WO2024075572A1 - Information processing method, program, information processing device, and data structure - Google Patents

Abstract

Provided is an information processing method executed by an information processing device that includes a processor, the processor executing the steps of extracting an edge part of a passage on the basis of point group data for a passage other than a roadway (S104), identifying attribute information relating to the passage on the basis of the point group data (S106), generating line segment data relating to the passage on the basis of the attribute information and edge part data relating to the edge part (S108), and associating the passage attribute information that corresponds to the line segment data with the line segment data (S110).

Description

Information processing method, program, information processing device, and data structure

The present invention relates to an information processing method, a program, an information processing device, and a data structure.

In recent years, there has been active development of technology to generate highly accurate map data and use this map data for autonomous driving and the like. Generally, when map data to be used for autonomous driving and the like is generated, features are detected by various sensors mounted on the vehicle, feature data relating to these features is generated, and this feature data is associated with the map data. This enables the vehicle to properly grasp the features around the vehicle using the feature data associated with the map data. It is known that this feature data includes information on the type of external sensor and environmental information on the environment when the features are detected for each type of external sensor (see, for example, Patent Document 1).

JP 2020-73893 A

In conventional technology, highly accurate map data is generated based on point cloud data and various other data collected when a measurement vehicle equipped with an MMS (Mobile Mapping System) measurement system drives along the roadway.

However, conventional high-precision map data was designed for vehicles traveling on roads and could not be used for vehicles traveling on roads other than roads.

The present invention aims to provide highly accurate map data that can be used by vehicles traveling on roads other than roadways.

An information processing method according to one aspect of the present invention is an information processing method executed by an information processing device including a processor, in which the processor executes the following operations: extracting edges of a passageway other than a roadway based on point cloud data of the passageway, identifying attribute information related to the passageway based on the point cloud data, generating line segment data related to the passageway based on the edge data related to the edges and the attribute information, and associating the line segment data with attribute information of the passageway corresponding to the line segment data.

The present invention makes it possible to provide highly accurate map data that can be used by vehicles traveling on roads other than roadways.

1 is a diagram illustrating an example of a configuration of an information processing system according to an embodiment of the present invention. 1 is a diagram showing an example of a configuration of a vehicle according to an embodiment of the present invention; 1 is a diagram illustrating an example of a configuration of an information processing device according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a database according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of information relating to feature data of a route according to an embodiment of the present invention; FIG. 1 is a diagram showing an example of a road without a sidewalk. 1A and 1B are diagrams showing examples of various obstacles on a walkway. FIG. 1 is a diagram showing an example of grating present on a sidewalk. FIG. 13 illustrates the cut-off for vehicle entry and exit. FIG. 13 is a diagram showing an example of a relationship between sidewalks and paved roads outside a road and attribute information such as road width. FIG. 13 is a diagram showing an example of a relationship between sidewalks and paved roads outside a road and attribute information such as road width. FIG. 2 is a diagram showing an example of a width code according to an embodiment of the present invention. 11 is a diagram showing an example of attribute information that can be associated with line segment data according to an embodiment of the present invention; FIG. FIG. 13 is a diagram showing an example of section cutting in case 1. FIG. 13 is a diagram showing an example of section cutting in case 2. FIG. 13 is a diagram showing an example of section cutting in case 3. FIG. 13 is a diagram showing another example of section cutting in case 3. 5 is a flowchart showing an example of a process related to map data generation according to an embodiment of the present invention. 10 is a flowchart showing an example of a process from identifying attribute information to associating processing according to an embodiment of the present invention.

[Embodiment]
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to the accompanying drawings, in which like reference numerals denote like or similar structures.

<System Overview>
Fig. 1 is a diagram showing an example of the configuration of an information processing system 1 according to an embodiment of the present invention. The information processing system 1 shown in Fig. 1 includes a measurement vehicle (hereinafter also referred to as "vehicle") 10 equipped with a Mobile Mapping System (MMS) measurement device, an information processing device 20, a positioning satellite 30 used in a Global Navigation Satellite System (GNSS), and an observation satellite 40 capable of acquiring satellite images, which are capable of transmitting and receiving data to and from each other via a network N. In addition, the number of vehicles 10 and information processing devices 20 may be one or more.

Vehicle 10 is a vehicle that performs three-dimensional measurements using MMS, and performs three-dimensional measurements of the surrounding terrain while moving. Vehicle 10 can also receive signals from GNSS positioning satellites 30 and detect its own vehicle's position information. The position information includes three-dimensional position information of latitude, longitude, and altitude, or two-dimensional position information of latitude and longitude.

The vehicle 10 is also equipped with various sensors and acquires data detected by the various sensors. The various sensors include an on-board camera, a vehicle speed sensor, an acceleration sensor, and the like. The vehicle 10 shown in FIG. 1 is an example in which measuring instruments and the like are mounted on a personal mobility device, but measuring instruments and the like may also be mounted on a general vehicle that travels on a roadway.

The positioning satellite 30 is a satellite that transmits signals necessary to measure position information. The observation satellite 40 uses, for example, a Synthetic Aperture Radar (SAR) sensor to observe the Earth and obtain satellite images by photographing it.

The information processing device 20 is, for example, a server, and acquires data measured by the MMS, data detected by various sensors, and the like from the vehicle 10. The information processing device 20 also acquires signals and images from each of the satellites 30, 40. The information processing device 20 generates map data using the data acquired from the vehicle 10 and each of the satellites 30, 40. Note that the information processing device 20 may be composed of multiple information processing devices.

