Information processing device

Abstract

To provide redundancy in pitch angle acquisition.SOLUTION: An information processing device 1 includes: a position information acquisition unit 200 that acquires position information of a vehicle V; a distance information acquisition unit 203 that acquires a distance from the vehicle V to a building; a height information acquisition unit 205 that acquires known height information H_Known of the building based on the position information of the vehicle V and a first distance to the building; a height information calculation unit 206 that calculates height information H_Cal of the building based on a second distance from the vehicle V to the building; and a pitch angle calculation unit 207 that calculates a pitch angle Pitch θ of the vehicle V using the known height information H_Known of the building and the height information H_Cal of the building.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an information processing apparatus.

Background Art

[0002] Patent Document 1 discloses a technique for estimating the attitude of a vehicle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

[0004] Various types of information are required for a vehicle to travel. One example of this type of information is the angle of the vehicle relative to the road surface (hereinafter referred to as the pitch angle). The pitch angle corresponds to the inclination angle of the vehicle relative to the horizontal plane. This pitch angle is used, for example, in object detection, object tracking, sensor fusion, and actuator control in Advanced Driver Assistance Systems (ADAS) and Automated Driving (AD).

Summary of the Invention

Problems to be Solved by the Invention

[0005] The pitch angle is generally provided by an Inertial Measurement Unit (IMU). The inertial measurement unit is a sensor that detects acceleration and angular velocity in three axial directions respectively. If the inertial measurement unit fails, or if there is an error in the information provided by the inertial measurement unit, there is a possibility that problems may occur in assistance provided by the advanced driver assistance system, automated driving, and the like. Therefore, in order to appropriately implement assistance by the advanced driver assistance system, automated driving, and the like, it is necessary to provide redundancy for the acquisition of the pitch angle.

Means for Solving the Problems

[0006] The present invention A location information acquisition unit that acquires location information of a moving object, A measuring instrument that irradiates a transmission wave toward a set irradiation range and receives the reflected wave reflected by an object within the irradiation range, Based on the measurement data from the aforementioned measuring instrument, A distance information acquisition unit that acquires the distance from the moving object to the target object, A height information acquisition unit acquires known height information of an object by referring to map data based on the position information of the moving body and the coordinates of the object identified based on a first distance from the moving body to the object. before A height information calculation unit that calculates the height information of the object to be recorded, The system includes a pitch angle calculation unit that calculates the pitch angle of the moving body using the known height information of the object and the height information of the object. 、 The irradiation area is divided into multiple layers in the vertical direction. The height information calculation unit, The second distance from the moving body to the object in a layer along a horizontal line based on the installation state of the measuring instrument on the moving body, Based on the third distance from the moving body to the top edge of the object in the layer from which the transmitted wave reflected off the top edge of the object was transmitted, the height information of the object is calculated. It was configured as an information processing device. [Effects of the Invention]

[0007] According to the present invention, redundancy can be introduced in the acquisition of the pitch angle. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram illustrating the general configuration of an information processing device. [Figure 2] This diagram illustrates the irradiation range of laser light from a sensor mounted on a vehicle. [Figure 3] This is a functional block diagram of the processing unit of an information processing device. [Figure 4] This diagram explains the principle for calculating building height information. [Figure 5] This is a flowchart explaining the pitch calculation process. [Figure 6]It is a schematic diagram for explaining the building detection process in an irradiation range. [Figure 7] It is a schematic diagram for explaining the calculation of a first distance to a building detected within an irradiation range. [Figure 8] It is a flowchart for explaining the calculation process of height information of a detected building. [Figure 9] It is a diagram for explaining the calculation of a first distance when a plurality of buildings are detected in an irradiation range. [Figure 10] It is a diagram for explaining the calculation of a first distance when a plurality of buildings are detected in an irradiation range. [Figure 11] It is a diagram for explaining another aspect of specifying a scan layer in which an array of points corresponding to the upper side of a building is generated. [Figure 12] It is a flowchart for explaining a modified example of a building height calculation step. [Figure 13] It is a diagram for explaining a scan layer according to a modified example. [Figure 14] It is a diagram for explaining another example of calculating a pitch angle. MODE FOR CARRYING OUT THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described by taking as an example a case where the embodiment is applied to an information processing apparatus 1 mounted on a vehicle V. FIG. 1 is a diagram illustrating a schematic configuration of the information processing apparatus 1. FIG. 2 is a diagram illustrating an irradiation range IR of laser light from a sensor 15 mounted on the vehicle V.

[0010] The information processing apparatus 1 has a function of generating information used for traveling control and the like of the vehicle V. The following are examples of functions of the information processing apparatus 1. (a) A function of acquiring known height information of a target object by using position information of the vehicle V and a distance from the vehicle V to the target object acquired by the sensor 15. (b) A function to calculate the pitch angle of the vehicle V (the inclination of the vehicle relative to the horizontal line) using the distance from the vehicle V to the object obtained by the sensor 15 and known height information of the object.

[0011] As shown in Figure 1, the information processing device 1 includes a processing unit 20, a storage unit 30, and an input / output port (I / O) 11. The processing unit 20, the storage unit 30, and the input / output port 11 are connected to each other via a bus 10. The input / output port 11 is connected to an IMU (Inertial Measurement Unit) 12, a GPS (Global Positioning System) unit 13, a communication unit 14, and a sensor 15.

[0012] IMU12 is an inertial measurement device that detects three-dimensional inertial motion (translational and rotational motion in three orthogonal directions). IMU12 consists of a three-axis gyroscope (not shown) and an accelerometer (not shown). IMU12 has the function of calculating angular velocity, acceleration, attitude angle (roll angle, pitch angle), and azimuth angle.

[0013] The GPS unit 13 has the function of receiving radio waves (GPS signals) from multiple positioning satellites ST. The communication unit 14 has the function of connecting to network NT via wireless, wired, and mobile phone communication networks. The communication unit 14 also has the function of connecting to an external server ES via network NT.

[0014] Sensor 15 is a measuring instrument having an irradiation unit (not shown) that irradiates a transmitted wave toward a set irradiation range IR, and a receiving unit (not shown) that receives reflected waves reflected by an object within the irradiation range. In this embodiment, a LiDAR sensor is used as the sensor 15. LiDAR works by shining a laser beam (transmitted wave) towards the IR (irradiated infrared) area and receiving the laser beam (reflected wave) that is reflected by objects within the IR area. By processing the measurement data (LiDAR data) from sensor 15, it is possible to measure the distance to objects within the IR irradiation range and the shape of those objects.

[0015] In this embodiment, as shown in Figure 2, the sensor 15 is installed at the front of the vehicle V, facing forward. On the vehicle V, the sensor 15 is installed at a position above the road surface R, at a height h. The sensor 15 emits laser light toward the front of the vehicle V. The sensor 15 emits laser light within a predetermined range in the vertical direction (see Figure 2) and within a predetermined range in the horizontal direction (width direction of the vehicle V) (see Figure 1). In this embodiment, the range to which the laser light is emitted is denoted as the "irradiation range IR".

[0016] In this embodiment, the laser beam irradiation area (IR) is divided into multiple layers (scanning layers) in the vertical direction. In Figure 2, for the sake of explanation, an example is shown where the total number of scan layers is "13". In the example irradiation range IR shown in Figure 2, six scan layers (L1 to L6) are set above the reference scan layer L0, and six scan layers (L-1 to L-6) are set below it. Here, the reference scan layer L0 is the scan layer caused by laser light emitted from the sensor 15 in the direction of the horizontal line HL when the pitch angle of the vehicle V on which the sensor 15 is mounted is 0°. Each scan layer has the same angular difference θ relative to other adjacent scan layers.

[0017] In this embodiment, the angular difference θ between adjacent scan layers is the same, but for example, the angular difference may be set to be different on the upper and lower sides of the reference scan layer L0. Furthermore, if the angular difference θ between adjacent scan layers is the same, the sensor may not have a reference scan layer L0. In such cases, the measurement value obtained by L0, which would be obtained if a reference scan layer L0 were present, can be estimated by calculating the midpoint of the distance between the object scanned by the L1 layer and the L-1 layer. In the following explanation, when specifying a scan layer, it may be denoted as scan layer Lx (where x is any integer). Also, when indicating the total number of scan layers, it may be denoted as Ln (where n is any integer).

