Map processing device and map processing method

Abstract

A map processing device according to the present invention comprises a map data read-ahead unit that: can acquire a plurality of types of map data including lane connection data indicating connection information about lane sections in an extension direction of a lane and lane group connection data indicating connection information about lane group sections in an extension direction of a lane group composed of one or more lanes; and acquires map data of a type corresponding to a condition of a prescribed spot when acquiring the map data of the prescribed spot ahead of a host vehicle from the plurality of types of map data.

Description

Map processing device and map processing method

[0001] The present invention relates to a map processing device and a map processing method.

[0002] Conventionally, a technique for pre-reading map data has been disclosed in a map creation application used in an electronic map display system or the like (see, for example, Patent Document 1). Patent Document 1 discloses a technique for selecting, from all available map data, map data tiles for an area including a first route and an area including a secondary route (such as a return route) selected based on the first route, and extracting map data. In the map data pre-reading technique disclosed in Patent Document 1, map data for a secondary route is pre-read based on the first route, but the amount of accessed map data is adjusted based on the priority of the secondary route.

[0003] Special Publication No. 2015-501956

[0004] As described above, technologies for pre-reading map data have been proposed for map processing devices in the past. However, in this technical field, there is a need to reduce the storage capacity required to store map data within the device, thereby lowering costs. To solve this problem, it is necessary to reduce the amount of map data stored within the map processing device. However, if the amount of stored map data is reduced too much, the performance of the map processing device may be reduced. Therefore, in the technical field of map processing devices, there is a need to develop technology that can achieve both a reduction in the storage capacity required to store map data within the device and maintaining the performance of the device.

[0005] In particular, map processing devices that enable lane-level driving assistance, which is necessary for automated driving of vehicles and has been developed in recent years, use map data with higher detail than conventional car navigation systems that can be used for road-level driving assistance. Therefore, map processing devices that handle such high-detail map data require a large amount of map data to be stored internally, which is why the above-mentioned demand is strong. For example, Patent Document 1 discloses a technology for predicting alternative routes (secondary routes) that are predicted based on the vehicle's position and route. However, on expressways, secondary routes tend to be long. Therefore, even if the technology disclosed in Patent Document 1 is applied to a map processing device that handles high-detail map data, it is difficult to predict all of the map data for secondary routes required for driving assistance.

[0006] The present invention has been made to meet the above-mentioned demands, and an object of the present invention is to provide a technology for a map processing device used for vehicle driving assistance that can achieve both reduction in the storage capacity for holding map data within the device and maintenance of device performance.

[0007] In order to solve the above problems, the map processing device of the present invention includes a map data pre-fetching unit capable of acquiring multiple types of map data. The multiple types of map data include lane connection data indicating connection information between lane sections in the direction in which the lanes extend, and lane group connection data indicating connection information between lane group sections in the direction in which a lane group made up of one or more lanes extends. When acquiring map data for a predetermined point ahead of the vehicle from the multiple types of map data, the map data pre-fetching unit acquires map data of a type that corresponds to the conditions of the predetermined point.

[0008] In order to solve the above-mentioned problem, the map processing method of the present invention includes a map data pre-reading unit of the map processing device of the present invention acquiring map data of a type corresponding to the conditions of the specified point when acquiring map data of a specified point ahead of the vehicle from multiple types of map data.

[0009] According to the present invention having the above configuration, in a map processing device used for vehicle driving assistance, it is possible to achieve both a reduction in the storage capacity for holding map data within the device and maintenance of the device's performance.

[0010] FIG. 1 is a schematic diagram of an in-vehicle system including a map processing device according to an embodiment of the present invention. FIG. 2 is a functional block diagram of a map processing device according to an embodiment of the present invention. FIG. 3 is a diagram illustrating the relationship between various map data stored in a memory unit of the map processing device according to an embodiment of the present invention and various applications that use the various map data. FIG. 4 is a hardware configuration diagram of a map processing device according to an embodiment of the present invention. FIG. 5 is a schematic diagram illustrating the relationship between various map data pre-read in the map processing device according to an embodiment of the present invention and the distance from the vehicle position to a pre-read target point. FIG. 6 is a flowchart illustrating the procedure for pre-reading various map data by the map processing device according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating the procedure for pre-reading various map data by the map processing device according to an embodiment of the present invention. FIG. 8 is a diagram illustrating an outline of a first route change operation example when a route deviation occurs, which is performed by the map processing device according to an embodiment of the present invention. FIG. 9 is a diagram illustrating an operation flow of the first route change operation example when a route deviation occurs, which is performed by the map processing device according to an embodiment of the present invention. FIG. 10 is a diagram illustrating an outline of a second route change operation example when a route deviation occurs, which is performed by the map processing device according to an embodiment of the present invention. FIG. 11 is a diagram illustrating an operation flow of the second route change operation when a route deviation occurs, which is performed by the map processing device according to an embodiment of the present invention.

[0011] A map processing device and a map processing method (map data prefetching method) according to an embodiment of the present invention will be specifically described below with reference to the drawings. The present invention is applicable to, for example, a vehicle control computing device (e.g., a map positioning unit (MPU)) with which an on-board ECU (Electronic Control Unit) for an Advanced Driver Assistance System (ADAS) or Autonomous Driving (AD) can communicate.

[0012] [Configuration of In-Vehicle System] Fig. 1 is a schematic diagram of an in-vehicle system including a map processing device according to an embodiment of the present invention. Note that Fig. 1 only shows components related to various processes performed by the map processing device.

[0013] 1, an in-vehicle system 1 of a vehicle equipped with an automatic driving function includes a map processing device 2, a car navigation system 3 (hereinafter referred to as "car navigation 3"), and an automatic driving control device 4, and these components are connected to each other within the in-vehicle system 1. The map processing device 2 is also connected via a communication network 6 to a map distribution server 5 that is provided externally, for example, in a cloud.

[0014] The map distribution server 5 stores various types of highly detailed map data at the lane level (e.g., lane connection data, lane attribute data, lane boundary data, etc., which will be described later; hereinafter, these data will be referred to as "high-precision map data") that are required for autonomous driving assistance of the host vehicle. During operation, the map processing device 2 obtains various types of high-precision map data of the area around the host vehicle on demand from the map distribution server 5. Although not shown in the figure, the car navigation system 3 is connected to a map distribution server for the car navigation system that is provided externally, for example, in the cloud, via a communication network, and obtains various types of map data of the area around the host vehicle at the road level from the map distribution server for the car navigation system on demand.

[0015] The map processing device 2 has a function of acquiring in advance (hereinafter referred to as "pre-fetching") various high-precision map data of a recommended route to a destination (hereinafter also referred to as a "planned driving route") and surrounding routes required for autonomous driving assistance of the host vehicle from the map distribution server 5. Note that the "surrounding routes of the planned driving route" here refers to routes that can branch off from the planned driving route. The map processing device 2 then searches for the planned driving route using the various high-precision map data that have been pre-fetched, and outputs information about the planned driving route determined by the search process to the autonomous driving control device 4. In this embodiment, as will be described later, when the host vehicle deviates from the planned driving route, the map processing device 2 also searches for and determines a new planned driving route using the various high-precision map data that have been pre-fetched. Note that the internal configuration and processing content of the map processing device 2 will be described later with reference to the drawings.

[0016] The car navigation system 3 searches for a recommended route to the destination using various road-level map data of the area around the vehicle acquired from a map distribution server for car navigation systems, and sets the recommended route obtained by the search process as the planned driving route. The car navigation system 3 also outputs various information, such as the planned driving route, destination, and route search conditions set in the car navigation system 3, to the map processing device 2.

[0017] The automatic driving control device 4 performs automatic driving control of the vehicle using information on the planned driving route to the destination set by the map processing device 2 or the car navigation system 3.