<Vehicle configuration>
Fig. 2 is a diagram showing an example of the configuration of a vehicle 10 according to an embodiment of the present invention. In the example shown in Fig. 2, the vehicle 10 travels in a measurement area using an MMS and measures three-dimensional coordinate values of features around the traveled roadway and/or around a traffic path other than the roadway. In order to measure these three-dimensional coordinate values, the vehicle 10 includes a GNSS receiver 101, an IMU 102, a laser scanner 103, and a camera 104. For example, these devices are attached to a top plate 100 provided on the top of the vehicle 10.

The vehicle 10 also has an odometer 105 attached to the wheel, and is equipped with a processor 106, various sensors 107, and a communication device 108 as on-board devices.

The GNSS receiver 101 is equipped with an antenna that receives positioning signals from the GNSS positioning satellites 30. The GNSS receiver 101 calculates the pseudo-range to the positioning satellites 30, the phase of the carrier wave carrying the positioning signal, and three-dimensional coordinate values based on the positioning signal received by the antenna.

The IMU 102 is equipped with a gyro sensor that measures angular velocity in three axial directions, and an acceleration sensor that measures acceleration. The IMU 102 acquires attitude data of the vehicle using these sensors.

The laser scanner 103 emits laser light in the width direction of the vehicle 10 while changing the emission angle, and receives the laser light reflected by a feature located at the destination of the emission. The laser scanner 103 measures the time from when the laser light is emitted to when it is received, and calculates the distance to the feature.

The camera 104 captures images of the outside of the vehicle 10, such as the front. The odometer 105 measures the distance traveled by the vehicle 10.

The processor 106 controls the driving of the vehicle 10, or controls the transmission and reception of data to and from external devices. The various sensors 107 include an on-board camera, a vehicle speed sensor, and an acceleration sensor. The communication device 108 transmits and receives data to and from external devices. For example, the communication device 108 transmits data acquired by MMS to the information processing device 20. Note that, for measuring point cloud data, etc., on sidewalks, within facilities, and on premises (including, but not limited to, airports and ports), measurements may be taken using measuring devices such as handheld, backpack, and fixed types.

<Configuration of information processing device>
3 is a diagram showing an example of the configuration of an information processing device 20 according to an embodiment of the present invention. The information processing device 20 includes one or more processing devices (CPU: Central Processing Unit) 210, one or more network communication interfaces 220, a storage device 230, a user interface 250, and one or more communication buses 270 for interconnecting these components. Note that the user interface 250 is not necessarily required.

Storage device 230 may be, for example, a high-speed random access memory such as a DRAM, SRAM, or other random access solid-state storage device, or may be a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices, or may be a non-transitory computer-readable recording medium.

Also, another example of storage device 230 may be one or more storage devices located remotely from CPU 210. In one embodiment, storage device 230 stores programs, modules, and data structures executed by CPU 210, or a subset thereof.

The storage device 230 stores data used by the information processing system 1. For example, the storage device 230 stores data related to the generation of high-precision three-dimensional map data (hereinafter also referred to as "HD (High Definition) maps"). As specific examples, dynamic maps, HD maps, feature data, etc. are stored in the storage device 230.

FIG. 4 is a diagram showing an example of a database according to one embodiment of the present invention. In the example shown in FIG. 4, the storage device 230 stores dynamic map data as a high-precision map database.

In the example shown in FIG. 4, the map data is, for example, high-precision three-dimensional map data used in autonomous vehicles. As a specific example, this map data is data on a map called a dynamic map that is provided in real time and to which more dynamic information, such as information on surrounding vehicles and traffic information, has been added. The map data that can be used in this embodiment is classified, for example, into four hierarchies. The map data is classified into static information SI1, quasi-static information SI2, quasi-dynamic information MI1, and dynamic information MI2.

Static information SI1 is high-precision, three-dimensional, basic map data (HD data) that includes road surface information, lane information, three-dimensional structures, etc., and is composed of three-dimensional position coordinates and linear vector data that indicate features. Quasi-static information SI2, semi-dynamic information MI1, and dynamic information MI2 are dynamic data that change from moment to moment, and are data that are superimposed on static information based on position information.

Semi-static information SI2 includes traffic regulation information, road construction information, wide-area weather information, etc. Semi-dynamic information MI1 includes accident information, congestion information, narrow-area weather information, etc. Dynamic information MI2 includes ITS (Intelligent Transport System) information, including information on nearby vehicles, pedestrians, traffic lights, etc.

The static information SI1 included in the dynamic map data includes HD data, and the HD data includes feature data. This feature data is basic information when an application uses a dynamic map, and is information that can contribute to improving performance when an application uses a dynamic map. Therefore, it is important to consider what information to include in the feature data. For example, dynamic maps can be used in automated driving systems such as AGVs (Automatic Guided Vehicles) and PMVs (Personal Mobility Vehicles) that are capable of autonomous driving. Appropriate information is detected from a large amount of information in order to contribute to improving the performance of the automated driving system, and the detected information is included in the feature data in this embodiment.

In addition, in this embodiment, the map data used for routing (route setting) and navigation (route guidance) of autonomous vehicles, remotely controlled vehicles, and vehicles controlled by the user that run on roads other than roadways may be static information S11 alone. In other words, if a vehicle can acquire at least the HD map of static information S11, it will be able to run autonomously or under remote control. The vehicle may be any vehicle that has wheels and is capable of running, including general vehicles, AGVs, PMVs, wheelchairs, strollers, two-wheelers, unicycles, etc.

Returning to FIG. 3, the CPU 210 that executes the process of generating feature data according to this embodiment will be described. The CPU 210 executes the programs stored in the storage device 230 to configure a map control unit 212, a transmission/reception unit 213, an acquisition unit 214, an extraction unit 215, an identification unit 216, a generation unit 217, and an association unit 218.