[0018] As shown in Figure 1, the processing unit 20 is a computing device such as a GPU (Graphics Processing Unit). The memory unit 30 is an information recording medium such as an SSD (Solid State Drive), an HDD (Hard Disk Drive), and memory. The memory unit 30 stores map data 31 (high-precision 3D map: HD map). The map data 31 contains information about infrastructure structures such as roads and tunnels, and objects in the surrounding environment such as buildings, which are registered in association with coordinate data. Here, coordinate data is data that represents the position of infrastructure structures and objects in terms of latitude and longitude in a geodetic datum. Object information includes information about the object's height (known height information) and building identifiers. Road information includes the angle of inclination of the road relative to the horizontal line. Here, a building identifier is a unique identifier assigned to an infrastructure structure or building. An identifier is, for example, an identification label that arbitrarily combines identification numbers, symbols, codes, etc. Different identifiers allow for the identification, distinction, and differentiation of infrastructure structures and buildings. Note that the identifier may also be the name assigned to the infrastructure structure or building.

[0019] In addition to the map data 31, the memory unit 30 stores the programs 32 for each process performed by the processing unit 20, as well as data generated during the processing by the programs 32. The map data 31 is also stored on the external server ES.

[0020] Figure 3 is a functional block diagram of the processing unit 20 of the information processing device 1. The determination unit 209 has an arbitrary configuration, but for the sake of explaining the application examples described later, it is included in the functional block. The location information acquisition unit 200 acquires the location information of the vehicle V (moving object). Specifically, the GPS unit 13 calculates position information indicating the current position of vehicle V based on radio waves (GPS signals) received from multiple positioning satellites ST (see Figure 1). Position information is coordinate data that represents the current position of vehicle V in terms of latitude and longitude in a geodetic datum.

[0021] The scanning control unit 201 controls the irradiation of the IR irradiation area from the sensor 15 with laser light and controls the scanning of the IR irradiation area.

[0022] The object detection unit 202 detects objects located within the IR irradiation range. The object detection unit 202 receives measurement data (LiDAR data) from the sensor 15. The object detection unit 202 processes the measurement data (LiDAR data) to detect objects located within the IR irradiation range.

[0023] As described above, the laser light emitted by sensor 15 is reflected by the surface of an object within the IR irradiation range. Sensor 15 receives the laser light (reflected light) reflected from the surface of the object. Sensor 15 generates point cloud data from the received laser light (reflected light), which is a collection of points on the object's surface that reflected the laser light. Here, each point in the point cloud data corresponds to the surface of the object. Therefore, the shape of the object's surface that reflected the laser light can be measured from the distribution of each point that makes up the point cloud data and the distance to each point. Therefore, when point cloud data is generated, objects located within the IR (irradiated infrared) range are identified as a collection of points. Each point in the point cloud data contains distance information from the sensor 15 and angle information (angle information) relative to the surface on which the sensor 15 is installed (horizontal line). In this embodiment, the sensor 15 generates positional information (XYZ coordinate data in 3D space) for each point on the surface of an object, based on the distance and angle to each point on the surface of the object. The measurement data output by the sensor 15 includes point cloud data and distance information, angle information, and position information for each point that makes up the point cloud data.

[0024] Therefore, the object detection unit 202, which receives measurement data from the sensor 15, detects objects located within the IR irradiation range based on the point cloud data.

[0025] In this embodiment, the objects to be detected are assumed to be buildings such as apartment buildings and office buildings. For example, these buildings have straight edges (such as building corners and top edges), and these parts corresponding to straight edges are easily identified in the point cloud data. As an example, in this embodiment, an edge filter or the like is used to detect buildings within the IR irradiation range based on the presence or absence of areas where points are connected in a straight line.

[0026] The distance information acquisition unit 203 acquires the distance from the sensor 15 to an object located within the IR irradiation range. Each point in the point cloud data contains information (distance information) that identifies the distance from the sensor 15. The distance to each point on the surface of an object (each point in the point cloud data) is calculated from the time from the irradiation of the laser beam into the irradiated IR area to the reception of the reflected wave reflected by an object within the irradiated area. The distance information acquisition unit 203 processes the point cloud data to acquire the distance to building BL within the irradiation range IR, specifically the distance for calculating the building's coordinates (first distance), the distance from vehicle V to the building (second distance), and the distance from vehicle V to the top edge of the building (third distance).

[0027] Figure 4 illustrates the calculation principle of building height information H_Cal. Figure 4(A) shows the case where the vehicle V and the reference scan layer L0 are parallel to the horizontal line HL, and the pitch angle Pitchθ=0. Figure 4(B) shows the case where the pitch angle of the vehicle V is slightly tilted relative to the horizontal line HL due to irregularities on the road surface, i.e., the pitch angle Pitchθ≠0.

[0028] The distance used to calculate the building's coordinates (first distance: D1) is the distance from vehicle V to any point on building BL. As an example, in Figure 6, the distance to the corner (edge) closest to the vehicle V (sensor 15) in building BLa is calculated. Here, a "corner" of a building refers to the point where two walls of the building meet. In point cloud data, building corners are detected as areas where multiple points are connected in a roughly straight line vertically. In this case, "connected in a straight line" does not mean that the points are connected in a straight line from the top to the bottom of the building. It is sufficient that the points are connected for a length that allows for the detection of a straight edge.

[0029] Furthermore, as shown in Figure 7, it is preferable that the distance D1 to the building BLa to be calculated is the distance to a point on the building that is higher than any installations around the road, such as trees or guardrails (in the case of Figure 7, point TG). Above trees and guardrails, the likelihood of objects other than buildings reflecting laser light is low. Therefore, the presence of trees and guardrails can be less likely to affect the calculated distance to buildings. Here, the distance D1 used to calculate the building's coordinates does not necessarily have to be calculated at the building's edges. If the distance can be calculated, it may be calculated at the building's wall surfaces. Furthermore, if there are no objects other than buildings around the road, distance D1 may be taken as the shortest distance to the building, i.e., the distance to the intersection point TG' with the reference scan layer L0.

[0030] The distance D2 (second distance) to the building and the distance D3 (third distance) to the top edge of the building are referenced when the height information calculation unit 206 calculates the building's height information H_Cal. As described above, the sensor 15 emits laser light towards the illumination range IR in front of the vehicle V. The distance information acquisition unit 203 calculates the distance from the sensor 15 to the intersection point of the reference scan layer L0 and the building, i.e., the distance D2 from the sensor 15 to the building, from the reflected wave of the laser light emitted from the sensor 15 along the reference scan layer L0 (see Figure 4). When the vehicle V and the building are located on the same horizontal plane, the laser beam emitted from the sensor 15 along the reference scan layer L0 reaches the building located on the front side of the vehicle V within the illumination range IR via the shortest distance. Therefore, the distance D2 calculated from the reflected wave of the laser beam emitted from the sensor 15 along the reference scan layer L0 is the shortest distance from the vehicle V to the building when the pitch angle is 0°. Furthermore, since each point in the point cloud data is associated with information about the intersection angle with the reference scan layer, the position information of the intersection point between the reference scan layer L0 and the building in the sensor coordinate system is calculated simultaneously with the calculation of this distance D2.

[0031] The distance D3 to the top edge of the building is calculated by the distance information acquisition unit 203, which identifies the scan layer that generated the point cloud corresponding to the top edge of the building, and then uses the reflected wave of the laser light irradiated from the sensor 15 along the identified scan layer to calculate the distance from the sensor 15 to the intersection point of the identified scan layer and the top edge of the building, i.e., the distance D3 from the sensor 15 to the top edge of the building (see Figure 4). Furthermore, since each point in the point cloud data is associated with information about the intersection angle with the reference scan layer, the positional information of the intersection point between the identified scan layer and the building in the sensor coordinate system is calculated simultaneously with the calculation of this distance D3.

[0032] The building coordinate calculation unit 204 generates location information (building coordinates), i.e., coordinate data of the building (object), based on the location information of the vehicle V (moving object) acquired by the location information acquisition unit 200 and the distance D1 (first distance) to the object acquired by the distance information acquisition unit 203.

[0033] The height information acquisition unit 205 refers to the map data 31 based on the coordinate data and acquires the known height information H_Known of the detected building (object). In the map data 31, building identifiers and building height information (known height information) are stored in association with building coordinate data. By referring to the map data 31 based on the coordinate data calculated by the building coordinate calculation unit 204, it is possible to obtain the building identifier and the building height information (known height information) associated with the coordinate data. The height information acquisition unit 205 acquires the height information of a building registered in a record that matches the coordinate data (latitude and longitude in the geodetic datum) as known height information for the detected building. Furthermore, in map data 31, if a flag is set in association with a building identifier indicating that it cannot be used for height calculation, it may be excluded from obtaining known height information H_Known.