[0018] [Configuration of Map Processing Device] Fig. 2 is a configuration diagram of functional blocks of a map processing device 2 according to one embodiment of the present invention. Note that Fig. 2 only shows components involved in the process of pre-reading various high-precision geographic data by the map processing device 2 and the process of searching for and determining a planned driving route to a destination.

[0019] 2, the map processing device 2 includes a control unit 10, a storage unit 11, a sensor unit 12, a first communication unit 13, a second communication unit 14, and a third communication unit 15. The control unit 10 is connected to the storage unit 11, the sensor unit 12, the first communication unit 13, and the second communication unit 14, and the storage unit 11 is connected to the third communication unit 15.

[0020] The control unit 10 searches for a recommended route to the destination at the lane level using various data (information) acquired from the storage unit 11, the sensor unit 12, and the first communication unit 13, and outputs information on the planned driving route obtained as a result of the search to the automatic driving control device 4 via the second communication unit 14. Furthermore, as will be described later, when the vehicle deviates from the planned driving route, the control unit 10 also searches for and determines a new planned driving route. The internal configuration of the control unit 10 will be described later.

[0021] The storage unit 11 is connected to an external map distribution server 5 via a third communication unit. The storage unit 11 stores various high-precision map data that enables lane-level automated driving assistance distributed from the map distribution server 5. In this embodiment, it is assumed that the map distribution server 5 stores various high-precision map data that enables lane-level automated driving assistance not only for expressways but also for ordinary roads.

[0022] The storage unit 11 also includes a lane connection and attribute data storage unit 31, a lane boundary data storage unit 32, a lane shape data storage unit 33, and a lane group connection data storage unit .

[0023] The lane connection and attribute data storage unit 31 stores information (hereinafter referred to as "lane connection data") regarding the connection relationships between lane sections that are connected to each other via branch points (e.g., intersections, etc., hereinafter referred to as "nodes") where the structure of the travel lane changes physically. Note that the lane connection data includes, for example, not only connection information between lane sections (between nodes) in the direction in which the lane extends, but also various other information such as information regarding the position coordinates of the nodes, lane identification information, travel direction, and information regarding whether or not the lane can be changed at a node.

[0024] The lane connection and attribute data storage unit 31 also stores information (hereinafter referred to as "lane attribute data") relating to the attributes (properties and characteristics) of each lane between nodes. The lane attribute data includes various types of information between nodes, such as lane type information, lane width, lane curvature, lane gradient type, and information indicating whether the lane is passable. The information indicating whether the lane is passable, which is included in the lane attribute data, includes real-time or time-bound traffic information such as congestion information and construction information.

[0025] Information relating to the boundaries between adjacent lanes in a direction perpendicular to the direction in which the lanes extend (hereinafter referred to as "lane boundary data") is stored in the lane boundary data storage unit 32. The lane boundary data includes information such as the type of boundary line between the lanes (e.g., white line, orange line, etc.) and the pattern type of the boundary line.

[0026] The lane shape data storage unit 33 stores information about the shape of each lane between nodes (hereinafter referred to as "lane shape data"). In this embodiment, the lane shape data includes separate data on the shape of the center line of each lane (hereinafter referred to as "lane center line shape data") and data on the shape of the boundary of each lane (hereinafter referred to as "lane boundary shape data"). Both the lane center line shape data and the lane boundary shape data are composed of coordinate point sequence data.

[0027] Furthermore, in the autonomous driving assistance provided by the map processing device 2, connection information is used that aggregates multiple lanes existing between nodes into one group (hereinafter referred to as a "lane group"). In this embodiment, a lane group is also set for a driving route where there is only one lane between nodes (for example, an ordinary road in FIGS. 8 to 10 described below). That is, in this embodiment, a lane group is made up of one or more lanes. The lane group connection data storage unit 34 stores information regarding the connections between lane groups that are connected to each other via nodes (hereinafter referred to as "lane group connection data"). The lane group connection data is only required to include the minimum connection information between lane groups, and in addition to the connection information between lane groups (between nodes) in the extension direction of the lane groups, various other information, such as lane group identification information, is also included.

[0028] The sensor unit 12 has various devices for recognizing structures and vehicles around the vehicle, the position of the vehicle, the driving conditions of the vehicle, etc. Specifically, the sensor unit 12 includes various devices such as a camera (photographing device) capable of photographing the conditions around the vehicle, a GPS (Global Positioning System) module capable of measuring the position of the vehicle at road level, an acceleration sensor and an angular velocity sensor capable of measuring the driving conditions of the vehicle, etc. The various pieces of information acquired by the sensor unit 12 are input to the control unit 10, and the control unit 10 performs a process of recognizing structures and vehicles around the vehicle and a process of estimating the vehicle's position based on this information.

[0029] The first communication unit 13 is connected to the car navigation system 3. Various information set in the car navigation system 3, such as a planned driving route, a destination, and route search conditions, is transmitted to the control unit 10 via the first communication unit 13. The second communication unit 14 is connected to the automatic driving control device 4. Information on a planned driving route to the destination determined by the control unit 10 is transmitted to the automatic driving control device 4 via the second communication unit 14. The third communication unit 15 is connected to an external map distribution server 5 via the communication network 6, receives various high-precision map data distributed from the map distribution server 5, and outputs the received various high-precision map data to the storage unit 11. The operations of the first communication unit 13 to the third communication unit 15 are controlled by the control unit 10.

[0030] [Configuration of control unit] As shown in FIG. 2 , the control unit 10 functionally includes a surroundings recognition unit 20, a self-position estimation unit 21, a candidate route search unit 22 (route search unit), a route selection unit 23 (route determination unit), a lane data pre-reading unit 24 (map data pre-reading unit), and a map access unit 25.

[0031] The functional processing connections between the functional blocks are as follows: The surroundings recognition unit 20 is connected to the sensor unit 12, the self-position estimation unit 21, and the lane data pre-reading unit 24. The self-position estimation unit 21 is connected to the sensor unit 12, the candidate route search unit 22, and the lane data pre-reading unit 24. The candidate route search unit 22 is connected to the first communication unit 13, the route selection unit 23, and the lane data pre-reading unit 24. The route selection unit 23 is connected to the second communication unit 14 and the map access unit 25. The lane data pre-reading unit 24 is connected to the map access unit 25. Furthermore, the map access unit 25 is connected to the lane connection and attribute data storage unit 31, the lane boundary data storage unit 32, the lane shape data storage unit 33, and the lane group connection data storage unit 34 in the storage unit 11, respectively.

[0032] The periphery recognition unit 20 recognizes the presence and position of structures and vehicles in the vicinity ahead of the vehicle based on an image of the vicinity ahead of the vehicle input from a camera (not shown) in the sensor unit 12 and various high-precision map data input from the lane data pre-reading unit 24. The periphery recognition unit 20 then outputs the recognition results of the structures and vehicles in the vicinity of the vehicle to the self-position estimation unit 21 and the lane data pre-reading unit 24.

[0033] The self-position estimation unit 21 estimates the vehicle's position at the lane level based on the recognition results of structures and vehicles around the vehicle input from the surrounding recognition unit 20, various sensor information input from the sensor unit 12, and various high-precision map data input from the lane data pre-reading unit 24.

[0034] In this embodiment, two types of applications are installed as applications for estimating the vehicle position by the self-position estimation unit 21. Specifically, an application that estimates the vehicle position using an image of the surroundings ahead of the vehicle and high-precision map data, and an application that estimates the vehicle position using high-precision map data are installed. Hereinafter, the former application will be referred to as a "high-precision locator (camera recognition)" and the latter application will be referred to as a "high-precision locator (map matching)."

