Artificial neural network-based relative positioning method and device, for recovering lost distance information in environment where fixed node is not present and all nodes are mobile

Abstract

The present invention relates to an artificial neural network-based relative positioning method and device, for recovering lost distance information in an environment where a fixed node is not present and all nodes are mobile, the artificial neural network-based relative positioning method, according to one embodiment of the present invention, comprising the steps of: for a plurality of nodes arranged into a formation, collecting distance information between each of the nodes; checking whether a loss has occurred in the collected distance information, and recovering the lost distance information in the collected distance information; storing distance data including the distance information obtained by the recovering of the lost distance information in the collected distance information; training a first artificial neural network by using the stored distance data as input; and predicting the formation by estimating relative coordinate information of each of the nodes on the basis of the trained artificial neural network.

Description

Artificial neural network-based relative positioning method and apparatus for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving

[0001] The present invention relates to a technology for restoring distance information when it is lost, and more specifically, to a technology for restoring lost distance information during distance-based relative positioning using an artificial neural network in an environment where no fixed node (anchor node) exists and all nodes are moving.

[0002] With the recent expansion of robot applications in fields such as military operations and environmental exploration, research is underway to operate unmanned swarm robots capable of delivering performance exceeding that of a single robot. When robots requiring position estimation are referred to as "mobile terminals," the Global Positioning System (GPS) is primarily used to determine their location. However, there are limitations to utilizing this in indoor environments or situations involving GPS jamming. In such cases, a new approach is required to estimate the precise locations of mobile terminals without the need for GPS.

[0003] Generally, positioning is performed based on reference nodes known as anchors, which are used for formation estimation. However, it is difficult to establish fixed reference nodes in situations where all nodes are in motion. Furthermore, in complex dynamic environments where all nodes move in real-time, the coordinates of all nodes and the distance information between them are updated at regular intervals. Consequently, the formation formed by all nodes changes whenever the distance information is updated. To achieve accurate positioning in such dynamic environments, real-time updated distance information must be continuously collected and analyzed.

[0004] Various positioning devices can be utilized to acquire distance information, such as UWB and Wi-Fi. However, during the process where moving nodes exchange distance information in real-time to estimate their location, there is a possibility that distance information may be lost due to signal loss or obstacles.

[0005] The embodiments of the present invention aim to provide an artificial intelligence-based relative position estimation method and apparatus that enables positioning using only distance information in a situation where no fixed node exists and all nodes are moving, and enables normal positioning by restoring the lost distance information even if the distance information is lost. Other unspecified objectives of the present invention may be additionally considered within the scope that can be easily inferred from the following detailed description and its effects.

[0006] The problem to be solved by the present invention is not limited thereto and may be extended in various ways within an environment that does not deviate from the spirit and scope of the present invention.

[0007] According to one embodiment of the present invention, a relative positioning method performed by a relative positioning device comprises: a step of collecting distance information between each node for a plurality of nodes forming a formation; a step of checking whether there is loss in the collected distance information and restoring the lost distance information in the collected distance information; a step of storing distance data including the distance information restored from the collected distance information; a step of training a first artificial neural network using the stored distance data as input; and a step of predicting the formation by estimating the relative coordinate information of each node based on the trained artificial neural network. An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving may be provided.

[0008] The above method may include a distance error in the distance information between each of the collected nodes.

[0009] The step of restoring the distance information above can restore the lost distance information by using an interpolation technique that estimates the lost current distance value using the past distance value of the section where the distance information was lost.

[0010] In the step of restoring the distance information, if the node with lost distance information moves in a straight line, the lost distance information can be restored using linear interpolation.

[0011] In the step of restoring the distance information, if the node with lost distance information moves in a curved shape, the lost distance information can be restored using polynomial interpolation.

[0012] The step of restoring the distance information above can restore the lost distance information using past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0013] The step of restoring the distance information above can restore the lost distance information by maintaining the final observation value in past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0014] The step of restoring the distance information above may restore the lost distance information by using the average value of the past distance information from past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0015] The step of restoring the distance information above can restore the lost distance information by using a moving average value calculated by assigning weights to the past distance information in past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0016] The step of restoring the distance information above can check for loss in the collected distance information and restore the lost distance information based on a second artificial neural network trained to detect and restore the existence of lost distance information.

[0017] Meanwhile, according to another embodiment of the present invention, an artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving may be provided, comprising: a memory for storing one or more programs; and a processor for executing the one or more programs stored therein. The processor collects distance information between each node for a plurality of nodes forming a large group, checks for loss in the collected distance information, restores lost distance information from the collected distance information, stores distance data containing distance information that has restored lost distance information from the collected distance information, trains a first artificial neural network using the stored distance data as input, and estimates relative coordinate information of each node based on the trained artificial neural network to predict the large group.