The CPU 210 controls the processing of the map control unit 212, which controls the processing of each unit described below and executes processing related to the generation of map data.

The map control unit 212 uses various data to control the generation of map data. For example, the map control unit 212 controls the generation of HD data and also controls the generation of feature data included in the HD data. The map control unit 212 may control the generation of HD maps of traffic routes other than roads, or may control the generation of HD maps of traffic routes other than roads in addition to HD maps of roads.

The transmission/reception unit 213 transmits and receives data to and from external devices via the network communication interface 220. For example, the transmission/reception unit 213 is configured as a receiving unit that receives data, signals, etc. from the vehicle 10 and each of the satellites 30, 40, and is also configured as a transmitting unit that transmits data, signals, etc. to the vehicle 10 and each of the satellites 30, 40. As a specific example, the transmission/reception unit 213 receives from the vehicle 10 various data measured by the MMS and various data sensed by various sensors mounted on the vehicle 10, and receives satellite images including a specified position from the observation satellite 40.

The acquisition unit 214 acquires various data received by the transmission/reception unit 213 as necessary, and acquires, for example, various data used to generate map data including at least roadways other than roads. As a specific example, the acquisition unit 214 acquires at least point cloud data measured by the vehicle 10.

The extraction unit 215 extracts the edges of the roadway based at least on the point cloud data of the roadway other than the roadway. For example, the extraction unit 215 may extract the edges of the roadway by detecting the edges of the sidewalk or the like included in the point cloud data.

In addition, when extracting the edges of the road, the extraction unit 215 may extract the following features using various data measured by the MMS and various data sensed by various sensors mounted on the vehicle 10.
・Center lines of traffic routes, virtual stop lines, etc. ・Sidewalks (both ends in the longitudinal direction)
・Crosswalks and intersections (areas where pedestrians cross)
- Road edges (pavement edges) on roads without sidewalks
- Features near sidewalks and road edges that may be obstacles to AGVs (pillars, facilities, roadside trees, curbs, etc.)
- Side gutters (whether or not they have covers or gratings) that exist in the longitudinal direction on the road edge, U-shaped gutters that cross the road, etc. - Entrances and exits of facilities, etc. When extracting a virtual line, the extraction unit 215 generates a virtual line based on feature data (e.g., a passageway, a sidewalk, etc.) extracted from various data, and extracts the generated virtual line. For example, the extraction unit 215 generates a virtual center line from line segment data on both ends of a sidewalk, and generates a virtual stop line based on the presence or absence of a curb, etc.

The extraction unit 215 may set the following conditions as conditions for extracting features on a route.
- Areas where pedestrians pass, such as sidewalks (within a specified distance)
(Even if the road does not have a sidewalk, there is an area where people can walk. However, if there is a gutter, AGVs and other vehicles cannot run on it, so the gutter area is not suitable for running on and is not included in the roadway.)

The identification unit 216 identifies attribute information related to the roadway based at least on point cloud data of roadways other than roadways. For example, the identification unit 216 identifies attribute information for each feature extracted by the extraction unit 215 based on point cloud data, etc. As a specific example, the identification unit 216 may measure at least the width of the roadway and assign the width data to an item of attribute information, as described below.

The generating unit 217 generates line segment data for a passageway based on the edge data for the edges extracted by the extracting unit 215 and the attribute information identified by the identifying unit 216. For example, the generating unit 217 divides or separates predetermined line segment data that runs along the edge data based on width data, feature data, etc., to generate each piece of line segment data. As a specific example, the generating unit 217 may classify width data according to the size of the width data, and generate line segment data such that adjacent line segment data have different classification values for width data.

The generating unit 217 may generate, as the line segment data, at least one of the edge data itself relating to the edge, line segment data along the edge data, line segment data located approximately in the center of the passageway that runs through the passageway, and line segment data approximately parallel to the edge data. The generating unit 217 may also generate virtual lines of the passageway described above, such as center lines and stop lines.

The matching unit 218 matches the line segment data generated by the generation unit 217 with attribute information of the route corresponding to the line segment data. For example, the matching unit 218 may match each piece of generated line segment data with attribute information that was the cause of generating the line segment data. As a specific example, when line segment data is generated according to the size of width data, the matching unit 218 may match the numerical value or code of the width data to the line segment data.

The above processing makes it possible to generate feature data contained in high-precision map data such as sidewalks, which is used to control the travel of AGVs, PMVs, wheelchairs, strollers, etc. (collectively referred to as "small vehicles"). This allows small vehicles to control their travel in the direction of travel along the line segment data of the route, and to perform at least one of the control processes of speed, whether or not to travel, and whether to stop, based on the attribute information associated with the line segment data. Note that feature data such as sidewalks can be used not only by small vehicles, but also by vehicles that can travel on sidewalks, such as motorcycles and narrow vehicles.

The attribute information identified by the identification unit 216 may include attribute information of multiple items of different types. For example, the attribute information may include the following items: pavement type, average cross slope, cross slope direction, average longitudinal slope, longitudinal slope direction, road surface notes, traffic notes, and obstacle notes.

For example, the pavement type can be identified based on at least one of point cloud data, satellite images, images captured by the vehicle 10, etc., and one of the options such as asphalt, concrete, colored pavement, gravel (unpaved), etc. is identified.

For example, the average transverse slope and the average longitudinal slope can be determined based on at least one of the point cloud data, the angular velocity sensor mounted on the vehicle 10, and the images captured by the vehicle 10, and are determined in percentages.