[0034] The height information calculation unit 206 calculates the building height information H_Cal detected within the irradiated IR range. Specifically, the height information calculation unit 206 identifies a scan layer Lx that generates a point cloud corresponding to the top edge of a building from which known height information H_Known has been obtained. (b) Obtain position information α at the intersection of the identified scan layer Lx and the top edge of the building from which known height information H_Known was obtained. (c) Obtain position information β at the intersection of the building from which known height information H_Known was obtained and the reference scan layer L0. (d) Calculate the building height information H_Cal from position information α, position information β, and the sensor installation height h on the vehicle V.

[0035] In the case of Figure 4(A), scan layer L4 becomes scan layer Lx, which generates a point cloud corresponding to the upper edge of building BL. Here, let point A be the emission point of the laser beam from sensor 15, point B be the intersection point with building BL in the direction along the reference scan layer L0 (reflection point of the laser beam), point C be the intersection point with the upper edge of building BL in the direction along the scan layer L4 (reflection point of the laser beam), and let the intersection angle ∠BAC between the reference scan layer L0 and the scan layer L4 be the intersection angle θx.

[0036] The height information calculation unit 206 obtains position information α of the intersection point C between the scan layer L4 and the building BL from the measurement data (point cloud data) input from the sensor 15. In position information α, the position information of the intersection point C in the sensor coordinate system is defined as XYZ coordinate data. Furthermore, the height information calculation unit 206 acquires position information β of the intersection point B between the reference scan layer L0 and the building BL from the measurement data (point cloud data) input from the sensor 15. In position information β, the position information of the intersection point B in the sensor coordinate system is defined as XYZ coordinate data.

[0037] Here, positional information α and β are coordinate data in a three-dimensional space based on the position of sensor 15. In other words, they are data from the same sensor coordinate system. Therefore, the distance Dx between points B and C in Figure 4(A) can be calculated simply by taking the difference between the Z component coordinates of position information α and the Z component coordinates of position information β. Specifically, the distance Dx can be calculated from the following formula (1), which subtracts location information β from location information α. Dx = Z coordinate of position information α - Z coordinate of position information β ... (1)

[0038] Here, considering the installation height h of the sensor 15 in vehicle V, the building height information H_Cal can be expressed by the following equation (2). H_Cal=Dx+h ···(2)

[0039] In this way, the height information calculation unit 206 calculates the building height information H_Cal detected within the irradiation range IR from (a) the position information α of the intersection point C with the building BL, which is generated from the scan layer Lx that generates a point cloud corresponding to the top edge of the building from which known height information H_Known was obtained, (b) the position information β of the intersection point B with the reference scan layer L0 and the building BL, and (c) the sensor installation height h on the vehicle V. Since the installation height of sensor 15 is small compared to the height of building BL, h can be considered zero if accuracy is not a concern.

[0040] Here, in Figure 4(A), if vehicle V is located on the horizontal line HL and the pitch angle, which is the inclination of vehicle V with respect to the horizontal line, is zero degrees (=0°), then the relationship between the known height information H_Known of the building and the calculated height information H_Cal of the building can be expressed by the following equation (3). H_Known = H_Cal ... (3)

[0041] On the other hand, as shown in Figure 4(B), if the vehicle V is tilted relative to the horizontal line HL due to road surface irregularities or the like, a difference H occurs between the known building height information H_Known and the calculated building height information H_Cal, corresponding to the pitch angle Pitchθ of the vehicle V. In this case, equation (3) above can be expressed as equation (4) below. H_Known = H_Cal + H ... (4)

[0042] From equation (4) above, the difference H can be expressed by the following equation (5). H = H_Known - H_cal ... (5)

[0043] The pitch angle calculation unit 207 calculates the pitch angle Pitchθ of the vehicle V using the difference H (see (B) in Figure 4) between the known building height information H_Known and the calculated building height information H_Cal.

[0044] The pitch angle Pitchθ when the vehicle V is tilted relative to the horizontal can be expressed by the following equation (6). Sin(Pitchθ) = H / D² ... (6) From equation (6), the pitch angle (Pitchθ) can be expressed by the following equation (7). Pitchθ = sin -1 (H / D2) ···(7)

[0045] From equation (5) above, H is a value calculated from the known height information of the building H_Known and the calculated height information of the building H_Cal. The known height information of the building, H_Known, is obtained from the map data 31 stored in the memory unit 30. The calculated height information of the building, H_Cal and D2, is obtained from the measurement data (point cloud data) of the sensor 15.

[0046] The map data update unit 208 connects to the external server ES via the communication unit 14, and if newer map data exists than the map data 31 stored in the storage unit 30, it downloads the new map data from the external server ES and updates the map data 31 stored in the storage unit 30.

[0047] The functions realized by the components of the functional block of the processing unit 20 described above are realized by the program 32 stored in the storage unit 30 and the program (not shown) stored in the external server ES. However, the functions realized by the processing unit 20 are not limited to the above-described embodiment.

[0048] An example of processing in the information processing device 1 according to this embodiment will be described below. Figure 5 is a flowchart illustrating the pitch calculation process performed by the information processing device 1. Figure 6 is a schematic diagram illustrating the building detection process in the IR irradiation range. Figure 7 is a schematic diagram illustrating the calculation of the distance D1 to the building detected within the IR irradiation range.

[0049] The information processing device 1 repeatedly performs a pitch calculation process at predetermined execution intervals while the vehicle V equipped with the information processing device 1 is in motion.

[0050] As shown in Figure 5, in the pitch calculation process, in step S101, the position information acquisition unit 200 acquires position information indicating the current position of the vehicle V (moving object). In step S102, the scanning control unit 201 uses the sensor 15 to irradiate laser light into the irradiation range IR set in front of the vehicle V and performs scanning of the irradiation range IR (step S102). When sensor 15 receives laser light (reflected wave) reflected from the surface of an object within its IR irradiation range, it generates point cloud data from the received laser light. The point cloud data is a collection of points (measurement points) that reflect the laser light from the surface of the object. The sensor 15 then outputs measurement data, including the generated point cloud data. Here, the point cloud data can be associated with the following information. (a) distance to each point that makes up the point cloud data, (b) The angle of each point that makes up the point cloud data (angle with respect to the horizontal line passing through sensor 15), (c) Position information of each point that makes up the point cloud data (XYZ coordinate data in the sensor coordinate system).

[0051] When measurement data is input from the sensor 15 (step S103, Yes), the object detection unit 202 checks for the presence or absence of a building within the IR irradiation range by processing the point cloud data. If the presence of at least one building is confirmed within the irradiated IR range (step S104, Yes), in step 105, the distance information acquisition unit 203 calculates the distance D1 (first distance) to the detected building.

[0052] In the case of Figure 6, building BLa is located within the IR irradiation range. Therefore, the object detection unit 202 detects the presence of building BLa through the processing in step S103. As shown in Figure 7, in step S105, the distance D1 to the building BLa is calculated in order to obtain the location information of the building BLa. Therefore, the measurement point TG on the building BLa is set to an arbitrary height position that is not affected by obstacles such as surrounding trees and guardrails. This allows the distance D1 to the measurement point TG on the building BLa to be calculated.

[0053] In step S106, the building coordinate calculation unit 204 calculates the coordinates of the detected building based on the position information of the vehicle V identified in step S101 and the distance D1 to the building calculated in step S105. In the case of Figure 7, the coordinates of the building BLa in the geodetic reference system are calculated.

[0054] In step S107, the height information acquisition unit 205 refers to the map data 31 based on the calculated coordinates of the building and acquires the known height information H_Known of the detected building.

[0055] In step S108, the height information calculation unit 206 calculates the height information H_Cal of the detected building by processing the point cloud data.

[0056] Figure 8 is a flowchart illustrating the calculation process for the detected building height information, H_Cal. The height information calculation unit 206 calculates the building height information H_Cal by following the procedure below. First, the height information calculation unit 206 determines the building for which the height information H_Cal will be calculated (step S201). Specifically, it determines the building for which known height information H_Known has been obtained as the building for which height information will be calculated. In Figure 6, only one building BLa has been detected, so building BLa is determined to be the building for which height information will be calculated.

[0057] Next, the height information calculation unit 206 searches the point cloud data corresponding to the building for which height information is calculated for the sequence of points corresponding to the top edge of the building to identify the top edge of the building (step S202). The height information calculation unit 206 identifies the scan layer Lx that generated the sequence of points corresponding to the top edge of the building (step S203). As mentioned above, the IR irradiation area is divided into multiple sections in the vertical direction. Therefore, we need to check which of the multiple scan layers Ln (where n is an arbitrary integer) detected the sequence of points corresponding to the top edge of the building. In the case of Figure 2, scan layer L4 coincides with the top edge of the building, and scan layer Lx detects the sequence of points corresponding to the top edge of the building.

