KR101956447B1 - Method and apparatus for position estimation of unmanned vehicle based on graph structure - Google Patents

Abstract

The present invention relates to a non-human body position estimation apparatus based on a graph structure for estimating a position of a human body using multiple sensors and a graph structure, and a method thereof, and can estimate the position of a human body even under a large structure.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for estimating a position of a human body based on a graph structure,

The present invention relates to a non-human body position estimation apparatus based on a graph structure and a method thereof, and more particularly, to a technique for estimating a position of a human body using multiple sensors and a graph structure.

The most fundamental prerequisite for improving the quality of life of the people and realizing a welfare society is building a safe society, which starts with securing a safe infrastructure (safety infrastructure).

However, due to the occurrence of subsequent structural accidents, the people have become negative viewpoints and distrust, and social anxiety is increasing due to the aging of infrastructure.

In order to prevent accidents of large structures such as bridges, safety diagnosis is carried out through specialized companies. However, there is a lack of investment and management supervision by the management, lack of expertise of the related field engineers, poor working environment, The difficulty of the appearance inspection due to the cost inconvenience, and the difficulty in securing the reliability by the subjective judgment intervention of the inspector.

In order to solve these problems, researches on a structure health monitoring system that monitors and installs various kinds of measuring instruments on a structure are actively conducted. However, the existing structural health monitoring system is mainly performed by visual inspection and non-destructive inspection, and there are insufficient application techniques for a wide area and there are limitations in performing inspection according to the location and condition of the inspection target facility.

In recent years, a method of using a human body as a means to overcome this limitation has been studied.

Until now, domestic UAV technology has been developed for military purposes to perform missions such as reconnaissance, surveillance and bombing.

The above UAV technology is expected to be applicable to various civilian fields. The civilian UAV market is expected to be developed mainly in the public sector, and it is expected to expand into various fields such as video, agriculture, and environmental monitoring depending on the utilization.

The structure diagnosis using the unmanned human body based on the domestic unmanned aerial vehicle technology can effectively perform the inspection even in areas where the inspector can not access directly. However, such areas as the inspection and monitoring of structures using the human body are in the early stage of the technology, and it is necessary to develop the human body and related technologies suitable for structural inspection and monitoring.

At present, the level of non-human body technology is excellent, but its competitiveness deteriorates because it relies on imports of core parts.

In addition, the existing structure monitoring technology using the human body has a limitation that it is impossible to estimate the location due to the shadow area of GPS when the unmanned person carries out the diagnosis task under the large structure.

Therefore, this specification proposes a technique for estimating the position even under a large structure.

Korean Registered Patent No. 101472392 (Registered on December 31, 2014), "UAV System with Precise Position Tracking Function and Control Method Thereof" Korean Patent Publication No. 1020170004556 (published on Jan. 11, 2017), "Relocation method and apparatus of mobile robot in indoor environment"

It is an object of embodiments of the present invention to provide a non-human body position estimation apparatus and a method thereof based on a graph structure capable of estimating the position of a non-human body even under a large structure.

Also, it is an object of embodiments of the present invention to provide a non-human body position estimation apparatus based on a graph structure capable of more accurately estimating a position of a non-human body in a shadow area of a GPS using multiple sensors.

Also, an object of embodiments of the present invention is to provide a non-human body position estimation apparatus based on a graph structure capable of improving autonomous driving performance of a human body by estimating the position of a human body, and a method thereof.

The non-human body position estimation method based on a graph structure according to an embodiment of the present invention includes the steps of measuring data according to a moving path of a human body using multiple sensors, Constructing a graph structure from a node, a constraint between each node, and estimating a position of a human body based on the graph structure.

The step of measuring the data may include measuring velocity data and initial position data generated based on sensor data from an IMU (Inertial Measurement Unit) having an acceleration sensor and a gyro sensor mounted on the human body, And measuring position data from a Global Positioning System (GPS) mounted on the human body.

