 153. In this case, the knowledge DB 251 does not have a field for at least one of the vehicle type and date and time not included in the interruption information 153. In addition, in the process of track generation, Figure 7 In step S611, at least one of the vehicle type and the date and time not included in the interruption information 153 is deleted from the search conditions.

[0140] (Variant 7)

[0141] In the first embodiment described above, a field for the automatic driving control point 1417 is provided in the knowledge DB 251. However, the knowledge DB 251 may not have a field for the automatic driving control point 1417. This is because, as described in the first embodiment, the coordinates of the automatic driving control point 1417 are required in the creation of the driving coordinates 1517, but the vehicle-mounted device 10 can drive without referring to the automatic driving control point 1518 by using the driving coordinates 1517.

[0142] (Variant 8)

[0143] In the above-mentioned first embodiment, the configurations of the respective vehicle-mounted devices 10 are the same. However, the configurations and operations of the respective vehicle-mounted devices 10 may be different as described below.

[0144] Fig.14 is a system structure diagram showing the automatic driving system of this modified example. Fig.14 In the embodiment, the structure of the server 20 is different from that of the first embodiment. Figure 1 The vehicle-mounted device 10 of this variant is the same, so detailed description is omitted. The vehicle-mounted device 10 of this modification is classified into a new-specification vehicle-mounted device 10A and an old-specification vehicle-mounted device 10B. The new-specification vehicle-mounted device 10A has a new-specification vehicle-mounted communication unit 14A corresponding to 5G. The old-specification vehicle-mounted device 10B has an old-specification vehicle-mounted communication unit 14B corresponding to 4G or 3G, which are communication standards older than 5G. Since the new-specification vehicle-mounted communication unit 14A corresponds to 5G, it has the advantages of fast communication speed, a large number of simultaneous connections, and less communication delay compared to the old-specification vehicle-mounted communication unit 14B. The old-specification vehicle-mounted device 10B does not communicate directly with the server 20, but communicates with the server 20 via the new-specification vehicle-mounted device 10A.

[0145] According to this modification example 8, the following effects can be obtained.

[0146] (7) The vehicle-mounted device 10 is classified into an old-standard vehicle-mounted device 10B and a new-standard vehicle-mounted device 10A. The vehicle-mounted communication unit 14 of the new-standard vehicle-mounted device 10A is a new-standard vehicle-mounted communication unit 14A corresponding to 5G. The vehicle-mounted communication unit 14 of the old-standard vehicle-mounted device 10B is an old-standard vehicle-mounted communication unit 14B corresponding only to a communication standard older than 5G. The old-standard vehicle-mounted communication unit 14B transmits information on the interruption position and manual driving information to the server 20 via the new-standard vehicle-mounted communication unit 14A. Therefore, the server 20 or the base station can collect the interruption information 153 from a plurality of vehicle-mounted devices 10 and distribute the knowledge DB 251 to a plurality of vehicle-mounted devices 10.

[0147] (Variant 9)

[0148] In the first embodiment described above, the control point is the point where the trajectory of the vehicle intersects the vehicle marking line. However, the control point may also be the point where the vehicle starts to change its traveling direction.

[0149] (Variant 10)

[0150] In the first embodiment described above, it is assumed that the operation mode of the extraction unit 217 and the creation unit 218 is set in advance to one of the first mode and the second mode by the operator of the server 20. However, the extraction unit 217 and the creation unit 218 may perform operations in both the first mode and the second mode to create two knowledge DBs 251. In this case, the server 20 may also transmit the knowledge DB 251 created in the operation mode specified by the vehicle-mounted device 10 to the vehicle-mounted device 10.

[0151] (Variant 11)

[0152] A certain vehicle-mounted device 10 connected to the server 20 may include only one of the interruption position 1531 and the travel information 1535 in the interruption information 153. In other words, the vehicle-mounted device 10 may not correspond to the second vehicle by not including the travel information 1535 in the interruption information 153, and may not correspond to the first vehicle by not including the interruption position 1531 in the interruption information 153. Furthermore, the vehicle-mounted device 10 may not correspond to the third vehicle by not receiving the knowledge information from the server 20.

[0153] (Variant 12)

[0154] As knowledge information suitable for use in each of the above-mentioned vehicle-mounted devices 10, when there are a plurality of pieces of knowledge information to be sent based on at least one of comparison with the vehicle type included in the knowledge information, comparison with the current date and time included in the knowledge information, and comparison with the position of the third vehicle and the interruption position included in the knowledge information, the knowledge information as teacher data for optimizing these pieces of knowledge information may be sent to the vehicle-mounted device 10. Thus, automatic driving that is more suitable for the vehicle-mounted device 10 of the third vehicle and does not feel awkward to the person riding in the third vehicle can be continued.

