Trajectory data generation device, three-dimensional data generation device and computer program

Abstract

To provide a technique capable of generating highly accurate trajectory data without generating point group data.SOLUTION: A trajectory data generation device includes: a movement data acquisition unit for acquiring movement data to be specified during movement of a mobile body, the movement data related to the movement of the mobile body; a relative position data acquisition unit for acquiring relative position data to be specified during the movement of the mobile body, the relative position data representing a relative position of the specific point relative to the mobile body; and a trajectory data generation unit for generating trajectory data representing a movement trajectory of the mobile body using absolute position data representing an absolute position of the specific point relative to a reference point, the relative position data and the movement data.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The technology disclosed in this specification relates to a trajectory data generating device, a three-dimensional data generating device, and a computer program. [Background technology]

[0002] Patent Document 1 discloses a data processing device that generates three-dimensional point cloud data using a vehicle's movement trajectory. The data processing device generates composite point cloud data using the vehicle's movement trajectory and measurement point cloud data acquired by a laser scanner attached to the vehicle. Next, the data processing device adjusts the trajectory obtained from the composite point cloud data based on the coordinates of known feature points. The data processing device also recalculates the trajectory by fusing the adjusted trajectory with inertial positioning data acquired by an IMU (Inertial Measuring Unit) using a Kalman filter. Furthermore, the recalculated trajectory is smoothed and combined with the measurement point cloud data to generate three-dimensional point cloud data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-138786 Summary of the Invention [Problem to be solved by the invention]

[0004] In the above-described technology, synthetic point cloud data is generated to identify a trajectory for generating three-dimensional point cloud data. When generating precise three-dimensional point cloud data, the data size of the point cloud data increases. Therefore, a configuration that generates point cloud data to identify a trajectory imposes a heavy processing load. This specification provides a technology that can generate highly accurate trajectory data without generating point cloud data. [Means for solving the problem]

[0005] A first aspect of the present disclosure discloses a trajectory data generation device, which may include: a movement data acquisition unit that acquires movement data identified during movement of a moving object, the movement data being related to the movement of the moving object; a relative position data acquisition unit that acquires relative position data identified during movement of the moving object, the relative position data representing a relative position of a specific point with respect to the moving object; and a trajectory data generation unit that generates trajectory data representing a movement trajectory of the moving object using absolute position data representing an absolute position of the specific point with respect to a reference point, the relative position data, and the movement data.

[0006] According to the above configuration, trajectory data can be generated using not only movement data but also relative position data that indicates the relative position of a specific point with respect to a moving object and absolute position data that indicates the absolute position of a specific point with respect to a reference point. By using the absolute position data as well as data obtained during the movement of the moving object, highly accurate trajectory data can be generated without generating point cloud data.

[0007] In a second aspect of the present disclosure, in the first aspect, the reference point may be a point fixed with respect to the Earth, and the trajectory data generation unit may generate the trajectory data representing the movement trajectory with respect to the reference point.

[0008] According to the above configuration, it is possible to generate trajectory data representing a movement trajectory using reference points whose positions on the earth are specified.

[0009] In a third aspect of the present disclosure, in the first or second aspect, the trajectory data generation unit may perform an optimization process to optimize the movement trajectory based on the difference between a replacement position obtained by replacing the relative position represented by the relative position data with a position relative to the reference point and the absolute position represented by the absolute position data.

[0010] According to the above configuration, by replacing the relative position with the replacement position, it is possible to match the reference between the position of a specific point obtained during the movement of the moving object and the absolute position represented by the absolute position data. This makes it possible to clarify the error between the relative position represented by the relative position data and the absolute position represented by the absolute position data. Furthermore, according to the above configuration, it is possible to generate highly accurate trajectory data by optimizing the movement trajectory using the relative position represented by the relative position data and the absolute position represented by the absolute position data.

[0011] In a fourth aspect of the present disclosure, in any one of the first to third aspects above, the trajectory data generation unit may generate the trajectory data using a position and attitude obtained using the movement data, the position and attitude of the moving body during movement.