The step of measuring the data may include measuring a visual odometry of a human body from a camera and measuring a human body by using a 3D rider (LiDAR; Light Detection And Ranging) and a 3D ICP (Iterative Closest Point Fitting) Of the moving image data.

Wherein the step of constructing the graph structure comprises: the node representing each position along the movement path of the human body based on the measured data from the multiple sensors including the IMU, the GPS, the camera and the 3D rider; Can be obtained.

The step of constructing the graph structure may construct the graph structure through SLAM (Simultaneous Localization and Mapping) based on the node according to the moving path of the human body and the constraint condition.

The step of estimating the position of the non-human body may estimate the position of the non-human body using graph optimization based on the constructed graph structure.

The non-human body position estimation apparatus based on a graph structure according to an embodiment of the present invention includes a measurement unit for measuring data according to a moving path of a human body using multiple sensors, A graph structure constructing unit for constructing a graph structure from nodes of the nodes and constraints between the nodes, and a position estimator for estimating the position of the human body based on the graph structure.

The measurement unit measures velocity data and initial position data (initial position) generated based on sensor data from an IMU (Inertial Measurement Unit) equipped with an acceleration sensor and a gyro sensor mounted on the human body, It is possible to measure position data from a mounted Global Positioning System (GPS).

The measurement unit measures the visual odometry of a human body from a camera and measures movement data of a human body through a 3D ICP (Iterative Closest Point Fitting) algorithm based on a 3D rider (LiDAR; Light Detection And Ranging) Can be measured.

Wherein the graph structure construction unit comprises: a node representing each position according to a moving path of the human body based on the measured data from the multiple sensors including an IMU, a GPS, a camera, and a 3D rider; Constraints can be obtained.

The graph structure constructing unit may construct the graph structure through SLAM (Simultaneous Localization and Mapping) based on the node according to the moving path of the human body and the constraint condition.

The position estimating unit may estimate the position of the non-human body using graph optimization based on the constructed graph structure.

According to the embodiment of the present invention, the position of the non-human body can be estimated even under the large structure.

Also, according to the embodiment of the present invention, it is possible to more accurately estimate the position of a non-human body in a shadow area of a GPS using multiple sensors.

In addition, according to the embodiment of the present invention, it is possible to improve autonomous driving performance of a human body by estimating the position of a human body.

1 shows an example of a moving path of a human body for diagnosing a large structure according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a non-human body position estimation method based on a graph structure according to an embodiment of the present invention.
FIG. 3 illustrates an example of a position estimation algorithm using a graph structure according to an embodiment of the present invention.
4A to 4D show an example of a graph structure according to an embodiment of the present invention.
FIG. 5 shows a result of estimating the position of a human body using a graph-based non-human body position estimation method according to an embodiment of the present invention.
FIG. 6 is a block diagram of a non-human body position estimation apparatus based on a graph structure according to an embodiment of the present invention.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to or limited by the embodiments. In addition, the same reference numerals shown in the drawings denote the same members.

Also, terminologies used herein are terms used to properly represent preferred embodiments of the present invention, which may vary depending on the viewer, the intention of the operator, or the custom in the field to which the present invention belongs. Therefore, the definitions of these terms should be based on the contents throughout this specification.

1 shows an example of a moving path of a human body for diagnosing a large structure according to an embodiment of the present invention.

Referring to FIG. 1, the human body 10 moves along the position of a sensor attached to each of a large structure or a concrete structure, and can monitor and diagnose the structure.

There is a limit in that the conventional technique can not estimate the position of the non-human body in the shadow area of the GPS when the unmanned person carries out the diagnosis task under the large structure or the concrete structure.

Accordingly, the non-human body position estimation apparatus and method based on a graph structure according to an embodiment of the present invention can more accurately estimate movement of a human body in a shadow area of a GPS, which is a large structure or a concrete structure.

The large structure or concrete structure may refer to a lower end of a pylon, a high-rise building, a building, a building or a dam, but the structure, shape and position of the large structure or concrete structure are not limited thereto, Is not detected or is in a vulnerable area.