[0018] The distance information between each node collected above may include distance errors.

[0019] The processor can restore the lost distance information by using an interpolation technique that estimates the lost current distance value using the past distance value of the section where the distance information was lost.

[0020] The above processor can restore the lost distance information using linear interpolation if the node with lost distance information moves in a straight line.

[0021] The above processor can restore the lost distance information using polynomial interpolation if the node with lost distance information moves in a curved shape.

[0022] The processor can restore the lost distance information using past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0023] The processor can restore the lost distance information by maintaining the final observation value in past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0024] The processor can restore the lost distance information by using the average value of the past distance information from past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0025] The processor can restore the lost distance information by using a moving average value calculated by weighting the past distance information in past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0026] The processor can detect whether there is a loss in the collected distance information and restore the lost distance information based on a second artificial neural network trained to detect and restore the existence of lost distance information.

[0027] The disclosed technology may have the following effects. However, this does not mean that a specific embodiment must include all of the following effects or only the following effects; therefore, the scope of the rights of the disclosed technology should not be understood as being limited by this.

[0028] Embodiments of the present invention enable positioning using only distance information in a situation where no fixed node exists and all nodes are moving, and even if distance information is lost, the lost distance information can be restored to enable normal positioning.

[0029] Embodiments of the present invention enable efficient system operation regardless of distance information loss by verifying and restoring distance information between cluster nodes when such information is lost, and can maximize the stability and flexibility of an artificial neural network-based relative position estimation system utilizing distance information in a situation where there are no fixed nodes and distance information between all nodes is updated at regular time intervals.

[0030] Embodiments of the present invention can contribute to improving the reliability and performance of robot systems used in military operations and various fields by determining the accurate relative positions of cluster nodes through the restoration of lost distance information. In particular, the embodiments of the present invention can be very useful in environments where GPS-based position estimation is impossible or restricted.

[0031] FIG. 1 is a diagram illustrating a relative positioning operation performed by an artificial neural network-based relative positioning device according to an embodiment of the present invention.

[0032] Figures 2 and 3 are diagrams illustrating the case where the measurement distance between nodes is lost.

[0033] FIG. 4 is a flowchart illustrating an artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, according to an embodiment of the present invention.

[0034] FIG. 5 is a diagram illustrating an example of restoring lost distance information using a linear interpolation technique according to an embodiment of the present invention.

[0035] FIG. 6 is a diagram illustrating an example of restoring lost distance information using a final observation value maintenance technique according to an embodiment of the present invention.

[0036] FIG. 7 is a diagram illustrating an example of restoring lost distance information using a past distance information averaging technique according to an embodiment of the present invention.

[0037] FIG. 8 is a diagram illustrating an example of restoring lost distance information using a weighted moving average technique according to an embodiment of the present invention.

[0038] FIG. 9 is a configuration diagram of an artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, according to an embodiment of the present invention. It is a diagram for explaining the operation of restoring lost distance information using an artificial neural network according to an embodiment of the present invention.

[0039] FIGS. 10a to 10c are drawings for explaining the operation of restoring lost distance information using an artificial neural network according to an embodiment of the present invention.

[0040] The present invention is capable of various modifications and may have various embodiments; specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the technical spirit and scope of the present invention. In describing the present invention, detailed descriptions of related prior art are omitted if it is determined that such detailed descriptions may obscure the essence of the present invention.

[0041] Terms such as "first," "second," etc., may be used to describe various components, but the components are not limited by these terms. The terms are used solely for the purpose of distinguishing one component from another.

[0042] The terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. While the terms used in this invention have been selected to be as widely used as possible in consideration of their functions within the invention, they may vary depending on the intent of those skilled in the art, case law, or the emergence of new technologies. Furthermore, in specific cases, terms have been arbitrarily selected by the applicant, and in such cases, their meanings will be described in detail in the relevant description of the invention. Therefore, the terms used in this invention should be defined not merely by their names, but based on their meanings and the overall content of the invention.

[0043] A singular expression includes a plural expression unless the context clearly indicates otherwise. In the present invention, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0044] [Explanation of the symbol]

[0045] 100: Relative positioning device

[0046] 110: Memory

[0047] 120: Processor

[0048] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing with reference to the accompanying drawings, identical or corresponding components are given the same reference numerals, and redundant descriptions thereof will be omitted.

[0049] FIG. 1 is a diagram illustrating a relative positioning operation performed by an artificial neural network-based relative positioning device according to an embodiment of the present invention.