The transverse gradient direction and longitudinal gradient direction can be similarly determined based on at least one of the point cloud data, the angular velocity sensor mounted on the vehicle 10, images captured by the vehicle 10, etc. The transverse gradient direction can be determined as being the roadway side or the private land boundary side, and the longitudinal gradient direction can be determined as being the starting point side or the end point side of the line segment data.

The special notes, such as road surface notes, traffic notes, and obstacle notes, may be set by the user. For example, the identification unit 216 may set the content entered by the user for each special note using the user interface 250 as attribute information.

The above processing makes it possible to include appropriate items in the attribute information for routes other than roadways. For example, attribute information suitable for vehicle driving control can be associated with line segment data of the route as feature data. This allows the vehicle to use the line segment data while also acquiring attribute information associated with the line segment data and using it for appropriate driving control.

The generating unit 217 may also include dividing (or separating) the edge data of the passageway to generate line segment data when the attribute information of at least one item satisfies a predetermined condition for division. For example, the generating unit 217 may include, as a predetermined condition for division, that the attribute information of at least one item is different. For example, when the attribute information includes different width data or gradients, the generating unit 217 may divide the data into edge data showing similar values (values within a predetermined range) and generate each line segment data.

The above process generates line segment data using edge data extracted from point cloud data, etc., making it easy to generate line segment data and reducing the processing load on the information processing device 20. Also, when creating line segment data for the center line of a roadway other than a roadway, as with roadways, the width of the roadway often differs depending on the area, so the information indicating the position of the line segment data indicating the center line becomes complex, increasing the processing load and memory capacity.

On the other hand, by generating line segment data from edge data, data for basically straight lines is treated as line segment data, so data such as positional information can be easily identified, and by using differential values for the positional information of each line segment data, it is possible to reduce the storage capacity of the positional information.

The path may also include a sidewalk, or a paved portion from the road outer line to the edge of the road pavement. In this case, the extraction unit 215 may include extracting the edge of the sidewalk on the opposite side of the roadway (e.g., the private land boundary side), or the pavement edge of the paved portion.

As described above, the extraction unit 215 extracts the outer edge of the sidewalk (also referred to as the "sidewalk outer edge") and the outermost edge of the paved portion (also referred to as the "road pavement edge") as edges, and the generation unit 217 generates line segment data from edge data based on these edges, thereby making it possible to generate simple and easy-to-manage line segment data.

The significance of using the outer sidewalk edge and the road pavement edge as line segment data is given below.
- The extraction unit 215 does not need to acquire new features separately. - There is less horizontal and vertical ingress and egress and change within the area in which AGVs, etc. travel compared to other linear shapes that exist on the route. - Because horizontal entrances and exits such as bus stops and cut-offs may be provided in some areas closer to the roadway, the line segment data for the outer edge of the sidewalk and the edge of the road pavement is straighter. - There are many obstructing features such as bus stop timetables, street trees and tree belts closer to the roadway, so it is possible for AGVs, etc. to travel safely and stably if they travel along the line segment data for the outer edge of the sidewalk (or the edge of the road pavement).

The route may also include a passageway within a building or private property. In this case, the extraction unit 215 may extract the edge of the passageway within the building or the edge of the passageway within the private property. For example, by driving a small vehicle capable of driving within a building or private property and capable of MMS measurement, the acquisition unit 214 can acquire point cloud data, etc., of the routeway within the building or private property.

By the above processing, the extraction unit 215 extracts edges of passages within buildings or private property as edges, the generation unit 217 generates line data from edge data based on these edges, and the association unit 218 associates attribute information corresponding to the line data with the line data. This makes it possible to control the travel of small vehicles even on passages within buildings or private property.

The generating unit 217 also generates map data of the route, including each line segment data associated with each attribute information of the route. For example, the generating unit 217 may generate map data showing only routes other than roads, or may generate map data in which roads and routes other than roads are integrated by adding feature data showing routes other than roads to conventional map data including roads.

The above processing makes it possible to distribute map data that includes feature data for routes other than roads, increasing the opportunities for using this map data and also increasing the availability of map data for routes other than roads.

<Data example>
Next, examples of data related to the map data generated by the above-mentioned process will be described with reference to Figures 5 to 9. Figure 5 is a diagram showing an example of information related to feature data of a route according to an embodiment of the present invention. The feature-related information shown in Figure 5 includes attribute information and various codes for features that can be extracted from point cloud data, for example.

In the example shown in Figure 5, in the case of roads that do not have separate sidewalks, features related to the outside of the roadway are extracted, and since it is anticipated that AGVs and other vehicles will also travel on crosswalks, crosswalk features are extracted even if they are inside the roadway.

In the example shown in Figure 5, the features extracted as roadway-related features are the outer roadway line, stop line, pedestrian crossing, curb (lower edge of roadway side), and road pavement edge (area without sidewalk). In addition, the code "1-1" written to the left of the target feature may be used as a feature code to identify the feature.

Furthermore, each target feature shown in FIG. 5 is associated with a "Shape name/form" related to the road, a code "class" related to the attribute, and "walkwidth" (width data), which is one attribute item. "Shape name/form" includes the type of feature such as road, sidewalk, obstacle, build, etc., and in parentheses, the type of shape such as line, poly (surface, point), etc. "Class" includes the attribute type code. "Walkwidth" (width data) includes width data of walkways, etc. that can be measured from point cloud data, etc. The code "class" may be used as a feature code.

The extraction unit 215 extracts target features included in the feature-related information shown in FIG. 5 from point cloud data, etc., and the identification unit 216 may identify various codes for the extracted target features by referring to the feature-related information shown in FIG. 5, and assign them to the target features as attribute information.