Hereinafter, a method for estimating the position of a non-human body in a GPS shaded area below the large structure or the concrete structure described above will be described.

FIG. 2 is a flowchart illustrating a non-human body position estimation method based on a graph structure according to an embodiment of the present invention.

Referring to FIG. 2, in step 210, data according to a moving path of a human body is measured using multiple sensors.

For example, the multiple sensors may include at least one of an IMU, a GPS, a camera, and a 3D rider (LiDAR). In step 210, data of movement paths . ≪ / RTI >

More specifically, step 210 measures velocity data and initial position data (initial position) generated based on sensor data from an IMU (Inertial Measurement Unit) equipped with an acceleration sensor and a gyro sensor mounted on a human body, It may be a step of measuring position data from a GPS (Global Positioning System) mounted on the human body.

In step 210, the visual odometry data of a human body is measured from a camera, and based on a 3D rider (Light Detection And Ranging), a human body is moved through a 3D ICP (Iterative Closest Point Fitting) And measuring the data.

Here, the IMU (Inertial Measurement Unit) uses a gyro sensor (gyroscope) as inertial measurement means. For example, the IMU calculates attitude-related sensor data such as forward rotation angle (roll axis direction), rightward direction (pitch axis direction) and gravity direction (yaw axis direction) Acceleration sensor data such as an inhuman body speed middle value can be generated using an acceleration sensor.

However, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention is not limited to this, and velocity data, initial position data, and attitude data can be acquired using various inertial measurement means.

In addition, GPS (Global Position System) can acquire position data (e.g., latitude, longitude and ellipsoid height as geodesic coordinates of a WGS84 coordinate system) by communicating with a satellite.

In addition, the camera can acquire image data and visual odometry data of a human body by capturing an image using a CMOS or a CCD (Charge Coupled Device) image sensor.

According to the embodiment, the RGB camera of Point Gray Flea3 can be used as the method for estimating the non-human body position based on the graph structure according to the embodiment of the present invention, but the type of the camera is not limited and various image photographing cameras can be used.

Also, 3D LiDAR (Light Detection And Ranging) can be a radar system that irradiates laser pulses to a reflector (human body) and measures the position coordinates of the reflector by measuring the return time from the reflector. As is well known, 3D LiDAR stores position coordinates measured and calculated from reflection intensity as measurement data, and measurement data can be utilized as basic information for imaging 3D graph structure. The 3D LiDAR is widely used in the related art, and its structure and operation principle will not be described.

According to the embodiment, 3DLDAR of Velodyne VLP16 may be used for the non-human body position estimation method based on the graph structure according to the embodiment of the present invention, but the type of 3D LiDAR is not limited thereto.

According to embodiments, the types of the above-described multiple sensors are not limited, and the multiple sensors of IMU, GPS, Camera, and 3D LiDAR can be located inside or outside each of a human body and a large structure, The number is not limited.

A graph structure is constructed from the nodes in the moving path of the human body and the constraints between the nodes based on the measured data in step 220.

For example, step 220 may include obtaining a node representing each location along the moving path of the human body based on the measured data from multiple sensors including IMU, GPS, camera and 3D rider, can do.

Accordingly, the step 220 may be a step of constructing a graph structure through a node according to the moving path of the human body and SLAM (Simultaneous Localization and Mapping) based on the constraint.

The SLAM refers to a method of detecting the non-human and non-human surroundings information from multiple sensors and processing the measured data to create a map corresponding to the movement and position of the human body and estimating the absolute position of the human body .

The non-human body position estimation method based on the graph structure according to the embodiment of the present invention is a means for recognizing the real-time position and the final position of a human body in step 220, wherein the SLAM process based on the measurement data acquired by the multiple sensors It is possible to construct a graph structure according to a movement path or movement trajectory of a human body.

According to an embodiment, step 220 may include updating the real-time location estimate of the inhuman body from a constraint between a node and a node indicating the location of the human body at a specified time based on data measured from multiple sensors And the distance measurement value of the human body can be obtained.