[0012] When a moving object moves, not only the position of the moving object but also the posture of the moving object changes. According to the above configuration, it is possible to generate trajectory data by further using both the position and posture of the moving object that change while the moving object is moving. This makes it possible to improve the accuracy of the trajectory data.

[0013] In a fifth aspect of the present disclosure, in any one of the first to fourth aspects, the movement data acquisition unit may acquire the movement data including position data representing a position of the moving object.

[0014] According to the above configuration, the position of the moving object that serves as a reference for the relative position can be used to generate the trajectory data, thereby improving the accuracy of the trajectory data.

[0015] In a sixth aspect of the present disclosure, in the fifth aspect above, the position data may include first position data obtained from a sensor device mounted on the mobile body and second position data obtained using a satellite positioning system.

[0016] The first position data may include a detection error of the sensor device. The second position data may include a measurement error of the satellite positioning system. By using both the first position data and the second position data, even if at least one of the first position data and the second position data includes an error, the effect of the error on the trajectory data can be reduced.

[0017] In a seventh aspect of the present disclosure, in any one of the first to sixth aspects, the movement data acquisition unit may acquire the movement data including spatial data representing a spatial shape around the moving body.

[0018] According to the above configuration, the spatial shape around the moving object can be used to generate trajectory data, thereby improving the accuracy of the trajectory data.

[0019] An eighth aspect of the present disclosure discloses a three-dimensional data generation device. The three-dimensional data generation device may include the trajectory data generation device of any one of the first to seventh aspects and a three-dimensional data generation unit that generates three-dimensional shape data that represents a spatial shape in three dimensions. The movement data acquisition unit may acquire the movement data including spatial data that represents a spatial shape around the moving object. The three-dimensional data generation unit may generate the three-dimensional shape data using the trajectory data and the spatial data generated by the trajectory data generation device.

[0020] According to the above configuration, three-dimensional shape data can be generated based on highly accurate trajectory data, so that highly accurate three-dimensional shape data can be generated.

[0021] A ninth aspect of the present disclosure discloses a computer program, which may cause a computer mounted on a trajectory data generation device to acquire movement data identified during movement of a moving object, the movement data being related to the movement of the moving object, acquire relative position data identified during movement of the moving object, the relative position data representing a relative position of a specific point with respect to the moving object, and generate trajectory data representing a movement trajectory of the moving object using absolute position data representing an absolute position of the specific point with respect to a reference point, the relative position data, and the movement data.

[0022] According to the above configuration, the computer can generate trajectory data using not only movement data but also relative position data that indicates the relative position of a specific point with respect to the moving object and absolute position data (for example, data determined using a surveying instrument) that indicates the position of the specific point with respect to a reference point. By using the absolute position data as well as data obtained during the movement of the moving object, highly accurate trajectory data can be generated without generating point cloud data. [Brief explanation of the drawings]

[0023] [Figure 1] FIG. 2 is a block diagram showing a three-dimensional data generation device 2. [Figure 2] FIG. 2 is a diagram schematically showing the relationship between a three-dimensional data generation device 2 and a measurement system 20. [Figure 3] 10 is a flowchart of a three-dimensional data generation process executed by the three-dimensional data generation device 2. [Figure 4] FIG. 2 is a diagram schematically showing a pose graph 70 constructed by the three-dimensional data generation device 2. [Figure 5] 2 is a diagram schematically showing a mathematical model 72 created by the three-dimensional data generation device 2. FIG. [Figure 6] 2 is a schematic diagram for explaining a process in which the three-dimensional data generation device 2 generates three-dimensional shape data 74. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0024] (Example) The three-dimensional data generation device 2 shown in FIG. 1 generates three-dimensional shape data that represents the spatial shape of a three-dimensional space. The three-dimensional shape data represents the spatial shape using a point cloud. In other words, the three-dimensional shape data includes data that represents the position coordinates of each point included in the point cloud. The three-dimensional data generation device 2 is used, for example, in the field of surveying, to generate three-dimensional shape data that represents terrain (shapes of buildings and roads) using a point cloud. The three-dimensional data generation device 2 includes a processor 4, a memory 6, a storage 8, a display unit 10, an operation unit 12, and a data input / output unit 14.