[0050] As illustrated in FIG. 1, an artificial neural network-based relative positioning device (100) according to one embodiment of the present invention can derive the relative positions of all nodes as coordinates using only distance information between nodes in a situation where all nodes are moving in real time. At this time, the relative positioning device (100) can predict the formation of nodes by estimating the two-dimensional coordinates of all nodes using artificial neural network-based distance information between nodes to restore lost distance information in an environment where no fixed nodes exist and all nodes are moving.

[0051] Here, the artificial neural network-based relative positioning device (100) collects distance information between nodes as input data in an environment where no fixed nodes exist. When the number of nodes is N, N There are C2 measured distance information points. For example, when the number of nodes is 5, there are 10 input distance information points between nodes.

[0052] And the artificial neural network-based relative positioning device (100) can estimate a position without using a fixed reference node by using an artificial neural network, and performs a relative positioning operation that is capable of positioning by restoring distance information when it is lost even when all nodes are moving in real time. Here, the artificial neural network can perform relative positioning in real time by restoring distance information even if it is lost.

[0053] The relative positioning device (100) estimates the relative coordinate information of each node based on a learned artificial neural network and outputs the relative coordinate information of each node as output data. When the number of nodes is N, there may be (2×N-3) output values. For example, when the number of nodes is 5, the size of the output relative coordinate information is 7 (=2×5-3).

[0054] Meanwhile, in one embodiment of the present invention, ambiguity may arise in the problem of finding coordinates based on distance information. For example, even though the nodes are of the same size, various coordinate information may exist during performance evaluation.

[0055] To resolve such ambiguity, a relative positioning device (100) according to one embodiment of the present invention can perform a data generation rule that eliminates ambiguity and minimizes limitations.

[0056] Nodes can be operated by applying the following three rules 1 to 3 for a single solution.

[0057] Rule 1: Assume node 1 is the origin (x1= y1= 0).

[0058] Rule 2: Assume that the x value of node 2 is positive and the y value is 0 (x2 > 0, y2 = 0)

[0059] Rule 3: Assume that node 3 exists in the 1st or 2nd quadrant (y3 > 0)

[0060] Here, x n , y n Each represents the x and y values ​​of the nth node.

[0061] Figures 2 and 3 are diagrams illustrating the case where the measurement distance between nodes is lost.

[0062] An example of an environment in which no fixed nodes exist and all nodes, from node 1 to node 5, move is illustrated in FIG. 2.

[0063] As illustrated in Fig. 3, distance information may be lost due to signal interference or obstacles in a situation where all nodes are moving in real time. A case where one piece of distance information is lost is shown. R represents the measured value, and X represents the lost value. For each sample, the lost distance information is d from sample 1. 1,4 , in sample 2, d 2,4 , in sample 3, d 1,5 , in sample 4, d 3,4 , in sample 5, d 1,4 , in sample 6, d 4,5 It can be.

[0064] To solve this, the relative positioning device (100) can estimate the 2D coordinates of all nodes using artificial neural network-based inter-node distance information to restore lost distance information in an environment where no fixed nodes exist and all nodes are moving. Here, the relative positioning device (100) can restore lost distance information by utilizing an interpolation technique, restore lost distance information by utilizing past data based on lost distance information, or restore lost distance information by utilizing artificial intelligence (AI).

[0065] FIG. 4 is a flowchart illustrating an artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, according to an embodiment of the present invention.

[0066] In step S101, the artificial neural network-based relative positioning device (100) collects distance information between each node for a plurality of nodes forming a relative group. Here, the artificial neural network-based relative positioning device (100) can collect and store distance information between all existing nodes. At this time, there are no limitations on the method or technique for collecting distance information. The collected distance information may include distance errors. For example, the distance measurement error between nodes may be Gaussian noise following a normal distribution. In such a case, in one embodiment of the present invention, the impact on prediction accuracy due to distance errors can be minimized by using a dataset containing Gaussian noise.

[0067] In step S102, the artificial neural network-based relative positioning device (100) checks for loss in the collected distance information and restores the lost distance information from the collected distance information. At this time, the relative positioning device (100) can check for loss of distance information by counting the number of distance information obtained based on the number of nodes, or by determining whether the obtained distance information is a normal value. If loss of distance information is confirmed, there are a total of three methods to restore it.

[0068] The first method is to use an interpolation technique. The relative positioning device (100) can use an interpolation technique to infer the lost current distance value by using the past distance value of the section where the distance information was lost when the distance information was lost. Various interpolation techniques can be used to restore the lost current distance value. Examples may include linear interpolation or polynomial interpolation.

[0069] The relative positioning device (100) can restore lost distance information using linear interpolation if the node moves in a straight line. Through linear interpolation, lost distance information can be linearly estimated based on past distance values.