[0058] The height information calculation unit 206 obtains the position information α of the intersection point between the identified scan layer Lx and the building from the measurement data (point cloud data) input from the sensor 15 (step S204). In the case of Figure 4(A), the positional information of the intersection point C between the identified scan layer L4 and building BL, i.e., the XYZ coordinate data of the sensor coordinate system, is acquired.

[0059] Next, the height information calculation unit 206 obtains positional information β of the intersection point between the reference scan layer L0 and the building from the measurement data (point cloud data) input from the sensor 15 (step S205). In the case of (A) in Figure 4, the positional information of the intersection point B between the reference scan layer L0 and the building BL, i.e., the XYZ coordinate data of the sensor coordinate system, is acquired.

[0060] The height information calculation unit 206 calculates the detected building height information H_Cal using the position information α of the intersection point between the identified scan layer Lx and the building, the position information β of the intersection point between the reference scan layer L0 and the building, and the installation height h of the sensor 15 (step S206). Specifically, since location information α and location information β are coordinate data (XYZ coordinate data) in the same sensor coordinate system, the difference between the Z coordinate of location information α and the Z coordinate of location information β corresponds to the height distance between point B and point C. Therefore, the distance Dx is calculated from the following formula (1), which subtracts the location information β from the location information α, and the building height information H_Cal is calculated by adding the installation height h of the sensor 15 to this (see formula (2) below). Dx = Z coordinate of position information α - Z coordinate of position information β ... (1) H_Cal=Dx+h ···(2)

[0061] Returning to the flowchart in Figure 5, once the building height information H_Cal is calculated, in step S109 the pitch angle calculation unit 207 calculates the pitch angle Pitchθ based on the known height information H_Known obtained in step S107 and the building height information H_Cal calculated in step S108.

[0062] Here, as shown in Figure 4(A), when the pitch angle of vehicle V is 0°, that is, when the vehicle is horizontal, the building height information H_Cal calculated by equation (2) above matches the known building height information H_Known. In this case, the pitch angle calculation unit 207 calculates "zero (=0)" as the pitch angle.

[0063] On the other hand, if the vehicle is not level, as shown in Figure 4(B), a difference H occurs between the vehicle and the known building height information H_Known, corresponding to the pitch angle. If there is a difference H, the pitch angle calculation unit 207 uses the difference H to calculate the pitch angle. Here, the difference H is expressed by the following equation (5). H = H_Known - H_cal ... (5)

[0064] Here, the pitch angle Pitchθ when the vehicle V is tilted relative to the horizontal can be expressed by the following equation (6). Sin(Pitch) = H / D² ... (6) From equation (6), the pitch angle (Pitchθ) can be expressed by the following equation (7). Pitchθ = sin -1 (H / D2) ···(7)

[0065] Thus, the difference H that arises from the tilt of the vehicle V is calculated from the known height information H_Known, the calculated height information H_Cal, and the distance D2. Therefore, since this information can be obtained from the distance to the building (distances D1, D2, D3) acquired by the sensor 15 and the location information of the building acquired using distance D1, the pitch angle Pitchθ can be calculated using the sensor 15 mounted on the vehicle V. Therefore, as described using Figures 4(A) and 4(B), the pitch angle Pitchθ of the vehicle V is calculated from the distance to the building (distances D1, D2, D3) acquired by sensor 15, the coordinate data of intersections B and C calculated based on distances D2 and D3, and the known height information H_Known acquired based on these.

[0066] [Example 1] Figures 9 and 10 illustrate the calculation of the distance D1 (D1a to D1c) when three buildings are detected within the IR irradiation range. In the above embodiment, an example was given in which one building is detected within the irradiated IR range. The following explanation will describe the case where multiple buildings are detected within the IR irradiation range, using the example of three buildings being detected. For example, as shown in Figures 9 and 10, if there are three buildings BLa, BLb, and BLc within the irradiated IR range, in step S105 of the pitch calculation process described above (see Figure 5), the distance D1 (D1a to D1c) to a point (point TG) on each of the three buildings BLa, BLb, and BLc is calculated. Then, in step S106, the coordinates of the three buildings BLa, BLb, and BLc are calculated from the position information of the vehicle V and the distances D1 (D1a to D1c) to each of the three buildings BLa, BLb, and BLc obtained in step S105. This retrieves the known height information H_Known for each of the three buildings, BLa, BLb, and BLc. Then, in step S108, for each of the three buildings BLa, BLb, and BLc, the building height information H_Cal is calculated from the measurement data (point cloud data) of sensor 15.

[0067] In step S109, the pitch angle of vehicle V is calculated using one of the following methods: (a) The average of the three pitch angles calculated from each of the three buildings BLa, BLb, and BLc is used as the pitch angle calculated by the pitch angle calculation unit 207. (b) Of the three pitch angles calculated from each of the three buildings BLa, BLb, and BLc, the pitch angle with the largest difference from the average value is excluded, and the average of the remaining two pitch angle differences is used as the pitch angle calculated by the pitch angle calculation unit 207.

[0068] This is expected to improve the accuracy of the calculated pitch angle Pitchθ compared to using a pitch angle Pitchθ calculated from only one building. Furthermore, since it can improve the accuracy of the pitch angle without using historical information such as moving averages, it is particularly useful for advanced driver assistance and autonomous driving systems where there are strict time constraints on input values.

[0069] In Modification Example 1, an example was given in which the building height information H_Cal is calculated from the measurement data (point cloud data) of sensor 15 for each of the three detected buildings BLa, BLb, and BLc. However, for at least two buildings, the building height information H_Cal may be calculated from the measurement data of sensor 15. For example, if four or more buildings are detected, reducing the total number of buildings for which building height information H_Cal is calculated can reduce the load on the calculation process in the processing unit 20. In this case, the buildings for which height information H_Cal is calculated can be narrowed down by methods such as excluding buildings that are close to vehicle V.

[0070] [Differentiation 2] Figure 11 illustrates another embodiment for identifying the scan layer Lx that generated the sequence of points corresponding to the top edge of a building when a scanning LiDAR sensor is applied.

[0071] In the above embodiment, an example was given in which a scan layer Lx that generated a sequence of points corresponding to the top edge of a building was identified from measurement data (point cloud data) acquired at a certain time. As mentioned above, the pitch calculation process is executed repeatedly at predetermined time intervals. Therefore, the measurement data (point cloud data) is also generated sequentially at predetermined time intervals. As shown in Figure 11, when the vehicle is moving towards building BL, the scan layer Lx that generated the sequence of points corresponding to the top edge of the building switches sequentially. Specifically, as the vehicle approaches the building, the scan layer Lx that generated the sequence of points corresponding to the top edge of the building switches. Therefore, the intersection angle between the identified scan layer Lx and the reference scan layer L0 increases each time the scan layer Lx switches.

[0072] In the second modification, as the vehicle moves, the scan layer Lx that generates the arrangement corresponding to the top edge of the building switches, and at that timing, the building height information H_Cal is calculated, and the pitch angle Pitchθ is calculated.

[0073] In the point cloud data calculated from the measurement data at the timing (time tn) shown in Figure 11(A), the scan layer Ln reflects at a position below the top edge of the building. Furthermore, the scan layer Ln+1, which is one level above scan layer Ln, passes above the top edge of the building. Therefore, in the point cloud data at this time, a position lower than the actual top of the building is identified as the top of the building. Consequently, the calculated building height information will contain errors.

[0074] If the building is located on the side of the direction of vehicle V's movement, then at another time (time tn+1) when vehicle V is closer to the building, the scan layer Ln+1 that previously passed over the top of the building will be reflected by the top edge of the building. Therefore, by taking advantage of this timing to identify the position of the top edge of the building and calculating the building's height information H_Cal, it becomes possible to calculate the building's height more accurately.

[0075] Therefore, by modifying step S108 (Figure 5), which calculates the height of the building as described above, it becomes possible to calculate the height of the building more accurately. Figure 12 is a flowchart illustrating the steps for calculating the building height in the case of Modification 2. As shown in Figure 12, the height information calculation unit 206 determines the building for which to calculate height information H_Cal in the point cloud data generated from measurement data acquired at a certain timing (time tn) (step S301). The building BL in Figure 11 is determined to be the building for which to calculate height information.

[0076] The height information calculation unit 206 determines the building for which height information is to be calculated (step S301), and then searches the point cloud data of the building for which height information is to be calculated for the sequence of points corresponding to the top edge of the building to identify the top edge of the building (step S302). The height information calculation unit 206 identifies the scan layer Lx that generated the sequence of points corresponding to the top edge of the building (step S303).