[0035] In this embodiment, the type of high-precision map data used in the vehicle position estimation process by the self-position estimation unit 21 also varies depending on the vehicle position estimation application being used. Fig. 3 is a diagram showing the relationship between the type of vehicle position estimation application and the type of high-precision map data being used, with the high-precision map data being indicated by circles. Fig. 3 also shows the high-precision map data used in the candidate route search process (described later) by the candidate route search unit 22 and the planned driving route determination process (described later) by the route selection unit 23 (see the "lane level search" column in the figure). In the figure, the high-precision map data used in the candidate route search process (described later) by the candidate route search unit 22 are indicated by circles, and the high-precision map data used in the planned driving route determination process (described later) by the route selection unit 23 are indicated by triangles.

[0036] As shown in FIG. 3 , the high-precision map data used by the high-precision locator (camera recognition) includes lane connection data, lane attribute data, lane boundary data, lane centerline shape data, lane boundary shape data, and lane group connection data. That is, the high-precision locator (camera recognition) uses all types of high-precision map data stored in the storage unit 11. On the other hand, the high-precision map data used by the high-precision locator (map matching) includes lane connection data, lane attribute data, lane centerline shape data, and lane group connection data. That is, the high-precision map data used by the high-precision locator (map matching) does not include high-precision map data related to lane boundaries.

[0037] Then, the self-position estimation unit 21 outputs information on the vehicle position estimated by the high-precision locator (camera recognition) or the high-precision locator (map matching) to the candidate route search unit 22 and the lane data pre-read unit 24 .

[0038] The candidate route searching unit 22 searches for candidates (hereinafter referred to as "candidate routes") that could be recommended routes to a set destination. The candidate route searching unit 22 searches for multiple candidate routes based on the vehicle position input from the self-position estimating unit 21, various high-precision map data input from the lane data pre-reading unit 24, and destination information (such as location information) input from the car navigation system 3. As will be described later, when the vehicle deviates from the set planned driving route, the candidate route searching unit 22 performs a candidate route search process using only lane group connection data as high-precision map data (see the circle in the "Lane Level Search" column in FIG. 3 ). The candidate route searching unit 22 then outputs information on the multiple candidate routes obtained by the search process to the route selecting unit 23.

[0039] The route selection unit 23 selects (determines) a planned driving route from among the multiple candidate routes input from the candidate route search unit 22 based on the lane connection data and lane attribute data input via the map access unit 25 (see the triangles in the "Lane Level Search" column in FIG. 3 ). In the planned driving route selection process by the route selection unit 23, candidate routes are selected from the multiple candidate routes in a predetermined order, and based on the lane attribute data of the selected candidate route, it determines whether the candidate route is actually suitable for automated driving. If the selected candidate route is suitable for automated driving (a drivable route), the route selection unit 23 determines the selected candidate route as the planned driving route. The route selection unit 23 then transmits information about the selected planned driving route (the "confirmed route" in FIG. 3 ) to the automatic driving control device 4 via the second communication unit 14.

[0040] In addition, in the process of selecting a planned driving route by the route selection unit 23, one of the following multiple types of candidate route pick-up orders (selection orders) is set in advance by the driver or the like: (1) in order of least amount of high-precision map data read (2) in order of shortest distance of candidate route (3) in order of shortest required time to the destination (4) in order taking into consideration the search conditions of the initial planned driving route set by the car navigation system 3 or the like.

[0041] In the above-described picking order (1), since the capacity of the map data to be loaded is not known in advance, candidate routes are picked in order of the number of lane groups (nodes) to be newly loaded. In the above-described picking order (2), the lane group connection data must include information on the distance between the lane groups (distance between nodes). In the above-described picking order (3), the lane group connection data must include information on the average speed and must be linked to traffic information. In the above-described picking order (4), the picking order of candidate routes is set according to the search conditions of the car navigation system 3, such as the use of expressways and arrival times, acquired via the first communication unit 13, etc. However, in the process of selecting the planned driving route by the route selection unit 23, candidate routes for which the distance to the next node (branch) of a lane group or lane is too short to load high-precision map data in time are excluded.

[0042] The lane data prefetching unit 24 prefetches (acquires) from the map distribution server 5 multiple types of high-precision map data for the planned driving route ahead of the vehicle and its surrounding routes, which are necessary for smooth autonomous driving assistance for the vehicle. Specifically, the lane data prefetching unit 24 selects tiles (hereinafter referred to as "map tiles") including the area of ​​the planned driving route ahead of the vehicle and / or its surrounding routes from a map divided into tiles, and prefetches various high-precision map data for the roads included in the map tiles. In this case, the type of high-precision map data prefetched varies depending on the conditions of a prefetch target point (predetermined point) ahead of the vehicle. Specifically, the type of high-precision map data prefetched varies depending on conditions such as the distance on the driving route from the vehicle position to the prefetch target point, whether the prefetch target point is on the planned driving route, and the likelihood of deviation from the planned driving route (e.g., the likelihood of becoming a secondary route). The specific content of the prefetching process performed by the lane data prefetching unit 24 will be described in detail later with reference to the drawings.

[0043] The map access unit 25 accesses the storage unit 11 to acquire various types of high-precision map data stored in the storage unit 11, and outputs the acquired various types of high-precision map data to the lane data pre-reading unit 24 and the route selection unit 23. Note that the map access unit 25 outputs high-precision map data of a type according to the conditions of the above-mentioned pre-read target points to the lane data pre-reading unit 24, and the map access unit 25 outputs lane connection data and lane attribute data of a candidate route to the route selection unit 23. Note that, as described above, in this embodiment, the route selection unit 23 determines a planned driving route based on lane attribute data, and therefore the high-precision map data input from the map access unit 25 to the route selection unit 23 may be only lane attribute data of a candidate route.

[0044] [Hardware Configuration of Map Processing Device] The map processing device 2 of this embodiment can be configured by an arithmetic processing device such as a computer device having an arithmetic function and a communication function. Fig. 4 is a block diagram showing an example of the hardware configuration of an arithmetic processing device 100 that can be used as the map processing device 2.

[0045] The arithmetic processing device 100 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, and a RAM (Random Access Memory) 103, all connected to a bus line 108. The arithmetic processing device 100 also includes a network I / F (Interface) 104, an operation unit 105, a display unit 106, and a non-volatile storage 107, all connected to the bus line 108. Although not shown in FIG. 4 , the arithmetic processing device 100 also includes various interfaces used when executing input / output processing of various data (various information) with external devices. Furthermore, although not shown in FIG. 4 , the arithmetic processing device 100 also includes a component corresponding to the sensor unit 12 in FIG. 2.

[0046] The CPU 101 reads out program code of software for realizing various processing functions of the map processing device 2 from the ROM 102 to the RAM 103 and executes it. At this time, variables and parameters generated during the calculation process are also temporarily written to the RAM 103. In other words, the control unit 10 of the map processing device 2 in FIG. 2 is included in the CPU 101.

[0047] The network I / F 104 is configured by, for example, a network interface card (NIC) or the like, and transmits and receives various data to and from each device connected thereto.

[0048] The operation unit 105 is composed of, for example, keys and buttons, and generates an operation signal according to the operation content input by the operator and supplies the operation signal to the CPU 101. For example, the order in which candidate routes are picked up in the process of selecting a planned driving route by the route selection unit 23 can be set by operating the operation unit 105. Such an operation may be performed via an operation unit (not shown) included in the car navigation system 3, in which case the map processing device 2 does not include the operation unit 105.

[0049] The display unit 106 is configured, for example, as a liquid crystal panel, and displays characters, images, etc. on the screen. The display unit 106 may also be configured as a touch panel, in which case the display unit 106 and the operation unit 105 are integrated. Note that the output information from the map processing device 2 may be displayed on a display unit (not shown) provided for displaying various information output from, for example, the car navigation system 3, in which case the map processing device 2 does not have the display unit 106.