(Target feature)
Next, an example of a target feature used for driving support of a small vehicle will be described with reference to Fig. 6. The extraction unit 215 extracts both ends of the sidewalk, the road pavement edge, and the road outer edge as the driving range of the small vehicle.

FIG. 6A is a diagram showing an example of a road without a sidewalk. When there is no sidewalk, as in the example shown in FIG. 6A, the drivable area is between the road pavement edge and the road exterior line. The road pavement edge may be a curb (bottom edge), and the road exterior line may not be a line. When there is a sidewalk, both ends of the sidewalk are extracted.

In addition, the extraction unit 215 may extract areas and lines related to driving and stopping at intersections, etc. For example, the following features are extracted.
・Crosswalk (area frame)
・Boundary of the sidewalk side of the curb (crosswalk connection)
・Stop lines on roads only when there are no sidewalks

The extraction unit 215 may also extract obstacles (area frames) within the travel range as target features. For example, the following obstacles may be extracted:
・Support-shaped objects (traffic signals, trees, bus stop timetable poles, etc.)
・Facilities (electrical facilities, planting frames, post boxes, telephone booths, etc.)
- Including step-eliminating blocks etc.

FIG. 6B is a diagram showing examples of various obstacles on a sidewalk. In the example shown in FIG. 6B, trees, traffic lights, bus stops, signs, etc. may be extracted as obstacles.

The extraction unit 215 may also extract unsuitable driving areas (area frames) within the driving range of a small vehicle. For example, the following features may be extracted as unsuitable driving areas:
- Access to drain covers, gratings, and sidewalks (cut-offs, passages to private homes, garages, etc.)

Fig. 6C is a diagram showing an example of gratings present on a sidewalk. In the example shown in Fig. 6C, there is one grating on the sidewalk and one at the edge of the roadway, indicated by a thick frame. Fig. 6D is a diagram showing a cut-off for vehicles to enter and exit. In the example shown in Fig. 6D, the oval part indicates the cut-off. It is advisable to arrange for the features shown in Figs. 6A to 6D to be extracted by the extraction unit 215.

(Generating line data, adding attributes to road width data)
7A and 7B are diagrams showing an example of the relationship between sidewalks and roadside pavements and attribute information such as road width. The generating unit 217 generates line segment data based on the positions of features (obstacles, etc.) that narrow the width of the sidewalks and roadside pavements (collectively referred to as "sidewalks, etc."). The generating unit 217 may use the sidewalk edge (private land boundary) or the road pavement edge as a reference for the line segment data that is the source of the generated line segment data.

Figure 7A shows an example in which a sidewalk is located above the road shown in the figure, and a paved road outside the road is located below the road shown in the figure. There are also obstacles on the sidewalk and in the paved road outside the road, and the width of the paved road outside the road further narrows as it goes to the right.

In the example shown in FIG. 7A, the generation unit 217 finally divides the road into sections A to C (line segment data). The type code for the outer edge (also called the "outer edge") of the sidewalk in sections A to C is given the type code "21" (sidewalk edge (private land side, outer edge)) shown in FIG. 5 by the identification unit 216. The width data (outer width), which is one of the attributes associated with the line segment data, is given a width code, described below, by the identification unit 216.

For example, the width code of sections A and C is "12" (1 m or more, less than 2 m), and the width code of section B is "11" (0.5 m or more, less than 1 m), which is narrower than the widths of sections A and C, due to the presence of an obstacle. In addition, the type code of the inside edge (also called the "inner edge") of the sidewalk of sections A to C (line segment data) is "22" (curb (sidewalk edge (sidewalk edge of the curb between the sidewalk and the roadway)) as shown in FIG. 5, which is assigned by the identification unit 216.

In the example shown in FIG. 7A, the generation unit 217 finally divides the sections J to M (line segment data) into the outer roadway edges (also referred to as "road edges") in the same way. The type code for sections J to M is given by the identification unit 216 the type code "11" (outer roadway line) shown in FIG. 5. The width data (road edge width), which is one of the attributes associated with the line segment data, is given a width code of "0" (less than 0.5 m), "11" or "12" by the identification unit 216 based on the width data of each section. The type code "13" may also be given to each section of the road pavement edges of the outer road pavement by the identification unit 216.

In the example shown in FIG. 7B, the identification unit 216 uses at least the point cloud data, etc. to identify the width data and the gradient of the cutoff included in the attribute information. For example, the identification unit 216 identifies the type code "23" corresponding to the cutoff in the area of section H. The identification unit 216 also identifies that the areas of sections O, Q to R have no outer lane markings and the distance from the outer lane line to the pavement edge is less than 0.5 m, and assigns a width code of "0". Note that the black markings located at the edge of the road in section O indicate obstacles that are constantly present, such as mailboxes. In this case, the generation unit 217 generates line segment data by dividing the obstacle by 1 m on both ends, and the association unit 218 associates the line segment data with width data "0", which indicates that the vehicle 10 cannot travel.

7A and 7B, the process up to the generation of line segment data will be described. First, the extraction unit 215 extracts edges of the passageway other than the roadway based at least on the point cloud data, and extracts features related to the passageway, such as obstacles. Next, the identification unit 216 identifies attribute information of the features extracted by the extraction unit 215. For example, the identification unit 216 identifies the code of the feature using the feature-related information shown in FIG. 5, and assigns the code to the extracted target feature. Next, the generation unit 217 divides the passageway into sections based on the extracted edges and the identified attribute information, and generates each line segment data. Finally, the association unit 218 associates the generated line segment data with the corresponding attribute information.