[0077] In this modified version, the scan layer Lx is identified multiple times before determining the scan layer Lx used to calculate the building height information H_Cal. Therefore, after the initial scan layer is identified (step S303), when newly acquired measurement data is input (step S304, Yes), the height information calculation unit 206 sets the building determined in step S301 again as the building for which height information is calculated (step S305).

[0078] Then, the height information calculation unit 206 searches the input measurement data (point cloud data) for a sequence of points corresponding to the top edge of the building, and the top edge of the building is identified (step S306). Then, the scan layer Lx that generated the sequence of points corresponding to the newly identified top edge of the building is identified (step S307).

[0079] The height information calculation unit 206 compares the old and new scan layers Lx (step S308). Specifically, it compares the previously identified scan layer Lx with the newly identified scan layer Lx. Then, if the newly identified scan layer Lx is a different scan layer from the previously identified scan layer Lx (step S309, Yes), the height of the building is calculated using a point (intersection) on the newly identified scan layer Lx. It is preferable that the process shown in Figure 12 is repeatedly performed at predetermined time intervals in conjunction with the pitch calculation process.

[0080] Here, we will specifically explain the case of Modification 2 using the example in Figure 11. As vehicle V moves towards the building, the scan layers Lx that receive the laser light reflected from the top edge of the building switch sequentially. In the case of Figure 11, at a certain time (time tn), scan layer Ln is the scan layer that is illuminated by laser light reflected from the top edge of the building. At another time after time tn (time tn+1), scan layer Ln+1 is the scan layer that is illuminated by laser light reflected from the top edge of the building. Therefore, as vehicle V moves, and as vehicle V approaches, the scanning layer that emits laser light reflected from the top of the building will switch at a certain point. In this modified version, at the timing of this switchover, the newly determined scan layer is selected as the scan layer used to calculate the building's height information.

[0081] As a result, the height information calculation unit 206 acquires position information α of the intersection point between the newly identified scan layer Lx and the building (step S310) and position information β of the intersection point between the reference scan layer L0 and the building (step S311). Then, the height information calculation unit 206 calculates the detected building height information H_Cal using the position information α of the intersection point between the identified scan layer Lx and the building, the position information β of the intersection point between the reference scan layer L0 and the building, and the installation height h of the sensor 15 (step S312).

[0082] Thus, in Modification 2, we demonstrated an example in which, when a vehicle is moving towards a building, the timing of the switching of the scanning layer of the laser light reflected from the top edge of the building is captured to determine the position of the top edge of the building and calculate the building's height information H_Cal. This allows for accurate calculation of the building's height by identifying the location of the building's top edge, provided that at least two scan layers are located above the reference scan layer L0 within the laser beam's IR irradiation range. While the accuracy of building height calculation improves with a larger total number of scan layers, the error is significant when using LiDAR with a limited number of scan layers, making the accuracy improvement achieved by Modification 2 particularly substantial.

[0083] In variation 2, an example was given of calculating the building height by capturing the timing of the scan layer switching. Here, if, after the initial identification of the scan layer (step S303), the scan layer does not switch and the identification of the scan layer (step S307) is performed a predetermined number of times, the intersection angle between the initially identified scan layer and the reference scan layer L0 may be calculated (step S310) to calculate the building height information H_Cal.

[0084] When vehicle V is traveling at low speeds, the switching of the scan layer is slow, so it is preferable to be able to calculate the pitch angle Pitchθ even in such cases.

[0085] [Difference 3] Figure 13 is a diagram illustrating the scanning layer in modified example 3. In the above embodiment, an example was given in which the scan layers above the reference scan layer L0 are set with the same angular difference. For example, as shown in Figure 13, the angular difference θ2 of the scan layers in the upper region of the irradiation range IR may be made smaller than the angular difference θ1 of the scan layers in the lower region of the irradiation range IR, so that the density of scan layers in the upper region of the building is greater than in the lower region. In such cases, the scan layer Lx that generated the sequence of points corresponding to the top edge of the building can be identified more accurately. In particular, when combined with the aforementioned modified example 2, an improvement in the accuracy of the calculated building height information H_Cal can be expected.

[0086] [Differentiation Example 4] Figure 14 illustrates another example of pitch angle calculation. In the above-described embodiment, as shown in Figures 4(A) and 4(B), an example of calculating the pitch angle Pitchθ was provided, assuming that the vehicle V and the building BL are on approximately the same plane. The present invention is also applicable when the vehicle V and the building BL are not on the same plane, for example, when the building BL is located on a road surface GD that is aligned with the horizontal line, and the vehicle V is traveling uphill on a slope SL toward the road surface GD.

[0087] For example, when vehicle V is traveling uphill on slope SL, there is a timing when the sensor 15 mounted on vehicle V intersects with a straight line LV that runs along the road surface GD in the vertical direction. At this point, the installation height h of the sensor 15 can be treated as zero (=0). Then, the building height information H_Cal calculated from the position information α of the intersection point between the scan layer Lx and the building at that point, and the position information β of the intersection point between the reference scan layer L0 and the building, is: The following relationship (8) holds between the known building height information H_Known and the value H determined according to the pitch angle described above. H = H_Known - Dx ... (8)

[0088] The pitch angle Pitchθ of the vehicle V at this time can be expressed by the following equation (6). Sin(pitch) = H / D² ... (6) From equation (6), the pitch angle (Pitchθ) can be expressed by the following equation (7). Pitchθ = sin -1 (H / D2) ···(7)

[0089] In the case of Figure 14, the pitch angle (Pitchθ) can also be calculated using the following formula (9). Pitchθ = Tan -1 (H_known-Dx / D4) ···(9) Here, D4 ​​is the distance from sensor 15 to the intersection point E of the straight line LV and building BL. The straight line LV is a horizontal line connecting sensor 15 and the lower edge of building BL.

[0090] Here, due to the influence of obstacles such as surrounding trees and guardrails, the bottom of building BL may not be directly measured by sensor 15, in which case the distance D4 between vehicle V and the bottom of building BL on a straight line parallel to the horizontal plane cannot be calculated. Furthermore, the timing at which D4 can be calculated based on the sensor 15 mounted on vehicle V while vehicle V is sloping towards building BL is limited. Therefore, it is preferable to estimate the pitch angle Pitchθ under the conditions of (A) and (B) in Figure 4 rather than under the conditions in Figure 14.

[0091] [Application Example 1] The embodiments and modifications described above illustrate the calculation of the vehicle's pitch angle, Pitchθ. Examples of applications of the calculated pitch angle, Pitchθ, are described below.

[0092] For example, the calculated pitch angle Pitchθ can be used to determine the validity of the pitch angle generated by the IMU12 mounted on the vehicle V. As a functional block that performs such decision processing, the processing unit 20 of the information processing device 1 may also include a determination unit 209 (see Figure 3).

[0093] For the purposes of this explanation, the pitch angle Pitchθ calculated by the pitch angle calculation unit 207 will be referred to as pitch angle B, and the pitch angle Pitchθ generated by the IMU 12 will be referred to as pitch angle A.

[0094] The determination unit 209 calculates the difference Δθ between the pitch angle B calculated by the pitch angle calculation unit 207 and the pitch angle A generated by the IMU 12. If the calculated difference △θ is less than the threshold Th, it can be determined that the pitch angle generated by IMU12 is appropriate. Therefore, if the pitch angle generated by IMU12 is used, for example, as a parameter of the vehicle V's automatic driving control system (not shown), and automatic driving control of the vehicle V is implemented, Furthermore, if the calculated difference △θ is greater than or equal to the threshold Th, it can be determined that the pitch angle generated by the IMU 12 is inappropriate. In such cases, the pitch angle calculated by the pitch angle calculation unit 207 is used as a parameter for the vehicle V's automatic driving control device (not shown), and automatic driving control of the vehicle V is performed.

[0095] Furthermore, if the calculated difference △θ is greater than or equal to the threshold Th, it may be possible to determine which of pitch angles A and B is inappropriate and then use the appropriate pitch angle. For example, the difference between the previous value and the latest value for each of pitch angles A and B may be checked, and the one in which a sudden change in value is observed may be determined to be inappropriate. Examples of using an appropriate pitch angle are not limited to the automated driving control described above. For example, it can be used to control the tilt of a vehicle equipped with active dampers, or to adjust the tilt of a cargo compartment in a vehicle that uses air suspension to support the cargo compartment. Furthermore, if pitch angle B is deemed inappropriate, the map provider may be provided with a statement indicating that the known height information H_Known should be updated, or with at least one of the building height information calculated backward from pitch angle A.