[0050] The non-volatile storage 107 can be configured, for example, with a hard disk drive (HDD), a solid state drive (SSD), a flexible disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, or a non-volatile memory. The non-volatile storage 107 stores an operating system (OS), various parameters, and programs for causing the arithmetic processing device 100 to function as the map processing device 2. In this embodiment, the various high-precision map data described above is stored in the non-volatile storage 107, and the storage unit 11 of the map processing device 2 in FIG. 2 is included in the non-volatile storage 107. Note that information (data) such as programs, tables, and files that realize the various functions of the map processing device 2 may be stored in a recording medium other than the ROM 102 or the non-volatile storage 107, such as an IC card, an SD card, or a DVD.

[0051] [Outline of High-Precision Map Data Read-Ahead Function] In this embodiment, the lane data read-ahead unit 24 is provided with a high-precision map data read-ahead function, as described in the following (A) to (C). (A) When the read-ahead target point is a point on the planned driving route, the lane data read-ahead unit 24 changes the type of high-precision map data to read ahead (acquire) depending on the distance on the driving route from the vehicle's position to the read-ahead target point. (B) When the read-ahead target point is a point on a peripheral route that can branch off from the planned driving route, the lane data read-ahead unit 24 reads ahead only lane group connection data. (C) For a point on a route that can deviate from the planned driving route and is within a predetermined distance from the deviation point, the lane data read-ahead unit 24 reads ahead not only lane group connection data but also various other high-precision map data necessary for autonomous driving assistance, such as lane connection data. In this case, the high-precision map data to be read ahead may be, for example, all types of high-precision map data shown in FIG. 3 , or high-precision map data other than high-precision map data related to lane boundaries.

[0052] Although the map processing device 2 of this embodiment will be described as having all of the above-described look-ahead functions (A) to (C), the present invention is not limited thereto. For example, the look-ahead function (C) (hereinafter referred to as the "deviation prediction function") may not be provided in some configurations depending on, for example, the processing performance, memory capacity, etc. of the map processing device 2. Alternatively, for example, a configuration may be provided with only one of the look-ahead functions (A) and (B). When only the look-ahead function (A) is provided, the type of high-precision map data read ahead at a look-ahead target point on a peripheral route also changes depending on the distance on the driving route from the vehicle position to the look-ahead target point. On the other hand, when only the look-ahead function (B) is provided, at a look-ahead target point on a planned driving route, not only lane group connection data but also various high-precision map data necessary for autonomous driving assistance, such as lane connection data, are read ahead, regardless of the distance from the vehicle position.

[0053] Here, the contents of the look-ahead function (A) above will be specifically described. Fig. 5 is a schematic diagram showing the relationship between the type of high-precision map data read ahead by the lane data look-ahead unit 24 and the distance on the travel route from the vehicle position to the look-ahead target point in the look-ahead function (A) above (when the look-ahead target point is a point on the planned travel route). Note that Fig. 5 illustrates an example of a road with two lanes on each side. Therefore, in the example shown in Fig. 5, four lanes are aggregated to form one lane group.

[0054] In the example shown in FIG. 5 , one lane group area 40, i.e., one lane group section, is represented by a substantially rectangular block. Lane group connection data 41 is represented by nodes (branching points) marked with open circles and provided at the entry and exit points of each lane group area 40. Lane connection data 42 is represented by nodes (branching points) marked with filled circles and provided at the entry and exit points of corresponding lanes within each lane group area 40. Lane attribute and shape data 43 (lane attribute data and lane centerline shape data) are represented by solid lines (links) connecting nodes in the lane connection data 42. Map data 44 related to lane boundaries (lane boundary data and lane boundary shape data) is represented by a pattern (white line pattern) of open rectangles arranged at predetermined intervals along the extension direction of the road. In addition to the white line pattern shown in the figure, the map data 44 related to lane boundaries may also include, for example, lines indicating boundaries with sidewalls and data on the sidewalls themselves.

[0055] If the look-ahead target point is a point 51 located close to the vehicle position (e.g., several kilometers ahead), as shown in FIG. 5 , the lane data look-ahead unit 24 acquires various high-precision map data used in the process of estimating the vehicle position using a high-precision locator (camera recognition). Specifically, in look-ahead for the close-up point 51, the lane data look-ahead unit 24 looks ahead (acquires) lane group connection data 41, lane connection data 42, lane attribute / shape data 43, and map data 44 related to lane boundaries. That is, in look-ahead for the close-up point 51, the lane data look-ahead unit 24 acquires all types of high-precision map data shown in FIG. 3 . In addition, in look-ahead for the close-up point 51, the lane data look-ahead unit 24 also acquires an image of the surroundings ahead of the vehicle captured by the camera in the sensor unit 12, thereby acquiring information about physical structures 45 and vehicles present in the surroundings ahead of the vehicle.

[0056] 5 shows an example in which, in pre-reading for a nearby point 51, the lane data pre-reading unit 24 pre-reads (acquires) not only high-precision map data for two lanes in the vehicle's traveling direction but also high-precision map data for two lanes in the opposite traveling direction. However, the present invention is not limited to this example. When pre-reading for a nearby point 51, for example, if a physical structure 45 is present in the vicinity ahead of the vehicle, the lane data pre-reading unit 24 may pre-read high-precision map data only for the range visible from the vehicle (for example, the range shown in the peripheral image). For example, if a median strip is provided as the physical structure 45 between the two lanes in the vehicle's traveling direction and the opposite lane, and the opposite lane is not visible from the vehicle, the lane data pre-reading unit 24 may pre-read high-precision map data only for the two lanes in the vehicle's traveling direction. In this case, there is no need to read high-precision map data for the opposite lane, and the amount of high-precision map data to be acquired can be further reduced.

[0057] If the look-ahead target point is a point 52 that is a medium distance from the vehicle position, the lane data look-ahead unit 24 acquires various high-precision map data to be used in the process of estimating the vehicle position using a high-precision locator (map matching). Specifically, in look-ahead for the medium-distance point 52, the lane data look-ahead unit 24 looks ahead (acquires) lane group connection data 41, lane connection data 42, and lane attribute / shape data 43. Note that in look-ahead for the medium-distance point 52, the lane data look-ahead unit 24 acquires only high-precision map data for two lanes in the traveling direction of the vehicle.

[0058] Furthermore, if the look-ahead target point is point 53, which is far from the vehicle position, the lane data look-ahead unit 24 looks ahead (acquires) only the lane group connection data 41. If the look-ahead target point is point 54, which is even farther away than point 53, the lane data look-ahead unit 24 does not look ahead (acquire) high-precision map data. Note that the boundary values ​​of the above-mentioned "short distance," "medium distance," and "long distance," as well as the boundary values ​​of distances for which high-precision map data is not looked ahead (first predetermined distance D1 to third predetermined distance D3 in FIG. 7 , which will be described later), can be set appropriately depending on, for example, the type of driving route (general road, expressway, etc.), the processing performance, memory capacity, etc. of the map processing device 2.

[0059] [High-precision map data pre-reading process by map processing device] Next, a description will be given of the processing flow of high-precision map data pre-reading process executed by the lane data pre-reading unit 24 of the map processing device 2. Note that the pre-reading process by the lane data pre-reading unit 24 described below is controlled by the CPU 101 in FIG.

[0060] The look-ahead processing by the lane data look-ahead unit 24 described below will cover the look-ahead processing when the look-ahead target point is on the planned driving route and when it is on a peripheral route of the planned driving route, i.e., the look-ahead functions (A) and (B) described above. The deviation prediction function (C) described above, which takes into account the possibility of deviation from the planned driving route, will be outlined as appropriate in the description of the processing flow below.