In the non-human body position estimation method based on the graph structure according to the embodiment of the present invention, when one sensor fails in the course of performing SLAM by using multiple sensors, the SLAM process is performed using the remaining sensors This can improve performance for SLAM.

In step 230, the position of the human body is estimated based on the graph structure.

Step 230 may be a step of estimating the position of a human body using graph optimization based on the constructed graph structure.

For example, in step 230, a non-human body position estimation method based on a graph structure according to an embodiment of the present invention includes a step of estimating a final position according to a moving path and time of a human body in a graph structure formed by nodes and constraints, The position of the human body can be estimated using the graph optimization technique.

However, the technique of acquiring the real-time position, the moving path, and the final position of the human body in the constructed graph structure can be applied variously in addition to the above-described graph optimization technique, and thus is not limited thereto.

FIG. 3 illustrates an example of a position estimation algorithm using a graph structure according to an embodiment of the present invention.

Referring to FIG. 3, a graph-based non-human body position estimation method according to an exemplary embodiment of the present invention is a method for estimating a human body position based on a graph using a plurality of sensors of an IMU 310, a GPS 320, a camera 330, and a 3D LiDAR 340 Constructs a structure 360, and provides an algorithm for estimating the position of a human body based on the constructed graph structure 360.

The IMU (Inertial Measurement Unit) 310 uses a gyro sensor (gyroscope) as inertial measurement means. For example, the IMU 310 can be used as a rotational angle of a human body such as a posture relating to the forward direction (Roll axis direction), the rightward direction of the advancing direction (the pitch axis direction), the gravity direction Acceleration-related sensor data such as sensor data and no-body speed middle value can be generated.

Accordingly, the IMU 310 can measure the speed data of the human body and the initial position data (initial position) 311 based on the sensor data from the acceleration sensor and the gyro sensor mounted on the human body.

GPS (Global Position System) 320 may communicate with the satellite to measure position data (e.g., latitude, longitude, and ellipsoid height as geodesic coordinates of a WGS84 coordinate system).

The camera 330 can measure a visual odometry 331 of a human body by capturing an image using a CMOS or a CCD (Charge Coupled Device) image sensor. The omometry data 331 ) Includes the x-axis coordinate of the human body, the y-axis coordinate, and the angle of the heading direction, and may include the estimated displacement.

The 3D LiDAR (Light Detection And Ranging) 340 irradiates a laser pulse to a human body, measures the return time from the human body, and calculates movement data of the human body through the 3D ICP (Iterative Closest Point Fitting, 341) Can be measured.

The 3D ICP 341 algorithm can be used to match the 2D or 3D point data measured with the 3D LiDAR 340 to the graph structure 360.

Thereafter, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention is based on data measured from multiple sensors of the IMU 310, the GPS 320, the camera 330 and the 3D LiDAR 340 A graph structure 360 can be constructed from nodes and constraints 350 in the path of the human body.

Depending on the embodiment, the node in the moving path of the human body may be the initial position 311 measured from the IMU 310.

For example, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention includes the IMU 310, the GPS 320, and the GPS 320 based on a node, which is an initial position 311 according to a moving path of a human body. Constraints 350 indicating the connection of the nodes can be constructed using the geometry data 331 and the 3D ICP 341 and the graph structure 360 can be constructed from the nodes and constraints 350. [ Can be constructed.

Hereinafter, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention can estimate (380) the position of a human body by applying a graph optimization technique 370 based on the constructed graph structure 360 .

4A to 4D show an example of a graph structure according to an embodiment of the present invention.

More specifically, FIG. 4A shows an example of a graph structure including odometry data by a camera, FIG. 4B shows an example of a graph structure including data by IMU and GPS, and FIG. And an example of a graph structure including data of 3D ICP by 3D LiDAR.

FIG. 4D shows an example of a graph structure incorporating FIGS. 4A to 4C.

The non-human body position estimation method based on the graph structure according to the embodiment of the present invention can form nodes (positions) according to the moving path of the human body. For example, in the basis of the initial time x _i form a free node 410 of the human body in position, and after a predetermined time x _{i + 1} where no human body in the node (420), x _{i + 2} position in The non-human body node 430, the human body node 440 at the x _{i + 3} position, and the human body node 450 at the x _{i + 4} position can be formed.