[0096] As described above, the information processing device 1 according to this embodiment has the following configuration. (1) The information processing device 1 is A location information acquisition unit 200 acquires location information of the vehicle V (moving object), A distance information acquisition unit 203 acquires the distance from the vehicle V to the building (object), A height information acquisition unit 205 acquires known height information H_Known of a building based on the position information of the vehicle V and the distance D1 (first distance) to the building. A height information calculation unit 206 calculates the building height information H_Cal based on position information α and β determined according to the distance D2 (second distance) and D3 (third distance) from the vehicle V to the building, The system includes a pitch angle calculation unit 207 that calculates the pitch angle Pitchθ of the vehicle V using known building height information H_Known and building height information H_Cal.

[0097] The pitch angle Pitchθ can be easily calculated using the position information of vehicle V, the distance D1 from vehicle V to the building, and position information α and β determined by the distances D2 and D3 from vehicle V to the building. Since the position information of vehicle V and the distance from vehicle V to the building can be obtained by existing equipment (sensor 15) mounted on vehicle V, the pitch angle Pitchθ of vehicle V can be calculated without requiring any additional equipment. The easily calculated pitch angle Pitchθ can be used in place of the pitch angle obtained by, for example, an inertial measurement unit (IMU12), thus providing redundancy in pitch angle acquisition.

[0098] (2) The information processing device 1 is A sensor 15 (measuring instrument) irradiates laser light (transmitted wave) towards a set irradiation range IR and receives reflected waves reflected by objects within the irradiation range IR, The system includes an object detection unit 202 that detects buildings (objects) within the IR irradiation range based on point cloud data generated from reflected waves received by the sensor 15. The distance information acquisition unit 203 calculates the distances D1, D2, and D3 to the detected buildings from the point cloud data of the detected buildings.

[0099] Each point in the point cloud data has a coordinate value within the IR (irradiation range), allowing the position of an object within the IR to be calculated. From the position of the building within the IR, the distance D1 from the vehicle V to the building can be calculated. Then, the position of the building (object) can be identified from the position information of the vehicle V acquired by the position information acquisition unit 200 and the distance D1 to the building. As a result, the position of the building on the map data can be calculated based on the position information of the vehicle V, so for example, by referring to the map data, the height information of the building within the IR can be obtained.

[0100] (I) Sensor 15 is a LiDAR sensor that irradiates the IR irradiation range with laser light and receives the laser light (reflected wave) reflected by an object within the IR irradiation range.

[0101] LiDAR sensors are frequently installed measuring instruments on vehicles (V). Using a LiDAR sensor mounted on a vehicle, height information of buildings within the illumination range (IR) can be acquired. Since the vehicle's pitch angle can be calculated without the need for separate dedicated equipment, it effectively prevents an increase in the vehicle's cost due to the need for pitch angle calculation.

[0102] (3) The IR irradiation range is divided into multiple scan layers Ln (where n is an integer) in the vertical direction based on the installation state of the sensor 15 on the vehicle V. The height information calculation unit 206 is: The coordinates of the first intersection point (intersection C in Figure 4(A)) between the scan layer Lx, which transmitted the reflected wave reflected off the top of the building, and the target object, building BL, and The reference scan layer L0 along the horizontal line based on the installation state of the sensor 15 on the vehicle V when the vehicle V is positioned on the horizontal plane, and the coordinate values ​​of the second intersection point (intersection point B in Figure 4(A)) with the target object, building BL, The height information of the building, H_Cal, is calculated using the installation height h of sensor 15 on vehicle V.

[0103] The point cloud data, which is the measurement data from sensor 15, allows us to obtain (a) the distance to each point that makes up the point cloud data, (b) the position information (coordinate value) of each point in the sensor coordinate system, and (c) the angle of each point relative to the reference scan layer L0. The laser beam irradiation angle when calculating the distance D2 to the building in the direction along the reference scan layer L0 is in the direction along the horizontal plane based on the mounting state of the sensor 15 on the vehicle V when the vehicle V is located on the horizontal plane, and is 0° when the pitch angle Pitchθ of the vehicle V is 0°. The coordinate value determined by the distance D2 to the building in the direction along the reference scan layer L0 is the position information β of the intersection B between the reference scan layer L0 and the building, and the coordinate value of the top edge of the building determined by the distance D3 to the top edge of the building is the position information α of the intersection C between the scan layer Lx passing through the top edge of the building and the building. Since the position information β at intersection B and the position information α at intersection C are values ​​in the same sensor coordinate system, the building height information H_Cal can be calculated from the difference between the Z component of the position information β at intersection B and the Z component of the position information α at intersection C, and the installation height h of sensor 15. The calculated building height information H_Cal theoretically corresponds to the building height information H_Known when the pitch angle of the moving object is 0°. Since the difference H between the calculated building height information H_Cal and the known building height information H_Known differs depending on the pitch angle of the moving object (the inclination of the moving object relative to the horizontal), the calculated building height information H_Cal can be used to calculate the pitch angle of the moving object.

[0104] (4) At least two scan layers Ln are set above a reference scan layer L0 that follows a horizontal line based on the installation state of the sensor 15 (measuring instrument) on the vehicle V when the vehicle V is located on a horizontal plane.

[0105] Increasing the number of divisions in the IR irradiation range improves the accuracy of calculating the height to the top edge of the object.

[0106] (5) The height information calculation unit 206 repeatedly identifies the scan layer Lx that transmitted the laser light reflected from the top of the building. If the newly identified scan layer Ln+1 is a different layer from the previously identified scan layer Ln, the coordinates of the first intersection with the target building BL are calculated using the newly identified scan layer Ln+1 (see Figure 11).

[0107] As vehicle V moves towards the building, the scan layers Lx that receive the laser light reflected from the top edge of the building switch sequentially. In the case of Figure 11, at a certain time (time tn), scan layer Ln is the scan layer illuminated by laser light reflected from the top edge of the building. At another time after time tn (time tn+1), scan layer Ln+1 is the scan layer illuminated by laser light reflected from the top edge of the building. Therefore, as vehicle V moves, and as vehicle V approaches, the scanning layer that emits laser light reflected from the top of the building will switch at a certain point. As vehicle V moves, the scan layer Lx that generates the arrangement corresponding to the top edge of the building switches, and at the timing of this switch, the building height information H_Cal and the pitch angle Pitchθ are calculated. This is expected to improve the accuracy of the calculated building height information H_Cal and pitch angle Pitchθ.

[0108] (7) If there are multiple buildings (objects) within the irradiation range IR, the distance information acquisition unit 203 acquires the distance D1 to at least two of the buildings. The height information acquisition unit 205 acquires known height information H_Known for each building whose distance D1 has been acquired. The height information calculation unit 206 calculates the height information H_Cal for each building whose distance D1 has been obtained. The pitch angle calculation unit 207 calculates the pitch angle Pitchθ for each building from which the distance D1 has been obtained, using the known height information H_Known and the height information H_Cal, and sets the average value of the calculated pitch angles Pitchθ as the pitch angle of the vehicle V (moving body).

[0109] If multiple buildings exist within the IR irradiation range, the accuracy of pitch angle calculation can be improved by calculating the pitch angle from each building and using the average of the calculated pitch angles as the pitch angle of the moving object.

[0110] (I) The pitch angle calculation unit 207 calculates the pitch angle Pitchθ for each building from which the distance D1 has been obtained, using known height information H_Known and height information H_Cal. Of the calculated pitch angles Pitchθ, one Pitch θ that differs significantly from the average Pitch θ is excluded, and the average of the remaining Pitch θ is taken as the pitch angle of the vehicle V (moving body).

[0111] Since pitch angles Pitchθ that deviate significantly from the average value are excluded, the pitch angle of the moving object, which is the average of the remaining pitch angles Pitchθ, becomes more accurate.

[0112] (8) If there are multiple buildings (objects) within the irradiation range IR, the distance information acquisition unit 203 acquires the distance D1 to at least two of the buildings. The height information acquisition unit 205 calculates known height information H_Known for buildings from which distance D1 has been acquired, for which known height information H_Known can be obtained. The height information calculation unit 206 calculates the height information H_Cal for each building from which known height information H_Known has been obtained. The pitch angle calculation unit 207 calculates the pitch angle Pitchθ for each building from which known height information H_Known has been obtained, using the known height information H_Known and the height information H_Cal.