[0061] <Overall Processing Flow of High-Precision Map Data Pre-reading Process> First, the overall processing flow of high-precision map data pre-reading process executed by the lane data pre-reading unit 24 will be described with reference to Fig. 6. Fig. 6 is a flowchart showing the overall processing procedure of high-precision map data pre-reading process executed by the lane data pre-reading unit 24. The overall high-precision map data pre-reading process shown in Fig. 6 starts when the map processing device 2 is started (powered on).

[0062] First, the lane data read-ahead unit 24 determines whether there is available time to perform the map data read-ahead process (S1). In this process, the lane data read-ahead unit 24 determines, based on the current processing load of the CPU 101, whether the CPU 101 has time available to perform the map data read-ahead process, i.e., whether the CPU 101 has sufficient processing load to perform the map data read-ahead process. For example, if the vehicle deviates from the planned travel route and a search process for a new route to the destination (e.g., the operations described in FIGS. 8 to 11 below) is being performed, the processing load on the CPU 101 for the route search process increases, and the CPU 101 no longer has sufficient time to perform the read-ahead process. Therefore, in this situation, there is no available time to perform the read-ahead process, the determination in S1 is NO, and the processes (read-ahead processes) from S3 onward (described below) are not performed.

[0063] If the lane data pre-reading unit 24 determines in S1 that there is no available time for performing the pre-reading process (if S1 is determined to be NO), the lane data pre-reading unit 24 performs a standby process for a certain period of time (S2). After the process of S2, the lane data pre-reading unit 24 returns the process to S1 and repeats the processes from S1 onwards.

[0064] On the other hand, if the lane data read-ahead unit 24 determines in S1 that there is free time for performing the read-ahead process (if S1 is determined as YES), the lane data read-ahead unit 24 determines whether or not to perform the read-ahead process on the planned driving route (S3). In this process, if there are any points on the planned driving route to the destination at the time of this process for which the read-ahead process has not been performed, the lane data read-ahead unit 24 determines to perform the read-ahead process on the planned driving route (determined as YES). On the other hand, if there are any points on the planned driving route to the destination at the time of this process for which the read-ahead process has not been performed, the lane data read-ahead unit 24 determines not to perform the read-ahead process on the planned driving route (determined as NO).

[0065] If the lane data pre-reading unit 24 determines in S3 that a pre-reading process on the planned driving route should be performed (if S3 is determined to be YES), the lane data pre-reading unit 24 acquires map tiles including points on the planned driving route for which high-precision map data has not yet been acquired (S4). That is, in this process, the lane data pre-reading unit 24 acquires map tiles including pre-reading target points on the planned driving route for which a pre-reading process has not yet been performed.

[0066] Next, the lane data pre-reading unit 24 performs a high-precision map data pre-reading execution process (S5). In this process, the lane data pre-reading unit 24 pre-reads (acquires) from the map distribution server 5 high-precision map data for each pre-read target point on the planned driving route that is included in the map tile acquired in the process of S4. At this time, as described with reference to FIG. 5, the type of high-precision map data pre-read by the lane data pre-reading unit 24 differs depending on the distance on the driving route from the vehicle position to the pre-read target point. Note that the map data pre-reading execution process of S5 will be described in detail later with reference to FIG. 7.

[0067] On the other hand, if the lane data pre-read unit 24 determines in S3 not to perform pre-read processing on the planned driving route (if S3 is determined to be NO), the lane data pre-read unit 24 acquires map tiles including points on the peripheral route of the planned driving route for which high-precision map data has not yet been acquired (S6). That is, in this processing, the lane data pre-read unit 24 acquires map tiles including pre-read target points on the peripheral route of the planned driving route for which pre-read processing has not yet been performed. In this case, the lane data pre-read unit 24 acquires the map tile closest to the vehicle position.

[0068] Next, the lane data pre-fetching unit 24 performs a lane group connection data pre-fetching process (S7). In this process, the lane data pre-fetching unit 24 pre-fetches (acquires) from the map distribution server 5 the lane group connection data of each pre-fetch target point on the peripheral route included in the map tile acquired in the process of S6. Note that if a departure prediction function is provided, in the process of S7, not only the lane group connection data but also various high-precision map data necessary for autonomous driving assistance, such as lane connection data, is pre-fetched (acquired) for points on the peripheral route that are within a predetermined distance from the departure point.

[0069] After the processing of S5 or S7, the lane data pre-reading unit 24 determines whether or not to continue the pre-reading of high-precision map data (S8). In this process, the lane data pre-reading unit 24 determines whether or not the power supply of the map processing device 2 has been turned off. If the power supply of the map processing device 2 has not been turned off, the pre-reading of high-precision map data continues, and the determination process of S8 is YES. On the other hand, if the power supply of the map processing device 2 has been turned off, the determination process of S7 is NO.

[0070] If the lane data pre-read unit 24 determines in S8 to continue the pre-reading process of high-precision map data (if the determination in S8 is YES), the lane data pre-read unit 24 returns the process to S1 and repeats the processes from S1 onwards. On the other hand, if the lane data pre-read unit 24 determines in S8 not to continue the pre-reading process of high-precision map data (if the determination in S8 is NO: if the power is turned off), the lane data pre-read unit 24 ends the pre-reading process of high-precision map data.

[0071] <Processing flow of high precision map data pre-reading execution processing (processing of S5 above)> Next, with reference to Fig. 7, a processing flow of the pre-reading execution processing performed in S5 in the overall processing flow of the high precision map data pre-reading processing shown in Fig. 6 will be described. Fig. 7 is a flowchart showing the procedure of the high precision map data pre-reading execution processing performed in S5 above.

[0072] First, the lane data pre-reading unit 24 determines whether the distance d from the vehicle position to a predetermined pre-read target point on the planned driving route included in the acquired map tile is less than a first predetermined distance D1 (S11). Note that here, the "distance d from the vehicle position to the predetermined pre-read target point" refers to the distance on the planned driving route from the vehicle position to the predetermined pre-read target point. Also, the "first predetermined distance D1" here is a threshold value for determining whether the distance d from the vehicle position to the predetermined pre-read target point is a short distance.

[0073] If the lane data pre-reading unit 24 determines in S11 that the distance d from the vehicle position to the predetermined pre-read target point is less than the first predetermined distance D1 (YES in S11), the lane data pre-reading unit 24 performs the process of S14, which will be described later. On the other hand, if the lane data pre-reading unit 24 determines in S11 that the distance d from the vehicle position to the predetermined pre-read target point is not less than the first predetermined distance D1 (NO in S11), the lane data pre-reading unit 24 determines whether the distance d from the vehicle position to the predetermined pre-read target point is less than a second predetermined distance D2 (S12). Note that the "second predetermined distance D2" here is a threshold value for determining whether the distance d from the vehicle position to the predetermined pre-read target point is a medium distance.

[0074] If the lane data pre-read unit 24 determines in S12 that the distance d from the vehicle position to the predetermined pre-read target point is less than the second predetermined distance D2 (if S12 is a YES determination), the lane data pre-read unit 24 performs the process of S15, which will be described later. On the other hand, if the lane data pre-read unit 24 determines in S12 that the distance d from the vehicle position to the predetermined pre-read target point is not less than the second predetermined distance D2 (if S12 is a NO determination), the lane data pre-read unit 24 determines whether the distance d from the vehicle position to the predetermined pre-read target point is less than a third predetermined distance D3 (S13). Note that the "third predetermined distance D3" here is a threshold value for determining whether the distance d from the vehicle position to the predetermined pre-read target point is a long distance.

[0075] If the lane data pre-read unit 24 determines in S13 that the distance d from the vehicle position to the predetermined pre-read target point is less than the third predetermined distance D3 (if S13 is a YES determination), the lane data pre-read unit 24 performs the process of S16, which will be described later. On the other hand, if the lane data pre-read unit 24 determines in S13 that the distance d from the vehicle position to the predetermined pre-read target point is not less than the third predetermined distance D3 (if S13 is a NO determination), the lane data pre-read unit 24 performs the process of S17, which will be described later.