[0155] - Second Embodiment -

[0156] Reference Fig.15 , describing the second embodiment of the automatic driving system. In the following description, the same reference numerals are given to the same components as those in the first embodiment, and the differences are mainly described. Points not specifically described are the same as those in the first embodiment. In this embodiment, the main difference from the first embodiment is that the knowledge DB is created by a vehicle-mounted device instead of a server. In addition, in this embodiment, a server may not exist. Since the hardware structure of the vehicle-mounted device in this embodiment is the same as that in the first embodiment, the description is omitted.

[0157] Fig.15This is a functional configuration diagram of the vehicle-mounted device 10-2 of the second embodiment. The vehicle-mounted device 10-2 of the second embodiment has, as its function, an extraction unit 117 and a production unit 118 in addition to the configuration of the first embodiment. The function and operation of the extraction unit 117 are the same as those of the extraction unit 217 of the server 20 of the first embodiment. The function and operation of the production unit 118 are the same as those of the production unit 218 of the server 20 of the first embodiment. The extraction unit 117 and the production unit 118 may be set in operation mode by a user mounted on the vehicle, or the operation mode may be specified in advance.

[0158] According to the above-mentioned second embodiment, the following effects can be obtained.

[0159] (8) The vehicle-mounted device 10-2 includes: a recording unit 112 that records information of an interruption position where the autonomous driving is interrupted and information of manual driving after the autonomous driving is interrupted; an extraction unit 117 that extracts a control point for starting a change in the traveling direction from the manual driving information; a creation unit 118 that creates an autonomous driving control point for controlling the autonomous driving based on the control point and creates knowledge information that can generate a track for changing the traveling direction at the autonomous driving control point; and a track creation unit 114 that creates a track for the vehicle using the knowledge information. Therefore, even in an environment where communication is not possible, the vehicle-mounted device 10-2 does not interrupt the autonomous driving during the next driving at the position where the autonomous driving was interrupted.

[0160] -Third Embodiment-

[0161] The third embodiment of the automatic driving system is described. In the following description, the same reference numerals are given to the same components as those in the first embodiment, and the differences are mainly described. Points not specifically described are the same as those in the first embodiment. This embodiment is different from the first embodiment mainly in that a path in which the automatic driving is interrupted is not generated. Since the hardware structure of the vehicle-mounted device of this embodiment is the same as that of the first embodiment, the description thereof is omitted.

[0162] In this embodiment, since the recording of the driving information is not necessary, the vehicle-mounted device 10-3 may not include the recording unit 112. The interruption information 153 of this embodiment does not include the driving information. Therefore, the interruption information sending unit 116 sends the interruption position, the interruption reason, the first node, and the second node to the server 20 as the interruption information 153.

[0163] In this embodiment, since calculation of control points, automatic driving control points, and driving coordinates is not necessary, the server 20 may not include the extracting unit 217. In addition, the creating unit 218 of this embodiment records the received interruption information 153 as a new record in the knowledge DB 251 as it is.

[0164] The route generation unit 113 of the vehicle-mounted device 10-3 refers to the knowledge DB 151 when generating a route, and excludes the combination of the first node and the second node recorded in the knowledge DB 151 from the generated route. When generating a route, the route generation unit 113 may treat the combination of the first node and the second node recorded in the knowledge DB 151 as a prohibited section.

[0165] According to the third embodiment, by generating a route where the autonomous driving is interrupted, it is possible to avoid the interruption of the autonomous driving.

[0166] In each of the above-mentioned embodiments and variations, the structure of the functional block is only an example. Some functional structures represented as different functional blocks may be integrally constituted, or a structure represented by one functional block diagram may be divided into two or more functions. In addition, a structure may be made in which a part of the functions of each functional block is possessed by other functional blocks. Furthermore, the hardware structure of each embodiment is only an example, and the device of each embodiment may also be implemented by multiple devices.

[0167] In each of the above-mentioned embodiments and variations, it is assumed that the program is stored in a ROM not shown in the figure, but the program may also be stored in a non-volatile storage area. In addition, the vehicle-mounted device or server may also be equipped with an input / output interface not shown in the figure, and when necessary, the program may be read from another device via a medium that can be used by the input / output interface. The medium described here is, for example, a storage medium or communication medium that is detachable from the input / output interface, that is, a wired, wireless, optical, or other network, or a carrier wave or digital signal transmitted in the network.