[0025] The processor 4 executes processing in accordance with a computer program stored in the storage 8. The memory 6 includes a RAM, and is used as a working memory when the processor 4 executes processing.

[0026] The storage 8 is configured by a hard disk drive or a solid state drive, and stores computer programs for causing the processor 4 to execute processes, and data used by the processor 4 when executing processes.

[0027] The display unit 10 is controlled by the processor 4 and configured to display various information to the user operating the three-dimensional data generation device 2. The display unit 10 includes, for example, a liquid crystal display.

[0028] The operation unit 12 is configured to input operation signals from a user to the three-dimensional data generation device 2 to the processor 4. The operation unit 12 includes, for example, a keyboard and a pointing device.

[0029] The data input / output unit 14 is used to input data from the outside and output data to the outside. The data input / output unit 14 includes, for example, a communication interface for communicating with an external device via a wired or wireless connection. The data input / output unit 14 includes, for example, a USB interface capable of reading and writing data from a USB memory. The data input / output unit 14 includes, for example, a media reader / writer capable of reading and writing data from a card-type recording medium.

[0030] 2, the storage 8 stores measurement data 50 measured by the measurement system 20. The measurement system 20 includes a moving object 22 (e.g., a vehicle or a person) and a measurement device 24 mounted on the moving object 22. The measurement system 20 is also called an MMS (Mobile Mapping System), and is known as a system that efficiently acquires the spatial shape around the moving object 22 with high accuracy while the moving object 22 is moving.

[0031] The measurement equipment 24 includes a laser scanner 26 , a satellite positioning unit 28 , and an inertial measurement unit 30 .

[0032] The laser scanner 26 irradiates the surroundings of the mobile object 22 with laser light and receives the laser light reflected by objects (for example, the ground or structures) around the mobile object 22. Based on the received laser light, the laser scanner 26 generates three-dimensional point cloud data that represents the spatial shape around the mobile object 22 while it is moving. The laser scanner 26 also generates scan data 52 that describes a combination of the generated point cloud data and time point data that represents the time point when the point cloud data was generated.

[0033] The satellite positioning unit 28 receives satellite signals transmitted from multiple artificial satellites and identifies the position of the mobile object 22 on Earth based on the received satellite signals. In this embodiment, the satellite positioning unit 28 identifies the position of the mobile object 22 in a geographic coordinate system (also referred to as the geographic coordinates of the mobile object 22) as the position of the mobile object 22 on Earth. The geographic coordinate system is a coordinate system that describes a position from an origin fixed relative to the Earth using three values: latitude, longitude, and altitude (i.e., polar coordinates). The satellite positioning unit 28 also generates satellite positioning data 54 that describes a combination of position data that indicates the identified position of the mobile object 22 on Earth and time data that indicates the time when the satellite signal for identifying the position was received.

[0034] The inertial measurement unit 30 includes an acceleration sensor and an angular velocity sensor (not shown) and detects the acceleration and angular velocity of the moving body 22. The inertial measurement unit 30 also generates inertial measurement data 56 that describes a combination of acceleration data and angular velocity data that represent the detected acceleration and angular velocity of the moving body 22, and time point data that represents the time point when the acceleration and angular velocity were detected. The acceleration of the moving body 22 corresponds to the change in the velocity of the moving body 22 over time. The angular velocity of the moving body 22 corresponds to the change in the attitude (roll angle, pitch angle, yaw angle) of the moving body 22 over time.