[0076] Returning to the explanation of the processing in S11, if the determination in S11 is YES, i.e., if the distance d from the vehicle position to the predetermined read-ahead target point is short, the lane data read-ahead unit 24 performs a process of read-ahead of lane boundary and boundary shape data (S14). Specifically, the lane data read-ahead unit 24 reads (acquires) the lane boundary data and lane boundary shape data of the predetermined read-ahead target point.

[0077] After the process of S14, or if the determination of S12 is YES (if the distance d from the vehicle position to the predetermined target point is a medium distance), the lane data read-ahead unit 24 performs a process of read-ahead of lane connection, attribute, and center line shape data (S15). Specifically, the lane data read-ahead unit 24 reads (acquires) the lane connection data, lane attribute data, and lane center line shape data of the predetermined target point.

[0078] After processing S15, or if S13 is judged as YES (if the distance d from the vehicle position to the specified pre-read target point is long), the lane data pre-reading unit 24 performs pre-reading processing of the lane group connection data of the specified pre-read target point (S16).

[0079] After the process of S16, or if the determination in S13 is NO, the lane data pre-reading unit 24 determines whether or not all of the pre-reading target points included in the acquired map tile have been selected (S17).

[0080] If the lane data pre-reading unit 24 determines in S17 that all of the pre-read target points have not been selected (if S17 is determined to be NO), the lane data pre-reading unit 24 returns the process to S11 and repeats the processes from S11 onwards. At this time, the lane data pre-reading unit 24 selects a pre-read target point for which pre-reading processing has not been performed as a new predetermined pre-read target point, and repeats the processes from S11 onwards.

[0081] On the other hand, if the lane data pre-reading unit 24 determines in S17 that all pre-read target points have been selected (if S17 is a YES determination), the lane data pre-reading unit 24 terminates the pre-reading execution process and moves the processing to S8 during the pre-reading process (see Figure 6).

[0082] As described above, in the process of look-ahead execution on the planned driving route by the lane data look-ahead unit 24, if the distance d from the vehicle position to the look-ahead target point is short (d<D1: distance within a predetermined range), the look-ahead processing of S14 to S16 is performed, and all types of high-precision map data are read in advance. In the process of look-ahead execution on the planned driving route, if the distance d from the vehicle position to the look-ahead target point is medium (D1≦d<D2), the look-ahead processing of S15 and S16 is performed, and high-precision map data other than data related to lane boundaries is read in advance. In addition, in the process of look-ahead execution on the planned driving route, if the distance d from the vehicle position to the look-ahead target point is long (D2≦d<D3: distance within a specific range), the look-ahead processing of S16 is performed, and only lane group connection data is read in advance. In the above-described look-ahead execution process of this embodiment, even if the target point is on the planned driving route, if the distance d from the vehicle position to the look-ahead target point is equal to or greater than the third predetermined distance D3, the look-ahead process of S14 to S16 is not performed, and high-precision map data is not read in advance (acquired).

[0083] [Route Change Operation by Map Processing Device When Route Deviation Occurs] Next, a description will be given of an operation of changing the route to the destination (planned driving route) when the vehicle deviates from the planned driving route in the map processing device 2 of this embodiment. The route change operation when the vehicle deviates from the route is performed by the candidate route search unit 22 and the route selection unit 23 in the map processing device 2, and the processing operation is controlled by the CPU 101 in FIG. 4.

[0084] 8 is a diagram showing a situation in which a route deviation occurs when the vehicle enters an expressway from an ordinary road, and an overview of a first route change operation performed by the map processing device 2 when the situation occurs. In the first route change operation example, consider the case in which the following route deviation situation occurs.

[0085] First, assume that the planned driving route of the vehicle (hereinafter referred to as the "original route Ro"), which is set in advance by the car navigation system 3, is a route from an entry point INa from a general road to an expressway to a merging point A with the expressway (as indicated by the solid arrow in the figure). Furthermore, assume that before the vehicle reaches the entry point INa, the driver, while driving automatically, sees a display on the general road that displays the expressway congestion status or hears the guidance voice of the car navigation system 3 and realizes that the section of the expressway from the merging point A to the next merging point B is congested. Then, assume that the driver determines that the original route Ro, which involves turning right at the entry point INa, is not a good option and quickly adjusts the steering (switches to manual driving) to continue straight ahead without turning right at the entry point INa. When such a route deviation occurs, after the route deviation, the map processing device 2 searches for a new planned driving route (hereinafter referred to as the "new route Rn") to the destination using the lane group connection data that has been read in advance. In the example shown in FIG. 8, after the route deviation, the return route (indicated by the dashed arrow in the figure) from the entry point INb to the expressway junction B is determined as the new route Rn.

[0086] Here, the first route change operation, i.e., the operation of returning to the expressway, performed by the map processing device 2 when the route deviation situation shown in Fig. 8 occurs will be described in more detail with reference to Fig. 9. Fig. 9 is a diagram showing the flow (situations) of the route change operation performed by the map processing device 2 when the route deviation situation shown in Fig. 8 occurs, and the relationship between the pre-fetched high precision map data and the acquired high precision map data in each situation of the route change operation.

[0087] 9 , as in FIG. 5 , each lane group area is represented by a substantially rectangular block, lane group connection data is represented by a node (branch point) marked with a white circle, and lane connection data is represented by a node marked with a black circle. Also, in FIG. 9 , lane attribute data and lane centerline shape data are represented by links (solid lines) connecting the nodes marked with black circles. Also, in FIG. 9 , map data related to lane boundaries (lane boundary data and lane boundary shape data) is omitted for simplicity's sake. Furthermore, in FIG. 9 , for lane group areas for which lane connection data (nodes marked with black circles) has been acquired (pre-read), the acquired lane group connection data (nodes marked with white circles) is omitted for simplicity's sake. Note that each lane group area shown in FIG. 9 is assumed to exist within a medium distance from the vehicle (see FIG. 5 ).

[0088] 9 shows the situation before the host vehicle deviates from the route. Therefore, in this situation, at least lane group connection data, lane connection data, lane attribute data, and lane centerline shape data are read ahead (acquired) in each lane group area along the original route Ro preset by the car navigation system 3. Note that in lane group areas located close to the host vehicle, lane boundary data, lane boundary shape data, and an image of the surrounding area ahead of the host vehicle captured by the camera in the sensor unit 12 are also acquired.

[0089] Furthermore, in situation (a1), in lane group areas on the peripheral route of the original route Ro, mainly only lane group connection data is pre-read (acquired) (see the nodes marked with open circles in the figure). However, at the expressway entry point INa, the route branches off onto a route different from the original route Ro. Therefore, in situation (a1), in consideration of the possibility of the vehicle deviating onto the different route, lane connection data, lane attribute data, lane centerline shape data, etc. are pre-read in several lane group areas located along the different route from the lane group area including the entry point INa (see "Departure Prediction" in FIG. 9).

[0090] After the situation (a1), in consideration of the congestion on the expressway, the vehicle proceeds straight (deviation) without turning right at the entry point INa, which is the situation shown in situation (b1) in Figure 9. In this case, the map processing device 2 (candidate route search unit 22) of the vehicle performs a search process for a new route Rn to the destination using the lane group connection data that has been read ahead in each lane group area ahead of the route after the deviation. Then, in situation (b1), this search process obtains multiple routes as candidate routes for the new route Rn, including a return route (candidate route Rc) from the entry point INb to the expressway toward the merging point B.