Hereinafter, a graph-based non-human body position estimation method according to an embodiment of the present invention includes a visual odometry of a human body measured from a camera, data measured from the GPS and the IMU, 3D data measured by 3D LiDAR Constraint conditions connecting the nodes 410, 420, 430, 440 and 450 of the human body can be formed by using the ICP data.

, x_i+1(420)과 x_i+2(430)를 연결하는

, x_i+2(430)과 x_{i+ 3}(440)를 연결하는

, x_i+3(440)과 x_i+4(450)를 연결하는

이다.Referring to FIG. 4A, a constraint formed by a visual odometry of a human body measured from a camera connects x _i (410) and x _{i + 1} (420)

, connecting x _{i + 1} (420) and x _{i + 2} (430)

, connecting x _{i + 2} (430) and x _{i + 3} (440)

, connecting x _{i + 3} (440) and x _{i + 4} (450)

to be.

, x_i(410)과 x_{i+ 3}(440)를 연결하는

, x_i(410)과 x_i+4(450)를 연결하는

이다.Referring to FIG. 4B, constraints formed by data measured from the GPS and the IMU are used to connect x _i (410) and x _{i + 2} (430)

, connecting x _i (410) and x _{i + 3} (440)

, connecting x _i (410) and x _{i + 4} (450)

to be.

, x_i+2(430)과 x_i+4(450)를 연결하는

이다.Referring to FIG. 4C, the constraint formed by the data of the 3D ICP measured by the 3D LiDAR connects x _{i + 1} (420) and x _{i + 3} (440)

, connecting x _{i + 2} (430) and x _{i + 4} (450)

to be.

4D, a non-human body position estimation method based on a graph structure according to an exemplary embodiment of the present invention includes estimating a position of each of the nodes 410 and 420 according to a moving path of a human body using IMU, GPS, , 430, 440, 450) to form a graph structure.

Accordingly, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention can estimate the position of a human body based on the constructed graph structure.

According to the embodiment, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention can construct a graph structure by forming a constraint condition after the formation of a node, . In addition, the measurement data of the multiple sensors for constructing the graph structure is based on the data received at a predetermined time.

FIG. 5 shows a result of estimating the position of a human body using a graph-based non-human body position estimation method according to an embodiment of the present invention.

More specifically, Fig. 5 (a) shows the results obtained by comparing the positions of the non-human body estimated using the actual moving body of the human body and the method according to the embodiment of the present invention on the X axis, 5C shows a result obtained by comparing the position of a non-human body estimated using a method according to an embodiment of the present invention with the actual moving body of a non-human body, And the position of a human body estimated using the method according to the embodiment of the present invention in the Z axis.

Referring to FIG. 5 (a), when the positions of the unthinking body estimated using the actual moving body of the human body and the method according to the embodiment of the present invention are compared in the X axis, the mean error / m is 0.1706, and the root mean square error (RMSE) is 0.4161.

Referring to FIG. 5 (b), when the positions of the unthinking body estimated using the actual moving body of the human body and the method according to the embodiment of the present invention are compared in the Y axis, the mean error / 0.0997, and a root mean square error (RMSE) of 0.4625.

Referring to FIG. 5 (c), when the positions of the unthinking body estimated using the actual moving body of the human body and the method according to the embodiment of the present invention are compared in the Z axis, the mean error / Represents 0.3363, and the root mean square error (RMSE) represents 0.6290.

Accordingly, the non-human body position estimation method based on the graph structure according to the embodiment of the present invention can confirm that a stable non-human body position is estimated by using the graph structure constructed using multiple sensors, and the distortion, It is possible to obtain a position estimation result with a significantly low error.

FIG. 6 is a block diagram of a non-human body position estimation apparatus based on a graph structure according to an embodiment of the present invention.