[0113] For example, buildings whose top edge is not straight, making it difficult to detect the position of the top edge, are registered in the map data 31 as buildings for which known height information H_Known cannot be obtained, while buildings whose top edge is easy to detect are registered in the map data 31 as buildings for which known height information H_Known can be obtained. This allows the system to select the building whose top edge is easiest to detect when multiple buildings are present within the IR irradiation range, and to acquire the known height information H_Known. This makes it easier to determine which building's height information H_Cal is easiest to acquire, thus improving the accuracy of the calculated height information H_Cal. Furthermore, a reduction in errors in calculating height information H_Cal is expected, allowing for faster and more accurate calculation of the pitch angle. Furthermore, buildings that are located on steep slopes or for other reasons where the distance between the road surface and the top edge changes significantly depending on the choice of building corners, resulting in errors that cannot be ignored, can also be registered in the map data 31 as buildings that are difficult to use.

[0114] (II) The IR illumination range is divided into multiple scan layers Ln (where n is an integer) in the vertical direction, based on the installation state of the sensor 15 on the vehicle V. In the IR illumination range, the density of the scan layers in the upper region in the vertical direction is higher than the density of the scan layers in the lower region.

[0115] This configuration allows for more accurate identification of the scan layer Lx, which generates the sequence of points corresponding to the top edge of the building. This is expected to improve the accuracy of the calculated building height information H_Cal.

[0116] (III) The pitch angle calculation unit 207 is: The pitch angle Pitchθ is calculated based on the known building height information H_Known, the calculated building height information H_Cal, and the distance D2 to the building in the direction along the reference scan layer L0.

[0117] By making the known height information of the building H_Known obtainable from the map data 31, the pitch angle Pitchθ can be calculated simply by calculating the building height information H_Cal and the distance D2 to the building in the direction along the reference scan layer L0 from the measurement data (LiDAR data) of the sensor 15 (LiDAR sensor). As a result, the only equipment that needs to be added to the vehicle V is the LiDAR sensor, and since LiDAR sensors are frequently installed on vehicles V, the cost incurred for calculating the pitch angle can be reduced in the case of vehicles that already have a LiDAR sensor installed.

[0118] Furthermore, the difference H between the calculated building height information H_Cal and the known height information H_Known differs depending on the pitch angle of the vehicle V (moving body) (the inclination of the moving body relative to the horizontal line). Therefore, by calculating the difference H from the known height information H_Known, and using the calculated difference H and the distance D2 to the building in the direction along the reference scan layer L0, the pitch angle Pitchθ of the vehicle V, which caused the difference H, can be calculated.

[0119] (IV) It has a storage unit 30 (storage unit) that stores map data 31 which associates the location information of a building (object) with the known height information of the building H_Known. The height information acquisition unit 205 calculates the building's location information from the vehicle V's location information and the distance D2 from the vehicle V to the building, and obtains the known height information H_Known of the building from the map data 31 based on the calculated location information.

[0120] By preparing map data 31 that associates building location information with known building height information H_Known, it is possible to obtain known building height information for buildings around vehicle V.

[0121] (V) An IMU12 that functions as an attitude sensor to measure the pitch angle of the vehicle V, It includes a determination unit 209 that determines whether the pitch angle output from the IMU 12 is appropriate, If the determination unit 209 determines that the pitch angle input from the IMU 12 is inappropriate, it replaces the pitch angle calculated by the pitch angle calculation unit 207 with the pitch angle input from the IMU 12.

[0122] With this configuration, if the IMU12 malfunctions as an attitude sensor, the pitch angle calculated by the IMU12 can be replaced with the pitch angle calculated by the pitch angle calculation unit 207, thereby improving the redundancy of the pitch angle calculation. Furthermore, if a device that controls the automatic driving of the vehicle using the pitch angle is installed, it is possible to effectively prevent interference with the automatic driving by such device when there is a problem with the IMU12's function as an attitude sensor.

[0123] (VI) The determination unit 209 uses the pitch angle calculated by the pitch angle calculation unit 207 to determine whether the pitch angle input from the IMU 12 is appropriate.

[0124] With this configuration, it is possible to determine whether the pitch angle calculated by IMU12 is appropriate. If there is a problem with the pitch angle calculated by IMU12, it can be replaced with the pitch angle calculated by the pitch angle calculation unit 207. This improves the redundancy of the pitch angle calculation.

[0125] [Differentiation Example 4] In the above embodiment, an example was given in which the calculation of building height information H_Cal detected within the irradiation range IR is performed using position information α of the intersection point between the scan layer Lx which identifies the top edge of the building and the building, position information β of the intersection point between the reference scan layer L0 and the building, and the installation height h of the sensor 15. The building height information H_Cal can also be calculated using the Law of Cosines. An example of using the Law of Cosines is explained below.

[0126] The height information calculation unit 206 (a) identifies the scan layer Lx that generated a point cloud corresponding to the top edge of the building from which known height information H_Known was obtained, and obtains the intersection angle θx (elevation angle) between the reference scan layer L0 of the sensor 15 and the identified scan layer Lx. (b) obtains the distance D3 to the building in the direction along the scan layer Lx that generated the point cloud corresponding to the top edge of the building from the point cloud data. (c) obtains the distance D2 to the building in the direction along the reference scan layer L0 from the point cloud data.

[0127] In the case of Figure 4(A), scan layer L4 becomes scan layer Lx, which generates a point cloud corresponding to the upper edge of building BL. Here, let point A be the emission point of the laser beam from sensor 15, point B be the intersection point with building BL in the direction along the reference scan layer L0 (reflection point of the laser beam), point C be the intersection point with the upper edge of building BL in the direction along the scan layer L4 (reflection point of the laser beam), and let the intersection angle ∠BAC between the reference scan layer L0 and the scan layer L4 be the intersection angle θx. In this case, the distance from point A to point C becomes the distance D3 to the building in the direction along the scan layer L4. The distance from point A to point C becomes the distance D2 to the building in the direction along the reference scan layer L0. The intersection angle θx, distance D2, and distance D3 can be calculated from the measurement data (point cloud data) of sensor 15.

[0128] Here, in triangle ABC formed by connecting points A, B, and C, the distances D2 and D3 are known, and the intersection angle θx between sides AB and AC is also known. If we define the distance of the remaining side BC as distance Dx, then by the Law of Cosines, equation (10) below holds. (Dx) 2 =(D2) 2 +(D3) 2 -2(D2)(D3)cosθx ···(10)

[0129] Here, considering the installation height h of the sensor 15 in vehicle V, the building height information H_Cal can be expressed by the following equation (2). H_Cal=Dx+h (2)

[0130] In this way, the height information calculation unit 206 calculates the building height information H_Cal detected within the irradiation range IR from the intersection angle θx (elevation angle) between the scan layer Lx, which generates a point cloud corresponding to the top edge of the building from which known height information H_Known was obtained, and the reference scan layer L0 of the sensor 15, the distance D3 to the building in the direction of the scan layer Lx from which the point cloud corresponding to the top edge of the building was generated, and the distance D2 to the building in the direction of the reference scan layer L0.

[0131] By doing so, the height of the building can be calculated using the values ​​obtained from the measurement data of sensor 15.

[0132] The information processing device according to Modification 4 has the following configuration. (1) The information processing device 1 is A location information acquisition unit 200 acquires location information of the vehicle V (moving object), A distance information acquisition unit 203 acquires the distance from the vehicle V to the building (object), A height information acquisition unit 205 acquires known height information H_Known of a building based on the position information of the vehicle V and the distance D1 (first distance) to the building. A height information calculation unit 206 calculates the building height information H_Cal based on the distances D2 (second distance) and D3 (third distance) from the vehicle V to the building, The system includes a pitch angle calculation unit 207 that calculates the pitch angle Pitchθ of the vehicle V using known building height information H_Known and building height information H_Cal.

[0133] The pitch angle Pitchθ can be easily calculated using the position information of vehicle V and the distances D1, D2, and D3 from vehicle V to the building. Since the position information of vehicle V and the distances from vehicle V to the building can be obtained using existing equipment (sensor 15) mounted on vehicle V, the pitch angle Pitchθ of vehicle V can be calculated without requiring any additional equipment. The easily calculated pitch angle Pitchθ can be used in place of the pitch angle obtained by, for example, an inertial measurement unit (IMU12), thus providing redundancy in pitch angle acquisition.

[0134] (6) The height information calculation unit 206 is: The height information of the object H_Cal is calculated using the intersection angle θx between the scan layer Lx that transmitted the laser light (transmitted wave) reflected from the top edge of the building and the reference scan layer L0 (a layer along the horizontal plane based on the installation state of the sensor 15 on the vehicle when the vehicle V is located on the horizontal plane), the distance D2 to the building in the direction along the reference scan layer L0, the distance D3 to the top edge of the building, and the installation height h of the sensor 15 on the vehicle V.