(Width Code)
Here, the width code will be described. FIG. 8 is a diagram showing an example of a width code according to an embodiment of the present invention. The advantage of setting a width code is that it reduces the cost of creating and managing width data, and reduces the processing load on the side that uses the width data. For example, if detailed width values are set for each sidewalk section, the processing load and cost increase, and it becomes complicated to grasp the width even when setting a route for a small vehicle. Therefore, the specification unit 216 may set an approximate width level (width code) based on the width value, and may assign the width code as attribute information for each section (each line segment data). The width code shown in FIG. 8 is an example, and other levels (ranges) may be set.

FIG. 9 is a diagram showing an example of attribute information that can be associated with line segment data according to one embodiment of the present invention. In the example shown in FIG. 9, the attribute information includes pavement type, average cross gradient, cross gradient direction, average longitudinal gradient, longitudinal gradient direction, road surface notes, traffic notes, and obstacle notes.

The pavement type, for example, the type of road surface such as asphalt, concrete, colored pavement, gravel, etc., can be identified by the identification unit 216 based on point cloud data, images captured by the vehicle 10, etc. The average transverse slope and average longitudinal slope can be identified by the identification unit 216 based on at least one of the point cloud data, an angular velocity sensor mounted on the vehicle 10, and images captured by the vehicle 10. For example, the average transverse slope and average longitudinal slope are identified in percentage.

The transverse gradient direction and longitudinal gradient direction can be identified by the identification unit 216 based on at least one of the point cloud data, the angular velocity sensor mounted on the vehicle 10, and the captured image of the vehicle 10. The transverse gradient direction can be identified as descending toward the roadway or toward the private land boundary, and the longitudinal gradient direction can be identified as descending toward the starting point or the end point of the line segment data.

The special notes, such as road surface notes, traffic notes, and obstacle notes, may be set by the user. For example, the identification unit 216 may set the content entered by the user for each special note using the user interface 250 as attribute information. For road surface notes, for example, pavement deterioration, cracks, and swelling caused by tree roots may be set. For traffic notes, for example, a lot of bicycle traffic, a lot of pedestrians during the day, etc. may be set. For obstacle notes, for example, a lot of parked bicycles, a lot of parked vehicles, a lot of parked signs, a lot of step-eliminating blocks, etc. may be set.

<Example of dividing a line>
Next, some examples will be given to explain generation of line segment data by the generation unit 217. In the examples shown below, it is assumed that point cloud data of a sidewalk or the like is also measured while traveling on a roadway, but data measured by the MMS of the vehicle while traveling separately on a sidewalk or the like may also be used.

(Case 1) Section cutting of features with a longitudinal length of less than 1 m on the road surface When there is a feature on the road surface with a longitudinal length of less than 1 m (approximately 20 to 40 cm) such as a utility pole or a lamp post, and this feature narrows the width of the sidewalk, it is possible that such a small feature section cutting may be overlooked when setting a route for a small vehicle. Therefore, in the case of a feature with a length of less than 1 m, the generation unit 217 extends a substantially perpendicular line from the road centerline (which is imaginary if there is none) to the sidewalk edge (private land boundary) or road pavement edge so as to pass through the position of the post, etc., and from the intersection, cuts the sidewalk edge or pavement edge into sections of about 1 m forward and backward (guideline) for a total of about 2 m.

FIG. 10 is a diagram showing an example of section division in case 1. In the example shown in FIG. 10, the generation unit 217 takes the intersection point of the sidewalk edge when a substantially perpendicular line is extended from the road centerline, passing through obstacle O10, as the center, and divides the area along the sidewalk edge to the left and right (or front and rear) from this center at a range of a predetermined distance, generating line segment data. For sections L10 and L14 shown in FIG. 10, the width data (width value) is M10 for both sections. On the other hand, because obstacle O10 is present in section L12, the width data M12 for this section is narrower than M10.

(Case 2) Cutting off sections of features whose longitudinal length on the road surface exceeds 1 m For features whose longitudinal section on the road surface exceeds 1 m, the generation unit 217 extends an approximately perpendicular line from the road centerline (which is imaginary if there is none) to the edge of the sidewalk (private land boundary) or the edge of the road pavement, so as to pass through the positions of both ends (maximum parts) of the longitudinal direction of the feature, and cuts off the section between the point where the width begins to change (from the point where it starts to narrow).

Fig. 11 is a diagram showing an example of section division in case 2. In the example shown in Fig. 11, the sidewalk is narrowed by a bus bay, parking area, etc. The generation unit 217 sets an approximately perpendicular line from the center line passing through both ends (widest parts) of the feature, and divides the area at the intersection of this perpendicular line and the sidewalk edge. For sections L20 and L24 shown in Fig. 11, the width data (width value) is M20. On the other hand, for section L22, the width data of the narrowest area is measured and set to M22.

(Case 3) Section cutting when features exist consecutively at a short distance When features that cause the road width to be narrowed exist consecutively at a certain interval in the longitudinal direction, the generation unit 217 does not cut sections for each feature, but cuts sections as a whole. The association unit 218 associates the narrowest road width information, taking into consideration that the road width information to be assigned to the section may not be passable. The distance between consecutive features here is about 5 to 10 m, but this distance is not limited depending on the sidewalk environment before and after.

For example, in a continuous section where there is no access to the sidewalk due to a building entrance or a lowering of the sidewalk, and the section can only be used as a through-passage, the association unit 218 associates the narrowest width data with the section (e.g., FIG. 12).

In addition, if there is a sidewalk drop between consecutive sections, the generation unit 217 will perform section cutting based on the section cutting definitions of the surrounding features (see, for example, Figure 13).