[0091] After the situation (b1), the map processing device 2 (route selection unit 23) of the vehicle selects candidate routes in ascending order of the amount of high-precision map data loaded (in descending order of the number of newly loaded nodes), and determines whether the candidate routes are suitable for autonomous driving of the vehicle. That is, the route selection unit 23 selects candidate routes in descending order of the fastest route to return to the expressway, and determines whether the candidate routes are suitable for autonomous driving. Therefore, in the example shown in FIG. 9 , the map processing device 2 (route selection unit 23) first acquires lane connection data and lane attribute data for each lane group area along the candidate route Rc. This situation is the situation (c1) in FIG. 9 . Therefore, in the situation (c1), the display mode of the acquired data for each lane group area between the entry point INb and the merging point B is represented by nodes marked with black circles and links between the nodes.

[0092] In situation (c1), the map processing device 2 (route selection unit 23) refers to the lane attribute data of each lane group area along the candidate route Rc to determine whether the candidate route Rc is actually suitable for autonomous driving (whether it is drivable). In the example shown in FIG. 9 , the candidate route Rc is determined to be suitable for autonomous driving, and the map processing device 2 (route selection unit 23) determines the candidate route Rc as a new route Rn. Thereafter, the map processing device 2 acquires other high-precision map data on the new route Rn, sequentially starting from the look-ahead target points closest to the vehicle.

[0093] When a route deviation shown in Figures 8 and 9 occurs, the map processing device 2 performs a route change operation as described above. In the first route change operation example, high-precision map data for each lane group area that becomes used or unused as the vehicle travels is deleted (discarded) as appropriate. In the first route change operation example, the original route Ro before the deviation was set by the car navigation system 3, but the present invention is not limited to this. For example, even if the original route Ro before the deviation was set by the map processing device 2 (control unit 10), a new route Rn is determined in the same manner as in the first route change operation example.

[0094] <Second Route Change Operation Example> FIG. 10 is a diagram showing a situation in which route deviation occurs when the vehicle enters an expressway from an ordinary road, and an overview of a second route change operation performed by the map processing device 2 when this situation occurs.

[0095] The route deviation occurrence situation shown in Figure 10 is similar to that explained in Figure 8, and therefore its explanation will be omitted here. The example shown in Figure 10 shows an example in which, after route deviation, a return route to the expressway from entry point INc toward junction C with the expressway is determined as the new route Rn. Note that in the example shown in Figure 10, a return route to the expressway from entry point INb, which is located between entry points INa and INc, toward junction B with the expressway is also determined as candidate route Rc1, but this candidate route Rc1 is shown as an example in which it is not possible to travel.

[0096] Here, with reference to Fig. 11 , the second route change operation performed by the map processing device 2 when the route deviation situation shown in Fig. 10 occurs, i.e., the operation of returning to the route to the expressway, will be described in more detail. Fig. 11 is a diagram showing the flow (situations) of the route change operation performed by the map processing device 2 when the route deviation situation shown in Fig. 10 occurs, and the relationship between the pre-fetched high precision map data and the acquired high precision map data in each situation of the route change operation. Note that the display modes of the lane group areas and various high precision map data shown in Fig. 11 are the same as those described in Fig. 9 , and each lane group area shown in Fig. 11 is also assumed to exist within a medium distance from the host vehicle (see Fig. 5 ).

[0097] 11 shows the situation before the host vehicle deviates from the route. Therefore, in this situation, at least lane group connection data, lane connection data, lane attribute data, and lane centerline shape data are pre-read (acquired) in each lane group area along the original route Ro preset by the car navigation system 3. Note that in lane group areas located close to the host vehicle, lane boundary data, lane boundary shape data, and an image of the surrounding area ahead of the host vehicle captured by the camera in the sensor unit 12 are also acquired.

[0098] Furthermore, in situation (a2), in lane group areas on the peripheral route of the original route Ro, mainly only lane group connection data is pre-read (acquired) (see the nodes marked with open circles in the figure). However, at the expressway entry point INa, the route branches off onto a route different from the original route Ro. Therefore, in situation (a2), in consideration of the possibility of the vehicle deviating onto the different route, lane connection data, lane attribute data, lane centerline shape data, etc. are pre-read in several lane group areas along the different route from the lane group area including the entry point INa (see "Departure Prediction" in FIG. 11).

[0099] After the situation (a2), in consideration of the congestion on the expressway, the vehicle proceeds straight (deviation) without turning right at the entry point INa, which is the situation shown in situation (b2) in Figure 11. In this case, the map processing device 2 (candidate route search unit 22) of the vehicle performs a search process for a new route Rn to the destination using the lane group connection data that has been read ahead in each lane group area ahead of the route after the deviation. Then, in situation (b2), this search process obtains multiple routes as candidate routes for the new route Rn, including a return route (candidate route Rc1) from the entry point INb to the expressway toward the merging point B and a return route (candidate route Rc2) from the entry point INc to the expressway toward the merging point C.

[0100] After situation (b2), the map processing device 2 (route selection unit 23) of the vehicle selects candidate routes in ascending order of the amount of high-precision map data loaded (in descending order of the number of newly loaded nodes), and determines whether or not the candidate routes are suitable for autonomous driving of the vehicle. That is, the route selection unit 23 selects candidate routes in descending order of the fastest route for returning to the expressway, and determines whether or not the candidate routes are suitable for autonomous driving. Therefore, in the example shown in FIG. 11 , the map processing device 2 (route selection unit 23) first acquires lane connection data and lane attribute data for each lane group area along candidate route Rc1. This situation is situation (c2) in FIG. 11 .

[0101] In situation (c2), the map processing device 2 (route selection unit 23) refers to the lane attribute data of each lane group area along the candidate route Rc1 to determine whether the candidate route Rc1 is actually suitable for automated driving (whether it is drivable). In the example shown in FIG. 11 , consider a case where the candidate route Rc1 is not suitable for automated driving of the vehicle due to, for example, construction, traffic congestion, or lane curvature exceeding a threshold value midway through the driving section from the entry point INb to the merging point B. In this case, in situation (c2), the map processing device 2 (route selection unit 23) determines that the candidate route Rc1 is not drivable and cannot be determined as the new route Rn (see the white cross in the figure).

[0102] After situation (c2), the map processing device 2 (route selection unit 23) of the host vehicle acquires lane connection data and lane attribute data for each lane group area along the candidate route Rc2. This situation is shown as situation (d2) in FIG. 11 . Then, in situation (d2), the map processing device 2 (route selection unit 23) refers to the lane attribute data for each lane group area along the candidate route Rc2 and determines whether the candidate route Rc2 is actually suitable for autonomous driving of the host vehicle. The example shown in FIG. 11 considers a case where the candidate route Rc2 is actually suitable for autonomous driving. In this case, in situation (d2), the map processing device 2 (route selection unit 23) determines the candidate route Rc2 as the new route Rn. Thereafter, the map processing device 2 acquires other high-precision map data on the new route Rn, sequentially starting from the look-ahead target points closest to the host vehicle.

[0103] 10 and 11 , when a route deviation occurs, the map processing device 2 performs a route change operation as described above. Note that, in the second route change operation example described above, high-precision map data for each lane group area that becomes used or unused as the vehicle travels is also deleted (discarded) as appropriate. In the second route change operation example described above, the original route Ro before the deviation was set by the car navigation system 3, but the present invention is not limited to this. For example, even if the original route Ro before the deviation was set by the map processing device 2 (control unit 10), a new route Rn is determined in the same manner as in the second route change operation example described above.