[0135] By identifying the scan layer Lx that was illuminated by the laser light reflected from the top edge of the building, the distance D3 to the top edge of the building, the distance D2 to the building in the direction along the reference scan layer L0, and the intersection angle θx between scan layer Lx and reference scan layer L0 can be calculated from the point cloud data. The height information H_Cal, obtained by adding the installation height h of the sensor 15 on the vehicle V to the distance Dx from the intersection point of the building and the reference scan layer L0 to the top edge of the building, theoretically becomes the known height information H_Known of the object when the pitch angle of the vehicle V is 0°. Since the difference H between the calculated object height information H_Cal and the known height information H_Known differs depending on the pitch angle of vehicle V (the inclination of vehicle V relative to the horizontal line HL), the calculated object height information H_Cal can be used to calculate the pitch angle of vehicle V.

[0136] (VII) The IR irradiation range is divided into multiple scan layers Ln (where n is an integer) in the vertical direction, based on the installation state of the sensor 15 on the vehicle V. The distance D2 to the building in the direction along the reference scan layer L0 is calculated using the reference scan layer L0 along the horizontal plane, based on the installation state of the sensor 15 on the vehicle V when the vehicle V is located on the horizontal plane.

[0137] When the pitch angle of vehicle V is 0°, the distance to the building in the direction along the layer (reference scan layer L0) that is aligned with the horizontal plane, based on the installation state of sensor 15 on vehicle V when vehicle V is located on the horizontal plane, is minimized. By calculating the building height information H_Cal using the reference scan layer L0 as a reference, the calculated building height information H_Cal and the acquired known building height information H_Known can be compared with a consistent reference. This is expected to improve the accuracy of pitch angle calculation.

[0138] Although embodiments and modifications of the present invention have been described above, the present invention is not limited to these and can be modified as appropriate within the scope of the technical idea of ​​the invention. Furthermore, an information processing device with a configuration that arbitrarily combines the embodiments and modifications may also be used.

[0139] In the above embodiment, an example was given in which the illumination range IR is set on the front side of the vehicle V. For example, the sensor 15 may be installed on the roof of the vehicle V so that the illumination range is set to a 360° area around the vehicle V. The pitch angle can also be calculated using a building located behind the vehicle V.

[0140] In the above embodiment, as shown in Figure 2, an example was given in which six scan layers are set above the reference scan layer L0 and six scan layers are set below it. The configuration of the irradiation range is not limited to this configuration. The number of scan layers above the reference scan layer L0 and the number of scan layers below it may be different. Alternatively, the irradiation area may be one without a reference scan layer L0. In this case, the midpoint between scan layers L1 and L-1 can be calculated as the value of the reference scan layer L0.

[0141] Furthermore, in the above-described embodiment, an example was given in which the sensor 15 is a LiDAR sensor that irradiates a laser beam into the irradiation range IR and receives the laser beam (reflected wave) reflected by an object within the irradiation range IR. Then, the case in which this LiDAR sensor is a scan-type LiDAR that generates point cloud data for each scan layer was given as an example for explanation. Sensor 15 may also be a flash-type LiDAR that generates point cloud data all at once from the laser light reflected by surrounding objects by irradiating the surroundings with laser light, without using the concept of a scanning layer.

[0142] In the above-described embodiment, the example given was that the object is a building such as an apartment building or office building. For optimal detection, the target object should preferably be a building with upper edges aligned horizontally in the point cloud data. However, even a building with a single vertex, such as a radio tower, can be used as long as height information to the vertex can be obtained from map data or similar sources. Furthermore, the target object may be something other than a building. For example, signs installed on or around roads can also be used as targets, and similarly, it is preferable that they have upper edges aligned horizontally.

[0143] In the above-described embodiment, an example was given in which the information processing device 1 mounted on the vehicle V performs the following: detection of a building, calculation of the distance to the detected building, calculation of the building height information H_Cal, and calculation of the pitch angle Pitchθ by comparing it with known height information H_Known. At least some of the functions of this information processing device 1 may be handled by an external server ES, and the vehicle V may only determine whether the pitch angle input from the IMU 12 is appropriate. In the embodiments described above, the example shown was that the sensor is a LiDAR sensor, but the sensor used in the present invention is not limited to LiDAR sensors. Any measuring device that has the same functionality as a LiDAR sensor, that is, a device that transmits a wave toward an illumination range, receives the wave reflected by an object within the illumination range, and processes the received wave (reflected wave) to measure the shape of the object and calculate the distance to the object, is acceptable. Furthermore, in the embodiments described above, a scanning LiDAR was used as an example of a LiDAR sensor, but a flash LiDAR can also be used. [Explanation of Symbols]

[0144] 1: Information Processing Device 13: GPS Unit 14: Communication Unit 15: Sensor (LiDAR sensor) 20: Processing Unit 30: Memory Unit 31: Map Data 200: Location information acquisition section 201: Scanning Control Unit 202: Object detection unit 203: Distance information acquisition section 204: Building Coordinate Calculation Unit 205: Height information acquisition unit 206: Height Information Calculation Unit 207: Pitch angle calculation unit 208: Map Data Update Section BL, BL1, BLa~BLc: Buildings D1~D3, Dx: Distance ES: External Server H: Difference HL: Horizontal line H_Cal: Height information (calculated height information) H_Known: Known height information IR: Irradiation range L0: Reference scan layer L1~L6, L-1~L-3, Ln, Lx: Scan layer Pitchθ: Pitch angle R: Road surface TG:Measuring point V: Vehicle h: Installation height h: height θ: Crossing angle

Claims

1. A location information acquisition unit that acquires location information of a moving object, A measuring instrument that irradiates a transmission wave toward a set irradiation range and receives the reflected wave reflected by an object within the irradiation range, A distance information acquisition unit acquires the distance from the moving object to the target object based on the measurement data of the measuring instrument, A height information acquisition unit acquires known height information of an object by referring to map data based on the position information of the moving body and the coordinates of the object identified based on a first distance from the moving body to the object. A height information calculation unit that calculates the height information of the aforementioned object, The system includes a pitch angle calculation unit that calculates the pitch angle of the moving body using the known height information of the object and the height information of the object, The irradiation area is divided into multiple layers in the vertical direction. The height information calculation unit, The second distance from the moving body to the object in a layer along a horizontal line based on the installation state of the measuring instrument on the moving body, An information processing device that calculates height information of an object based on a third distance from the moving body to the upper edge of the object in the layer from which the transmitted wave reflected off the upper edge of the object was transmitted.

2. In claim 1, It includes an object detection unit that detects objects within the irradiation range based on point cloud data generated from the received reflected waves, The distance information acquisition unit, An information processing device that calculates the distance to a detected object from the point cloud data of the detected object.

3. In claim 2, The height information calculation unit, The layer from which the transmitted wave reflected off the upper edge of the object was transmitted, and the coordinates of the first intersection point with the object, The coordinates of the second intersection point between the layer along the horizontal line based on the installation state of the measuring instrument and the object, An information processing device that calculates height information of an object using the installation height of the measuring instrument on the mobile body.

4. In claim 3, An information processing device having at least two layers set above a layer that lies along a horizontal line based on the installation state of the aforementioned measuring instrument.

5. In claim 3, The height information calculation unit repeatedly performs the task of identifying the layer from which the transmitted wave reflected off the upper edge of the object was transmitted. An information processing device that, when a newly identified layer is a different layer from a previously identified layer, calculates the coordinates of the first intersection point using the newly identified layer.

6. In claim 2, The height information calculation unit, An information processing device that calculates height information of an object using the distance to the object in the direction along a horizontal line based on the installation state of the measuring instrument on the mobile body, the distance to the top of the object from the layer that transmitted the transmitted wave reflected from the top of the object, the intersection angle between the layer that transmitted the transmitted wave reflected from the top of the object and the layer along the horizontal line, and the installation height of the measuring instrument on the mobile body.

7. In any one of claims 1 to 6, The distance information acquisition unit acquires the distances to at least two objects if multiple objects exist within the irradiation range. The height information acquisition unit acquires known height information for each of the objects whose distance has been acquired, The height information calculation unit calculates the height information for each object whose distance has been obtained, The pitch angle calculation unit calculates the pitch angle for each object whose distance has been acquired, using the known height information of the object and the height information of the object, and sets the average value of the calculated pitch angles as the pitch angle of the moving body.

8. In any one of claims 1 to 6, The distance information acquisition unit acquires the distances to at least two objects if multiple objects exist within the irradiation range. The height information acquisition unit acquires known height information of objects from which known height information can be acquired among the objects whose distance has been measured, The height information calculation unit calculates the height information of the object from which the known height information has been obtained. The pitch angle calculation unit is an information processing device that calculates the pitch angle for each object from which known height information has been acquired, using the known height information of the object and the height information of the object.

Images