Fig. 12 is a diagram showing an example of a section cut in Case 3. The example shown in Fig. 12 shows a section in which a feature (such as obstacle O30) narrows the width of the sidewalk at a fairly regular interval over a relatively short distance, and the sidewalk is cut down or otherwise prevented from entering or exiting the sidewalk.

For example, as shown in FIG. 12, in a section where it is not possible to enter or exit the sidewalk, there may be scattered features that narrow the sidewalk, and the narrowed width may vary depending on the section. In this case, if there is no access road (entrance, etc.) to the private land within this section, the matching unit 218 treats the sections L32 at both ends as a single unit, and matches the narrowest width M32 (M32 < M34 = M36) caused by the obstacle O30 as an attribute of section L32. This is because by knowing the narrowest width, it is possible to determine whether small vehicles can pass through.

Fig. 13 is a diagram showing another example of section division in Case 3. In the example shown in Fig. 13, there is a relatively short distance of feature (such as obstacle O40) that narrows the width of the sidewalk at a regular interval, but within that section, the sidewalk is lowered, etc., and there is a section L44 where it is possible to enter and exit the sidewalk.

In the example shown in FIG. 13, the association unit 218 assigns sidewalk width data as attribute information to sections L40, L44, and L48, and assigns narrow sidewalk width data M42 and M46 to sections L42 and L46, respectively, as attribute information.

Sections L42 and L44 are not treated as a continuous section because section L44 has a drop-off allowing access to the sidewalk. On the other hand, section L46, which includes obstacle O40, is treated as a continuous section because there is no drop-off allowing access to the sidewalk. The curved arrows in Figure 13 indicate that width code M40 corresponds to the attributes of sections L40 and L48, and width code M44 corresponds to the attributes of section L46.

The above cases 1 to 3 show examples of dividing line segment data into sections, but the examples are not limited to those described above, and the generation unit 217 may generate line segment data (sections) using other conditions, etc.

<Operation>
Next, each process related to map data generation in the information processing system 1 will be described. Fig. 14 is a flowchart showing an example of a process related to map data generation according to an embodiment of the present invention. The process shown in Fig. 14 is executed by the information processing device 20.

In step S102, the acquisition unit 214 of the information processing device 20 acquires at least point cloud data on roads other than roadways. In addition to the point cloud data, the acquisition unit 214 may also acquire data sensed by various sensors. The acquisition unit 214 may also acquire point cloud data and various data related to roadways.

In step S104, the extraction unit 215 extracts the edges of the roadway based at least on the point cloud data of the roadway other than the roadway.

In step S106, the identification unit 216 identifies attribute information related to the route based on the point cloud data acquired by the acquisition unit 214. For example, the identification unit 216 refers to the feature-related information shown in FIG. 5 and identifies attribute information from the features extracted by the extraction unit 215.

In step S108, the generating unit 217 generates line segment data related to the passageway based on the edge data related to the edges of the passageway extracted by the extracting unit 215 and the attribute information identified by the identifying unit 216. For example, the generating unit 217 may generate the line segment data using the method described in FIGS. 10 to 13.

In step S110, the matching unit 218 matches the line segment data generated by the generation unit 217 with attribute information of the route corresponding to the line segment data. The attribute information may be, for example, width data or at least one of the items of data shown in FIG. 9. It is preferable to set width data as the attribute information. This makes it possible to determine whether small vehicles can travel along the route, and can be used for routing and navigation of small vehicles.

In step S112, the generating unit 217 generates map data of the route including each line segment data associated with each attribute information of the route by the associating unit 218. This map data may include an HD map of the roadway.

FIG. 15 is a flowchart showing an example of a process from identifying attribute information to associating the information according to one embodiment of the present invention. The process shown in FIG. 15 shows a specific example of the process of steps S106 to S110 shown in FIG. 14.

In step S202, the identification unit 216 measures the width of each area of the roadway other than the roadway, and identifies the width data (numerical value or width code).

In step S204, the generation unit 217 divides edge data of sidewalks, etc. based on feature data (e.g., obstacles) or attribute information (e.g., width data, gradient, etc.) that exists in the roadway other than the roadway, and generates data for each line segment.

In step S206, the association unit 218 associates (sets) corresponding attribute information with each line segment data generated by the generation unit 217.

In step S208, which is processed in step S206, the association unit 218 inputs the measured path width data as one item of attribute information to be associated with the line segment data. Note that the attribute information items may include the items shown in FIG. 9.

The above process makes it possible to provide highly accurate map data that can be used by vehicles traveling on roads other than roadways. Vehicles that use highly accurate map data do not need to be able to travel autonomously, and this map data can be used for route routing, navigation, etc.

Although one embodiment of the present invention has been described in detail above, the present invention is not limited to the above embodiment, and various modifications and changes are possible within the scope of the claims. For example, the present invention may transfer some of the processes executed by the information processing device 20 to another information processing device, or may appropriately integrate multiple information processing devices.

<Modification 1>
In the first modified example, the above-described embodiment may be configured by a data structure of map data. For example, the data structure is a data structure of map data related to a roadway other than a roadway, which is used in a computer having a processor and a memory and stored in the memory. This data structure includes each line segment data and attribute information. Each line segment data is each line segment generated using edge data related to an edge extracted based on point cloud data on the roadway, and is used by the processor for control processing related to the progress of the vehicle. For example, the vehicle is controlled to run along each line segment data. Therefore, since the line segment data is preferably as straight as possible, it is preferable that the line segment data be generated by dividing the edge data of a straight line such as a sidewalk edge.