[0035] In the measurement system 20, the measurement device 24 generates measurement data 50 while the mobile object 22 is moving (i.e., generation of scan data 52 by the laser scanner 26, generation of satellite positioning data 54 by the satellite positioning unit 28, and generation of inertial measurement data 56 by the inertial measurement unit 30). Here, a series of events in which the measurement device 24 generates the measurement data 50 while the mobile object 22 is moving is also referred to as a "measurement event." Furthermore, the measurement data 50 generated by the measurement device 24 in a measurement event is input to the three-dimensional data generation device 2 via the data input / output unit 14 (see FIG. 1) and stored in the storage 8. For example, the measurement data 50 is input from the measurement device 24 to the three-dimensional data generation device 2 via communication between the three-dimensional data generation device 2 and the measurement device 24. Alternatively, the measurement data 50 is input from the measurement device 24 to the three-dimensional data generation device 2 via a USB memory or other recording medium.

[0036] The storage 8 also stores absolute position data 58 indicating the position of a specific point (e.g., the edge of a white line on a road, the base of a utility pole) on the earth. In this embodiment, the absolute position data 58 indicates the position of the specific point in a geographic coordinate system (the geographic coordinates of the specific point) as the position of the specific point on the earth. The position of the specific point indicated by the absolute position data 58 on the earth is identified using a means other than the measuring device 24 (e.g., a surveying instrument such as a total station). In this specification, a point whose position on the earth is determined by a means other than the measuring device 24 is also referred to as a GCP (short for Ground Control Point). Information regarding the position of the GCP may be information made public by an administrative agency. The absolute position data 58 is input to the three-dimensional data generation device 2 through the data input / output unit 14 and stored in the storage 8. For example, the absolute position data 58 is input to the three-dimensional data generation device 2 via an input operation by a user performed on the operation unit 12.

[0037] (3D data generation processing) When the user causes the three-dimensional data generation device 2 to generate three-dimensional data, the user executes a predetermined generation operation on the operation unit 12. When the generation operation is executed by the user, the processor 4 executes the three-dimensional data generation process shown in FIG.

[0038] In S2, the processor 4 acquires the measurement data 50 (see FIG. 2). Specifically, the processor 4 reads the measurement data 50 stored in the storage 8 from the storage 8. Unless otherwise specified, the measurement data 50 described in the following S4 to S10 is the measurement data 50 acquired by the processor 4 in S2.

[0039] Next, in S4, the processor 4 constructs a pose graph 70 as shown in FIG. 4 by scan matching using the scan data 52 and the inertial measurement data 56 from the measurement data 50. The pose graph 70 is data describing the position and orientation of the moving body 22 at each point in the measurement event in a coordinate system (hereinafter also referred to as a graph coordinate system) in which an arbitrary position (e.g., the position of the moving body 22 at the start of the measurement event) is the origin for position and an arbitrary orientation (e.g., the orientation of the moving body 22 at the start of the measurement event) is the origin for orientation. In the graph coordinate system, the orientation coordinate represents the amount of change from the orientation origin. Note that in the example of FIG. 4, the position and orientation of the moving body 22 are schematically shown using pentagons. For ease of explanation, arrows are used to indicate the order of movement of the moving body 22.

[0040] In S6, the processor 4 creates a mathematical model 72 for optimizing the pose graph 70, as shown in FIG. 5. Specifically, the processor 4 sets a coordinate system (hereinafter also referred to as a model coordinate system) in which a specific point on the Earth is set as the origin for position and an arbitrary attitude (for example, the attitude of the moving body 22 at the start of the measurement event) is set as the origin for attitude. Of the model coordinate systems, the coordinate system for position corresponds to a geographic coordinate system (i.e., a coordinate system in which each position is described by three values: latitude, longitude, and altitude). For example, the coordinate system for position may coincide with the geographic coordinate system. Alternatively, the coordinate system for position may differ from the geographic coordinate system. In this case, the coordinate system for position may be a coordinate system that can be replaced with the geographic coordinate system. In the model coordinate system, the processor 4 describes the position (latitude, longitude, altitude) of the moving body 22 at each time point of the measurement event as variables Va, Vb, and Vc, and describes the attitude (roll angle, pitch angle, yaw angle) of the moving body 22 at each time point of the measurement event as variables Vψ, Vθ, and Vφ. The processor 4 sets residuals r1, r2, and r3 as functions that describe the difference between a predetermined observed value (details will be described later) and a predicted value corresponding to the observed value. The residuals r1, r2, and r3 are set for different observed values, respectively.