[0104] [Various Effects] As described above, when the map processing device 2 of this embodiment pre-reads (acquires) high-precision map data from the map distribution server 5, the type of high-precision map data to be pre-read (acquired) along the planned driving route is changed depending on the distance from the vehicle position to the pre-read target point. For example, as described above, if the distance along the driving route from the vehicle position to the pre-read target point is short, all types of high-precision map data are pre-read, whereas if the distance is long, only lane group connection data is pre-read. Therefore, the map processing device 2 of this embodiment can reduce the amount of high-precision map data stored within the device, thereby reducing the storage capacity for storing map data within the device and lowering costs. Furthermore, in this embodiment, all types of high-precision map data, i.e., various high-precision map data necessary for assisting autonomous driving and the like, are pre-read at pre-read target points located close to the vehicle position, thereby maintaining the performance of the map processing device 2. As described above, the map processing device 2 of this embodiment can achieve both a reduction in the storage capacity for storing map data within the device and maintenance of device performance.

[0105] In the map processing device 2 of this embodiment, when pre-reading (acquiring) map data from the map distribution server 5, if a pre-read target point is not on the planned driving route, only the lane group connection data of the pre-read target point is pre-read. Therefore, in this embodiment, the amount of high-precision map data stored in the device can be further reduced.

[0106] The map processing device 2 of this embodiment acquires a peripheral image ahead of the vehicle for points located close to the vehicle, but does not acquire high-precision map data for lanes that are not shown in the peripheral image. Therefore, in this embodiment, the volume of high-precision map data stored in the device can be further reduced.

[0107] As described above, the map processing device 2 of this embodiment has a route deviation prediction function. Specifically, if there is a branch point on the planned driving route from which deviation may occur, not only lane group connection data but also various high-precision map data necessary for assistance with automated driving, such as lane connection data, are read in advance for points on the route that are within a predetermined distance from the branch point (part of points after the branch point). Therefore, in this embodiment, even if the driving route deviates from the planned driving route, various high-precision map data such as lane connection data has been read in advance for part of the determined new route, so that automated driving on the new route can be resumed more quickly.

[0108] In the map processing device 2 of this embodiment, when the vehicle deviates from the planned driving route during autonomous driving, a new planned driving route is searched for using the lane group connection data that has been read ahead for the route after the deviation. Therefore, in this embodiment, even if the vehicle deviates from the planned driving route during autonomous driving, a new planned driving route can be quickly determined, and autonomous driving can be quickly resumed on the new planned driving route.

[0109] Furthermore, in the map processing device 2 of this embodiment, when multiple candidate routes are found in the process of searching for candidate routes in the event of a route deviation, the map processing device 2 picks up each candidate route in a predetermined order and determines whether the picked candidate route is drivable. In this case, in this embodiment, multiple criteria are provided as criteria for the order in which the settable candidate routes are picked up. Specifically, as described above, the criteria for the order include the order in which the amount of high-precision map data loaded is smallest, the order in which the candidate route distance is shortest, the order in which the required time to reach the destination is shortest, and the order in which the search conditions for the initial planned driving route set in the car navigation system 3 or the like are taken into consideration. In this embodiment, the driver or the like can select a predetermined criterion from among these criteria. Therefore, when such a function is provided, a new route can be determined according to the user's needs and priorities.

[0110] [Various Modifications] In the above embodiment, a configuration example has been described in which the car navigation system 3 is provided separately from the map processing device 2, but the present invention is not limited to this. The car navigation system 3 may be provided with the various functions described above of the map processing device 2. In this case, the car navigation system 3 functions as the map processing device 2. Furthermore, the functions of the car navigation system 3 may be provided in the map processing device 2, in which case there is no need to provide the car navigation system 3. Similarly, the automatic driving control device 4 may be provided with the various functions described above of the map processing device 2.

[0111] In the above embodiment, an example has been described in which lane connection data, lane attribute data, lane boundary data, lane centerline shape data, lane boundary shape data, and lane group connection data are separately provided as high-precision map data to be processed, but the present invention is not limited to this. Among these high-precision map data, some of the high-precision map data may be combined into one piece of high-precision map data, or one piece of high-precision map data may be divided into multiple pieces of high-precision map data. Furthermore, the high-precision map data to be processed may be, for example, similar map data containing information similar to the various high-precision map data described above, or related data from which the various high-precision map data described above can be derived.

[0112] In the above embodiment, an example of installing a map processing device 2 in a vehicle with an automatic driving function has been described, but the present invention is not limited to this, and the map processing device 2 of the above embodiment can also be applied to vehicles that do not have an automatic driving function.

[0113] Furthermore, the above-described embodiment has been a detailed and specific description of the device configuration in order to clearly explain the present invention, and the present invention is not necessarily limited to an apparatus having all of the described configurations. Furthermore, the present invention can take on various other applications and modifications without departing from the gist of the present invention as set forth in the claims.

[0114] 1...In-vehicle system, 2...Map processing device, 3...Car navigation system, 4...Automatic driving control device, 5...Map distribution server, 10...Control unit, 11...Memory unit, 12...Sensor unit, 20...Surroundings recognition unit, 21...Self-position estimation unit, 22...Candidate route search unit, 23...Route selection unit, 24...Lane data pre-reading unit, 25...Map access unit, 31...Lane connection / attribute data storage unit, 32...Lane boundary data storage unit, 33...Lane shape data storage unit, 34...Lane group connection data storage unit, 40...Lane group area, 41...Lane group connection data, 42...Lane connection data, 43...Lane attribute data, lane center line shape data, 44...Lane boundary data, lane boundary shape data, 45...Physical structure, 51, 52, 53, 54...Pre-read target points

Claims

1. A map processing device capable of acquiring multiple types of map data including lane connection data indicating connection information between lane sections in the extension direction of the lanes, and lane group connection data indicating connection information between lane group sections in the extension direction of a lane group made up of one or more lanes, and comprising a map data pre-reading unit that, when acquiring map data of a specified point ahead of the vehicle from the multiple types of map data, acquires the type of map data according to the conditions of the specified point.

2. The map processing device according to claim 1, wherein the condition for the specified point is the distance from the vehicle position to the specified point, and the map data pre-reading unit acquires the multiple types of map data if the distance from the vehicle position to the specified point is within a specified range, and acquires only the lane group connection data if the distance from the vehicle position to the specified point is within a specific range that is farther than the specified range.

3. The map processing device according to claim 1, wherein the condition for the specified point is whether or not the specified point is on a planned driving route for the vehicle, and the map data pre-reading unit acquires the lane connection data when the specified point is on a planned driving route for the vehicle, and acquires the lane group connection data when the specified point is on a route branching off from the planned driving route for the vehicle.

4. The map processing device according to claim 3, further comprising a route search unit that, when the vehicle deviates from the planned driving route, searches for candidate routes that can become a new planned driving route by referring to the lane group connection data already acquired at points on the route on which the vehicle can travel after the deviation.

5. A map processing device as described in claim 4, further comprising a route determination unit which, when a plurality of candidate routes are obtained by the route search unit, sequentially selects one of the plurality of candidate routes in accordance with a preset selection order, obtains the lane attribute data of the selected candidate route, and determines whether the selected candidate route is suitable for driving the vehicle.

6. The map processing device according to claim 3, wherein the map data pre-reading unit acquires the lane connection data for points on a route branching off from a planned driving route of the vehicle and for some points subsequent to the branching point.

7. The map processing device according to claim 1, further comprising an image capturing device for capturing an image of the surroundings ahead of the vehicle, wherein the map data pre-reading unit does not acquire the map data relating to lanes that are not captured in the surroundings image.

8. A map processing method comprising: a map data pre-reading unit of a map processing device capable of acquiring multiple types of map data including lane connection data indicating connection information between lane sections in the extension direction of the lanes, and lane group connection data indicating connection information between lane group sections in the extension direction of a lane group made up of one or more lanes, the map data pre-reading unit acquiring, when acquiring map data of a specified point ahead of the vehicle from the multiple types of map data, a type of map data corresponding to conditions of the specified point.

Images