[0168] The above-mentioned embodiments and modifications can also be combined. In the above, various embodiments and modifications are described, but the present invention is not limited to these contents. Other forms that can be thought of within the scope of the technical idea of ​​the present invention are also included in the scope of the present invention.

[0169] The present invention claims priority based on Japanese Patent Application No. 2019-214763 (filed on November 27, 2019), the entire contents of which are incorporated herein by reference.

Claims

1. A computing device, have: a server communication unit that receives information of an interruption position, which is a position where the automatic driving is interrupted, from an onboard device mounted on a first vehicle, and receives manual driving information, which is information of manual driving of the second vehicle in an area of ​​the interruption position, from an onboard device mounted on a second vehicle; an extracting unit that extracts a control point as a location for changing the traveling direction from the manual driving information in the area of ​​the interruption position; as well as a production unit that produces knowledge information including travel information of the vehicle on a trajectory passing through the control point and capable of generating a trajectory for changing the traveling direction of the vehicle; The server communication unit further transmits at least one of the knowledge information and information indicating a trajectory of the vehicle generated using the knowledge information to an onboard device of the vehicle.

2. The computing device according to claim 1, The creation unit creates an autonomous driving control point used for autonomous driving control, and creates knowledge information capable of generating a trajectory for changing a traveling direction of a vehicle at the autonomous driving control point.

3. The computing device according to claim 1, The production unit is capable of operating in at least two modes: a first mode and a second mode; The production unit uses the control point as the automatic driving control point in the first mode, and moves the control point forward by a first distance relative to the traveling direction to produce the automatic driving control point in the second mode.

4. The computing device according to claim 3, The first distance is determined based on the legal speed limit at the control point and the number of required lane changes.

5. The computing device according to claim 1, The control point is either a point where the change of the traveling direction is started or a point where the trajectory of the vehicle intersects the vehicle marking line.

6. The computing device according to claim 1, The server communication unit further receives information on vehicle type and date and time from the first vehicle, the second vehicle, and the vehicle-mounted device of the vehicle; The knowledge information includes information on the vehicle types, date and time of the first vehicle and the second vehicle, and the interruption position; The above-mentioned computing device also has an object determination unit, which determines the above-mentioned vehicle-mounted device as the sending object vehicle-mounted device to which the above-mentioned knowledge information is sent based on at least one of a comparison between the vehicle type of the above-mentioned vehicle and the vehicle type included in the above-mentioned knowledge information, a comparison between the current date and time and the date and time included in the above-mentioned knowledge information, and a comparison between the position of the above-mentioned vehicle and the interruption position included in the above-mentioned knowledge information.

7. A method for producing a database, receiving information of an interruption position, which is a position where the automatic driving is interrupted, from an onboard device mounted on the first vehicle, receiving, from an on-vehicle device mounted on a second vehicle, manual driving information as information on manual driving of the second vehicle in the area of ​​the interruption position; extracting a control point as a location for changing the traveling direction from the manual driving information in the area of ​​the interruption position; Creating knowledge information including the vehicle's travel information on a trajectory passing through the control point and capable of generating a trajectory for changing the vehicle's travel direction; At least one of the knowledge information and information indicating a trajectory of the vehicle generated using the knowledge information is stored in a database.

8. The method for producing a database according to claim 7, An autonomous driving control point for controlling the autonomous driving is created, and knowledge information capable of generating a trajectory for changing the traveling direction of the vehicle at the autonomous driving control point is created.

9. The method for producing a database according to claim 7, The production unit that produces the knowledge information can operate in at least two modes, namely, a first mode and a second mode; The production unit uses the control point as the automatic driving control point in the first mode, and moves the control point forward by a first distance relative to the traveling direction to produce the automatic driving control point in the second mode.

10. The method for producing a database according to claim 9, The first distance is determined based on the legal speed limit at the control point and the number of required lane changes.

11. The method for producing a database according to claim 7, The control point is either a point where the change of the traveling direction is started or a point where the trajectory of the vehicle intersects the vehicle marking line.

12. The method for producing a database according to claim 7, further receiving information on vehicle type and date and time from the first vehicle, the second vehicle, and the vehicle-mounted device of the vehicle; The knowledge information includes information on the vehicle types, date and time of the first vehicle and the second vehicle, and the interruption position; Based on at least one of a comparison between the vehicle type of the above-mentioned vehicle and the vehicle type included in the above-mentioned knowledge information, a comparison between the current date and time and the date and time included in the above-mentioned knowledge information, and a comparison between the position of the above-mentioned vehicle and the interruption position included in the above-mentioned knowledge information, the vehicle-mounted device to which the above-mentioned knowledge information is to be sent is determined.

Images