[0041] (Regarding residual r1) The processor 4 sets the residual r1 as the difference between the positions SPa, SPb, SPc (observed values) of the moving body 22 on the earth at each time point identified from the satellite positioning data 54 (see FIG. 2) and the positions Va, Vb, Vc (predicted values) of the moving body 22 at each time point in the mathematical model 72. For example, as shown in FIG. 5, the residual r1(i) at time point i is written as the following equation (1).

[0042]

number

[0043] The lower the accuracy of the positions Va, Vb, and Vc of the moving object 22, the larger the absolute value of the residual r1. Conversely, the higher the accuracy of the positions Va, Vb, and Vc of the moving object 22, the smaller the absolute value of the residual r1.

[0044] (Regarding residual r2) Scan matching using the scan data 52 and inertial measurement data 56 from the measurement data 50 may allow the positional displacements Δa, Δb, Δc and the attitude displacements Δψ, Δθ, Δφ of the moving body 22 between two points in time to be determined. In this case, the processor 4 determines displacements ΔVa, ΔVb, ΔVc, ΔVψ, ΔVθ, ΔVφ corresponding to the displacements Δa, Δb, Δc, Δψ, Δθ, Δφ in the mathematical model 72. The processor 4 sets the difference between the position and attitude displacements Δa, Δb, Δc, Δψ, Δθ, Δφ (observed values) of the moving body 22 determined by scan matching and the position and attitude displacements ΔVa, ΔVb, ΔVc, ΔVψ, ΔVθ, ΔVφ (predicted values) of the moving body 22 on the mathematical model 72 as the residual r2. For example, as shown in FIG. 5, when the displacement amounts Δa(i,j), Δb(i,j), Δc(i,j), Δψ(i,j), Δθ(i,j), and Δφ(i,j) between time point i and time point j are identified, the displacement amounts ΔVa(i,j), ΔVb(i,j), ΔVc(i,j), ΔVψ(i,j), ΔVθ(i,j), and ΔVφ(i,j) identified from the mathematical model 72 are identified as the differences between the position and orientation Va(i), Vb(i), Vc(i), Vψ(i), Vθ(i), and Vφ(i) of the moving body 22 at time point j and the position and orientation Va(j), Vb(j), Vc(j), Vψ(j), Vθ(j), and Vφ(j) of the moving body 22 at time point j. That is, ΔVa(i,j) = Va(j) - Va(i), ΔVb(i,j) = Vb(j) - Vb(i), ΔVc(i,j) = Vc(j) - Vc(i), ΔVψ(i,j) = Vψ(j) - Vψ(i), ΔVθ(i,j) = Vθ(j) - Vθ(i), ΔVφ(i,j) = Vφ(j) - Vφ(i). The residual r2(i) set in this case is written as in the following equation (2).

[0045]

number

[0046] The lower the accuracy of the positions Va, Vb, Vc and orientations Vψ, Vθ, Vφ of the moving body 22, the larger the absolute value of the residual r2. Conversely, the higher the accuracy of the positions Va, Vb, Vc and orientations Vψ, Vθ, Vφ of the moving body 22, the smaller the absolute value of the residual r2.