The attribute information is attribute information of the route identified based on the point cloud data and associated with each corresponding line segment data, and is used by the processor for at least one control process of the vehicle speed, whether or not it is possible to travel, and whether or not it is possible to stop. For example, the attribute information can be used for route search such as adjusting the vehicle speed, stopping, or making a detour. Specifically, the vehicle speed is adjusted according to the magnitude of the width data in the attribute information, and for example, the vehicle may adjust its speed to be slower as the road width becomes narrower. In addition, the vehicle may be controlled to stop temporarily before the area indicated by the reduction in the attribute information, since there is a possibility that vehicles may enter or exit the area.

As a result, by providing map data having the above-mentioned data structure to a vehicle or a device that remotely controls the vehicle, the vehicle will be able to travel autonomously on routes other than roadways.

<Modification 2>
In the above example, the map data of the route is used for a vehicle, but the map data may be used for other applications. For example, the map data may be provided to a mobile device used by a visually impaired person, and may be used for detailed route guidance, for VR (Virtual Reality) or AR (Augmented Reality), for digital advertising, or for games in which a precise location can be identified.

<Modification 3>
Also, the HD map of a passageway other than a roadway can be associated with a conventional HD map of a roadway. For example, the identification unit 216 identifies the positional relationship between the roadway line segment data and the passageway line segment data, and the association unit 218 associates the roadway line segment data with information capable of identifying the position of the passageway line segment data. For example, for the roadway line segment data that is closest to the positional information of the starting point of the passageway line segment data, a difference value between the positional information of the starting point of the roadway line segment data and the positional information of the starting point of the passageway line segment data is associated with the roadway line segment data. Also, data indicating this difference value may be included in the ID of the feature data of the roadway line segment data.

<Modification 4>
The passageway other than the roadway may include a river on which ships and the like navigate. For example, a measuring device may be mounted on a ship to acquire point cloud data and the like on the river, enabling the generation of a three-dimensional high-precision map of the river. In this case, as described above, line segment data is generated based on attribute information of the river (such as the river width, the position of bridges, and the position of moored ships), and attribute information is assigned to each of the divided line segment data. For example, in a ship-steering support system that supports the operation of a ship, the HD map of the river is used, and the line segment data and/or attribute information associated with the line segment data are used to determine whether or not the river is passable. The HD map of the river can also be used for determining the ship's route and for route guidance (navigation). Therefore, in the case of the fourth modification, the above-mentioned passageway may be replaced with a river, and the vehicle may be replaced with a ship.

1...information processing system, 10...measurement vehicle, 20...information processing device, 30...positioning satellite, 40...observation satellite, 100...top plate, 101...GNSS receiver, 102...IMU, 103...laser scanner, 104...camera, 105...odometer, 106...processor, 107...various sensors, 108...communication device, 210...CPU, 212...map control unit, 213...transmitting/receiving unit, 214...acquisition unit, 215...extraction unit, 216...identification unit, 217...generation unit, 218...association unit, 230...storage device, 250...user interface, 220...network communication interface

Claims (10)

1. An information processing method executed by an information processing device including a processor,
   The processor,
   Extracting edges of a roadway other than a roadway based on point cloud data of the roadway;
   Identifying attribute information regarding the route based on the point cloud data;
   generating line segment data regarding the travel route based on edge data regarding the edge and the attribute information;
   Associating the line segment data with attribute information of the route corresponding to the line segment data;
   An information processing method for performing the above.
2. The attribute information includes attribute information of a plurality of items of different types,
   The generating of the line segment data includes:
   2. The information processing method according to claim 1, further comprising dividing the edge data to generate the line segment data when attribute information of at least one item satisfies a predetermined condition regarding division.
3. The said passageway includes a sidewalk or a paved portion from the road outer line to the road pavement edge,
   The extracting step comprises:
   The information processing method according to claim 1 , further comprising extracting an edge of the sidewalk on an opposite side to the roadway, or a pavement edge of the paved portion.
4. The passageway includes a passageway within a building or private property;
   The extracting step comprises:
   The method of claim 1 , further comprising extracting edges of the passageway within the building or edges within the private property.
5. The generating of the line segment data includes:
   The information processing method according to claim 1 , further comprising generating line segment data based on at least one of the attribute information items of pavement type, gradient, and width.
6. The information processing method according to claim 1, wherein the attribute information includes at least one of road surface conditions, traffic conditions, and obstacle conditions set by a user.
7. The processor,
   The information processing method according to claim 1 , further comprising the step of generating map data of the route, the map data including line segment data associated with each attribute information of the route.
8. A processor included in the information processing device includes:
   Extracting edges of a roadway other than a roadway based on point cloud data of the roadway;
   Identifying attribute information regarding the route based on the point cloud data;
   generating line segment data regarding the travel route based on edge data regarding the edge and the attribute information;
   Associating the line segment data with attribute information of the route corresponding to the line segment data;
   A program that executes the following.
9. An information processing device including a processor,
   The processor,
   Extracting edges of a roadway other than a roadway based on point cloud data of the roadway;
   Identifying attribute information regarding the route based on the point cloud data;
   generating line segment data regarding the travel route based on edge data regarding the edge and the attribute information;
   Associating the line segment data with attribute information of the route corresponding to the line segment data;
   An information processing device that executes the above.
10. A data structure of map data relating to a road other than a roadway, which is used in a computer having a processor and a memory and is stored in the memory, comprising:
    Each line segment data is generated using edge data related to an edge extracted based on the point cloud data on the travel route, and the each line segment data is used by the processor for a control process related to a travel of a vehicle;
    Attribute information of the route identified based on the point cloud data, which is associated with each of the corresponding line segment data, and which is used by the processor for at least one of control processes of the speed of the vehicle, whether or not the vehicle can travel, and whether or not the vehicle can stop;
    A data structure for map data, including:

Images