Referring to FIG. 6, the non-human body position estimating apparatus 600 based on a graph structure according to the present invention measures data according to a moving path of a human body from multiple sensors, Construct a graph structure from constraints to estimate the position of the human body.

For this, the non-human body position estimating apparatus 600 based on a graph structure according to an embodiment of the present invention includes a measuring unit 610, a graph structure constructing unit 620, and a position estimating unit 630.

The measuring unit 610 measures data according to the moving path of the human body using multiple sensors.

For example, the measuring unit 610 measures velocity data and initial position data (initial position) generated based on sensor data from an IMU (Inertial Measurement Unit) equipped with an acceleration sensor and a gyro sensor mounted on a human body , And can measure position data from a GPS (Global Positioning System) mounted on a human body.

The measuring unit 610 measures the visual odometry of the human body from the camera and measures the visual odometry of the human body through a 3D ICP (Iterative Closest Point Fitting) algorithm based on a 3D rider (LiDAR; Light Detection And Ranging) The movement data of the human body can be measured.

Here, the IMU (Inertial Measurement Unit) uses a gyro sensor (gyroscope) as inertial measurement means. For example, the IMU calculates attitude-related sensor data such as forward rotation angle (roll axis direction), rightward direction (pitch axis direction) and gravity direction (yaw axis direction) Acceleration sensor data such as an inhuman body speed middle value can be generated using an acceleration sensor.

In addition, GPS (Global Position System) can acquire position data (e.g., latitude, longitude and ellipsoid height as geodesic coordinates of a WGS84 coordinate system) by communicating with a satellite.

In addition, the camera can acquire visual odometry data of a human body by capturing an image using CMOS or a CCD (Charge Coupled Device) image sensor.

Also, 3D LiDAR (Light Detection And Ranging) can be a radar system that irradiates laser pulses to a reflector (human body) and measures the position coordinates of the reflector by measuring the return time from the reflector. As is well known, 3D LiDAR stores position coordinates measured and calculated from reflection intensity as measurement data, and measurement data can be utilized as basic information for imaging 3D graph structure. The 3D LiDAR is widely used in the related art, and its structure and operation principle will not be described.

Depending on the embodiment, the multiple sensors of the IMU, GPS, Camera and 3D LiDAR may be located inside or outside each of the non-human body and the large structure, and the type and number of sensors to be located are not limited.

The graph structure construction unit 620 constructs a graph structure from nodes and constraints between nodes on the moving path of the human body based on the measured data.

The graph structure construction unit 620 constructs a graph structure based on measured data from multiple sensors including an IMU, a GPS, a camera, and a 3D lidar, a node representing each position along a moving path of the human body, Can be obtained.

The graph structure construction unit 620 can construct a graph structure through nodes based on the moving path of the human body and SLAM (Simultaneous Localization and Mapping) based on constraint conditions.

The SLAM detects the surrounding environment information of the human body and the human body from multiple sensors and processes the measured data to create a map (MAP) corresponding to the movement and position of the human body and estimates the absolute position of the human body .

For example, the graph structure construction unit 620 is a means for recognizing a real-time position and a final position of a human body. The graph structure construction unit 620 constructs a graph structure by using a SLAM process based on measurement data acquired by multiple sensors, Can be constructed.

The position estimation unit 630 estimates the position of the human body based on the graph structure.

The position estimation unit 630 can estimate the position of a human body using graph optimization based on the constructed graph structure.

For example, the position estimating unit 630 estimates the position of a human body using the graph optimization technique that can estimate a final position according to a moving path and a time of a human body in a graph structure formed by nodes and constraints .