[0047] (Regarding residual r3) In this embodiment, feature points having characteristic shapes (e.g., the edges of white lines on roads or the bases of utility poles) are selected as specific points (i.e., GCPs) for storing terrestrial positions (absolute position data 58) in storage 8. Information regarding the shapes of the feature points is stored in storage 8. For this reason, processor 4 identifies the shapes representing the feature points from the spatial shapes represented by scan data 52 (see FIG. 2 ) and specifies the relative positions da, db, and dc of the feature points with respect to the moving object 22 at a certain point in time. Processor 4 can replace the relative positions da, db, and dc of the feature points with terrestrial positions EPa, EPb, and EPc based on the mathematical model 72. Processor 4 sets the difference between the terrestrial positions APa, APb, and APc of the feature points represented by absolute position data 58 (observed values) and the positions EPa, EPb, and EPc of the feature points specified from scan data 52 (predicted values) as residual r3. For example, as shown in Figure 5, when positions da(i), db(i), dc(i) at time point i are identified, the corresponding positions EPa(i), EPb(i), EPc(i) are identified as the sum of the positions Va(i), Vb(i), Vc(i) of the moving object 22 on the mathematical model 72 at time point i and the positions da(i), db(i), dc(i). That is, EPa(i) = Va(i) + da(i), EPb(i) = Vb(i) + db(i), EPc(i) = Vc(i) + dc(i). The residual r3(i) set in this case is written as in the following equation (3).

[0048]

number

[0049] The lower the accuracy of the positions Va, Vb, and Vc of the moving object 22, the larger the absolute value of the residual r3. Conversely, the higher the accuracy of the positions Va, Vb, and Vc of the moving object 22, the smaller the absolute value of the residual r3.

[0050] (On the objective function in mathematical model 72) The processor 4 sets an objective function to be optimized based on the residuals r1, r2, and r3. The following formula (4) describes an objective function F based on the least squares method as an example of an objective function in the mathematical model 72. Specifically, the objective function F is described as the sum of the squares of the weighted residuals r1·w1, r2·w2, and r3·w3 obtained by multiplying the residuals r1, r2, and r3 by weighting coefficients w1, w2, and w3. The weighting coefficients w1, w2, and w3 are set according to the reliability of each of the residuals r1, r2, and r3.

[0051]

number

[0052] The above-mentioned objective function F is a function that indicates, depending on its numerical value, the accuracy of the positions Va, Vb, Vc and orientations Vψ, Vθ, Vφ of the moving body 22. Therefore, in this embodiment, the numerical values ​​of the positions Va, Vb, Vc and orientations Vψ, Vθ, Vφ of the moving body 22 that minimize the above-mentioned objective function F indicate the position and orientation of the moving body 22 with the highest accuracy.

[0053] When the creation of the mathematical model 72 is completed, S6 shown in FIG. 3 ends and the process proceeds to S8.

[0054] In S8, processor 4 optimizes the pose graph 70 (see FIG. 4) based on the mathematical model 72 (see FIG. 5) created in S6. Specifically, processor 4 derives, as an optimal solution to the mathematical model 72, numerical values ​​for the positions Va, Vb, and Vc and the orientations Vψ, Vθ, and Vφ of the moving body 22 that minimize the objective function F of equation (4). Then, processor 4 replaces the numerical values ​​representing the position and orientation of the moving body 22 described in the pose graph 70 with the numerical values ​​derived as the optimal solution to the mathematical model 72. Note that the pose graph 70 optimized in S8 represents the movement trajectory of the moving body 22 on the Earth (in this embodiment, the movement trajectory in a geographic coordinate system). In this regard, the pose graph 70 constructed in S4 represents the movement trajectory of the moving body 22 relative to an arbitrary origin (e.g., the position of the moving body 22 at the start of the measurement event). Therefore, in this embodiment, the pose graph 70 constructed in S4 is sometimes called the "local pose graph," and the pose graph 70 optimized in S8 is sometimes called the "global pose graph," to distinguish between the two.

[0055] In S10, the processor 4 applies the scan data 52 (see FIG. 2) to the global pose graph to generate three-dimensional shape data 74. Specifically, the processor 4 arranges the point cloud data at each time point represented by the scan data 52 as viewed from the position and orientation of the moving object 22 at each time point represented by the global pose graph. This generates three-dimensional shape data 74, which represents the spatial shape of a three-dimensional space using a point cloud, as shown in FIG. 6. The three-dimensional shape data 74 in this embodiment is global three-dimensional shape data 74 in which each point included in the point cloud is associated with a position on the Earth (geographic coordinates). Note that in the example of FIG. 6, the three-dimensional shape is simplified to a two-dimensional shape.