However, the technique of acquiring the real-time position, the moving path, and the final position of the human body in the constructed graph structure can be applied variously in addition to the above-described graph optimization technique, and thus is not limited thereto.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented by any suitable means, including, for example, a processor, a controller, a device described above, a hardware component, a software component, and / or a combination of hardware components and software components Lt; / RTI > For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA) A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CDROMs and DVDs, magnetic optical media such as floppy disks, magnetooptical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

10: human body (or drone)
410, 420, 430, 440, 450:

Claims (13)

Measuring data according to a moving path of a human body using multiple sensors such as an IMU (Inertial Measurement Unit), a GPS (Global Positioning System), a camera, and a 3D rider (LiDAR; Light Detection And Ranging);
Constructing a graph structure based on the measured data based on a node in the path of the human body and a constraint between the nodes; And
Estimating a position of a human body based on the graph structure,
The step of constructing the graph structure
Constraint conditions formed by the human body's odometry data measured from the camera, velocity data measured from the GPS and the IMU, constraints formed by initial position data and position data, and 3D ICP (Iterative Closest Point fitting ) Constraint conditions for each node are optimized using constraint conditions formed by the data, the graph structure is constructed,
The step of estimating the position of the human body
Wherein the optimization of the constructed graph structure for estimating the position of the non-human body using graph optimization is performed based on the constructed graph structure, . The method according to claim 1,
The step of measuring the data
The velocity data and the initial position data generated based on the sensor data from the IMU (Inertial Measurement Unit) equipped with the acceleration sensor and the gyro sensor mounted on the human body are measured, Wherein the position data is measured from a GPS (Global Positioning System). 3. The method of claim 2,
The step of measuring the data
The visual odometry of the human body is measured from a camera and the movement data of the human body is measured through a 3D ICP (Iterative Closest Point Fitting) algorithm based on a 3D rider (LiDAR; Light Detection And Ranging) Non - body position estimation based on graph structure. The method of claim 3,
The step of constructing the graph structure
The node representing each position along the movement path of the human body based on the measured data from the multiple sensors including the IMU, the GPS, the camera and the 3D rider, and the constraint condition indicating the connection of the node Wherein the non-human body position estimation method is based on a graph structure. 5. The method of claim 4,
The step of constructing the graph structure
Wherein the graph structure is constructed through the node according to the moving path of the human body, and SLAM (Simultaneous Localization and Mapping) based on the constraint condition. delete A computer program stored in a computer-readable medium for performing the method of any one of claims 1 to 5. A measuring unit for measuring data according to a moving path of a human body using multiple sensors of an IMU (Inertial Measurement Unit), a GPS (Global Positioning System), a camera, and a 3D rider (LiDAR; Light Detection And Ranging);
A graph structure constructing unit for constructing a graph structure from a node and a constraint between nodes in the moving path of the human body based on the measured data; And
And a position estimator for estimating a position of a human body based on the graph structure,
The graph structure constructing unit
Constraint conditions formed by the human body's odometry data measured from the camera, velocity data measured from the GPS and the IMU, constraints formed by initial position data and position data, and 3D ICP (Iterative Closest Point fitting ) Constraint conditions for each node are optimized using constraint conditions formed by the data, the graph structure is constructed,
The position estimating unit
Wherein the optimization of the constructed graph structure for estimating the position of the non-human body is performed using a graph optimization method based on the constructed graph structure, . 9. The method of claim 8,
The measuring unit
The velocity data and the initial position data generated based on the sensor data from the IMU (Inertial Measurement Unit) equipped with the acceleration sensor and the gyro sensor mounted on the human body are measured, Wherein the position data is measured from a GPS (Global Positioning System). 10. The method of claim 9,
The measuring unit
The visual odometry of the human body is measured from a camera and the movement data of the human body is measured through a 3D ICP (Iterative Closest Point Fitting) algorithm based on a 3D rider (LiDAR; Light Detection And Ranging) Non - human body position estimator based on graph structure. 11. The method of claim 10,
The graph structure constructing unit
The node representing each position along the movement path of the human body based on the measured data from the multiple sensors including the IMU, the GPS, the camera and the 3D rider, and the constraint condition indicating the connection of the node Wherein the non-human body position estimating apparatus is based on a graph structure. 12. The method of claim 11,
The graph structure constructing unit
Wherein the graph structure is constructed through the node according to the moving path of the human body, and SLAM (Simultaneous Localization and Mapping) based on the constraint condition. delete

Images