[0056] As described above, according to the process shown in FIG. 3 , the pose graph 70 is constructed by scan matching using the scan data 52 and the inertial measurement data 56. This makes it possible to obtain a pose graph 70 that is consistent with both the scan data 52 and the inertial measurement data 56. Furthermore, according to the process shown in FIG. 3 , the pose graph 70 is optimized based on the discrepancy (residual r3) between the relative positions da, db, and dc of the feature points identified from the scan data 52 and the positions APa, APb, and APc of the feature points on Earth represented by the absolute position data 58. This allows for a highly accurate pose graph 70 to be obtained, thereby enabling the generation of highly accurate three-dimensional shape data. Furthermore, according to the process shown in FIG. 3 , not only the position of the moving body 22 but also the orientation of the moving body 22 is optimized in the pose graph 70. This makes it easier for the spatial shapes represented by the point cloud data at each time point to be consistent when the point cloud data at each time point of the measurement event is arranged as viewed from the position and orientation of the moving body 22 at each time point represented by the pose graph 70. As a result, it is possible to generate more accurate three-dimensional shape data.

[0057] (Correspondence) In the embodiment, the three-dimensional data generation device 2 is an example of a "trajectory data generation device" and a "three-dimensional data generation device." The feature point (GCP) is an example of a "specific point." The positions da, db, and dc are an example of a "relative position of a specific point with respect to a moving body." The positions EPa, EPb, and EPc are an example of a "replacement position." The positions APa, APb, and APc are an example of an "absolute position." Any point on the earth is an example of a "reference point." The absolute position data 58 is an example of "absolute position data." The processor 4 is an example of a "movement data acquisition unit," a "relative position data acquisition unit," a "trajectory data generation unit," and a "three-dimensional data generation unit." The measurement data 50 is an example of "movement data." The scan data 52 is an example of "spatial data" and "relative position data." The inertial measurement data 56 is an example of "position data" and "first position data." The satellite positioning data 54 is an example of "position data" and "second position data." The global pose graph is an example of “trajectory data.” Steps S4 to S8 of the three-dimensional data generation process shown in Figure 3 are an example of “optimization processing.”

[0058] (Variation) The measurement device 24 shown in FIG. 2 may further include measurement equipment other than the laser scanner 26, the satellite positioning unit 28, and the inertial measurement unit 30. For example, the measurement device 24 may further include a camera capable of generating image data. The image data may be data describing a combination of a visible light image and time point data indicating the time point at which the visible light image was generated. In this case, the processor 4 of the three-dimensional data generation device 2 may be configured to construct a pose graph 70 (see FIG. 4) from the temporal change in the visible light image represented by the image data in S4 of the process shown in FIG. 3. Furthermore, the processor 4 may generate three-dimensional shape data by applying the image data generated by the camera to the pose graph 70 in S10 of the process shown in FIG. 3. Specifically, the processor 4 may generate three-dimensional shape data by arranging the visible light images at each time point of the measurement event included in the image data as viewed from the position and orientation of the moving object 22 at each time point represented by the pose graph 70.

[0059] The absolute position data 58 shown in FIG. 2 may indicate the position of the characteristic point (GCP) relative to a reference point specified by the user (for example, the base of a specified utility pole) instead of indicating the geographic coordinates of the characteristic point (GCP).

[0060] The absolute position data 58 shown in FIG.

[0061] In S4 of the process shown in FIG. 3, processor 4 of three-dimensional data generation device 2 may construct pose graph 70 (see FIG. 4) without performing scan matching. For example, processor 4 may construct pose graph 70 using only scan data 52, without using inertial measurement data 56. In this case, processor 4 may construct pose graph 70 from temporal changes in point cloud data represented by scan data 52. Alternatively, processor 4 may construct pose graph 70 using only inertial measurement data 56, without using scan data 52. In this case, processor 4 may construct pose graph 70 from temporal changes in acceleration and angular velocity of moving object 22 represented by inertial measurement data 56.

[0062] In the series of processes shown in Fig. 3, the processor 4 of the three-dimensional data generation device 2 does not need to use the satellite positioning data 54. Specifically, in S6 of the processes shown in Fig. 3, the processor 4 does not need to set the residual r1. Accordingly, a term related to the residual r1 may be deleted from the objective function to be optimized (for example, the objective function F shown in equation (4)).

[0063] 3, the processor 4 may not set the residual r2. Accordingly, the term related to the residual r2 may be deleted from the objective function to be optimized (for example, the objective function F shown in equation (4)).

[0064] The technical elements described in this specification or drawings exhibit technical utility either alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Furthermore, the technologies illustrated in this specification or drawings can achieve multiple objectives simultaneously, and achieving one of those objectives is itself technically useful. [Explanation of symbols]

[0065] 2: Three-dimensional data generation device, 4: Processor, 6: Memory, 8: Storage, 10: Display unit, 12: Operation unit, 14: Data input / output unit, 20: Measurement system, 22: Moving object, 24: Measurement device, 26: Laser scanner, 28: Satellite positioning unit, 30: Inertial measurement unit, 50: Measurement data, 52: Scan data, 54: Satellite positioning data, 56: Inertial measurement data, 58: Absolute position data, 70: Pose graph, 72: Mathematical model, 74: Three-dimensional shape data

Claims

1. a movement data acquisition unit that acquires movement data identified during movement of a moving object, the movement data being related to the movement of the moving object; a relative position data acquisition unit that acquires relative position data identified during the movement of the moving body, the relative position data representing the relative position of a specific point with respect to the moving body; A trajectory data generation device comprising: a trajectory data generation unit that generates trajectory data that represents a movement trajectory of the moving body using absolute position data that represents the absolute position of the specific point relative to a reference point, the relative position data, and the movement data.

2. the reference point is a point fixed relative to the Earth; The trajectory data generating device according to claim 1 , wherein the trajectory data generating unit generates the trajectory data representing the movement trajectory relative to the reference point.

3. 2. The trajectory data generation device according to claim 1, wherein the trajectory data generation unit executes an optimization process to optimize the movement trajectory based on a difference between a replaced position obtained by replacing the relative position represented by the relative position data with a position relative to the reference point and the absolute position represented by the absolute position data.

4. The trajectory data generating device according to claim 1 , wherein the trajectory data generating unit generates the trajectory data using the position and the orientation of the moving object during movement, the position and the orientation being obtained using the movement data.

5. The trajectory data generating device according to claim 1 , wherein the movement data acquiring unit acquires the movement data including position data representing a position of the moving object.

6. The trajectory data generating device according to claim 5 , wherein the position data includes first position data obtained from a sensor device mounted on the moving body, and second position data obtained using a satellite positioning system.

7. The trajectory data generating device according to claim 1 , wherein the movement data acquiring unit acquires the movement data including spatial data representing a spatial shape around the moving object.

8. The trajectory data generating device according to claim 1 ; a three-dimensional data generation unit that generates three-dimensional shape data that represents a spatial shape in three dimensions; the movement data acquisition unit acquires the movement data including spatial data representing a spatial shape around the moving object; The three-dimensional data generating device, wherein the three-dimensional data generating unit generates the three-dimensional shape data using the trajectory data generated by the trajectory data generating device and the spatial data.

9. A computer installed in the trajectory data generating device acquiring movement data identified during movement of a moving object, the movement data being related to the movement of the moving object; acquiring relative position data identified during the movement of the moving body, the relative position data representing the relative position of a specific point with respect to the moving body; A computer program that generates trajectory data representing a movement trajectory of the moving object using absolute position data representing an absolute position of the specific point relative to a reference point, the relative position data, and the movement data.

Images