[0070] Alternatively, if the node moves in a curved shape rather than a straight line, the relative positioning device (100) can restore lost distance information using polynomial interpolation. The relative positioning device (100) can create a polynomial that includes all of the points based on the multiple distance information points corresponding to the preceding section of the lost distance information centered on the lost distance information, and estimate and restore the lost distance information based on this.

[0071] Second, there is a method of using past data. Past data refers to data prior to or accumulated distance information based on the lost distance information. The relative positioning device (100) can restore the lost distance information by using the past data immediately prior to the lost distance information as is. Alternatively, the relative positioning device (100) can restore the lost distance information by using the average of previously measured distance values. For example, the relative positioning device (100) can restore the lost distance information by calculating a centered moving average. Additionally, the relative positioning device (100) can restore the lost distance information by calculating a moving average by assigning weights to recently measured values.

[0072] The third method involves using an artificial neural network. This method applies a second artificial neural network, separate from the first artificial neural network used in step S104, to detect the presence or absence of lost distance information and perform restoration. When using the second artificial neural network, it is possible to predict the section where distance information is lost by utilizing all distance information other than the lost information, and to restore the lost distance values ​​based on existing data. However, this method requires the consideration that training must sufficiently assume various situations and environments where distance information is lost.

[0073] In step S103, the artificial neural network-based relative positioning device (100) stores distance data containing distance information in which lost distance information is restored from the collected distance information.

[0074] In step S104, the artificial neural network-based relative positioning device (100) trains the first artificial neural network using stored distance data as input. At this time, only distance information between each existing node is used as input data, or additional information may be used. The output data of the artificial neural network is the relative coordinate information of each node.

[0075] In step S105, the artificial neural network-based relative positioning device (100) estimates the relative coordinate information of each node based on the learned artificial neural network to predict the size.

[0076] FIG. 5 is a diagram illustrating an example of restoring lost distance information using a linear interpolation technique according to an embodiment of the present invention.

[0077] As shown in FIG. 5, distance information between nodes 1 through 5 is collected for each sample measuring distance information. In this case, there is a case where one piece of distance information is lost in the 5th sample, and in a situation where all nodes are moving in real time, due to signal interference or obstacles, the distance value d between nodes 1 and 41,4 This can be lost.

[0078] Here, the relative positioning device (100) has a distance value d 1,4 In the event of this loss, the lost distance information can be restored using linear interpolation according to [Mathematical Formula 1] below. Here, linear interpolation is a method of restoring the lost information by linearly estimating it using past values ​​based on the lost distance information. In the example illustrated in FIG. 5, the relative positioning device (100) is Lost distance values ​​can be restored as follows.

[0079]

[0080] Here, The number of distance measurements, is the lost distance value, is the n-1th distance value, represents the n-2nd distance value.

[0081] FIG. 6 is a diagram illustrating an example of restoring lost distance information using a final observation value maintenance technique according to an embodiment of the present invention.

[0082] As shown in FIG. 6, the relative positioning device (100) has a distance value d 1,4 In the event of this loss, the lost distance information can be restored by utilizing the Last Observation Carried Forward (LOCF) technique, which is the first technique that utilizes past data based on the lost distance information. Here, the Last Observation Carried Forward technique uses the last measured distance value as is. That is, the relative positioning device (100) can restore the lost distance information by maintaining the last observation from past distance data that includes at least one past distance value.

[0083] In the example illustrated in FIG. 6, the relative positioning device (100) measures the distance value d measured in the fourth sample. 1,4Using 5 as is, the distance value d of the 5th sample 1,4 It can be restored.

[0084] FIG. 7 is a diagram illustrating an example of restoring lost distance information using a past distance information averaging technique according to an embodiment of the present invention.

[0085] As illustrated in FIG. 7, the relative positioning device (100) has a distance value d 1,4 In the event of this loss, the lost distance information can be restored using a past distance information averaging technique according to [Equation 2] below. Here, the past distance information averaging technique uses the average value of past distance information. The relative positioning device (100) can restore the lost distance information by calculating the average of previously measured distance values. That is, the relative positioning device (100) can restore the lost distance information by using the average value of past distance information from past distance data that includes at least one past distance value.

[0086] In the example illustrated in FIG. 7, the relative positioning device (100) is Lost distance values ​​can be restored as follows.

[0087]

[0088] Here, The number of distance measurements, is the i-th distance value, represents the n-1th distance value.

[0089] FIG. 8 is a diagram illustrating an example of restoring lost distance information using a weighted moving average technique according to an embodiment of the present invention.

[0090] As illustrated in FIG. 8, the relative positioning device (100) has a distance value d 1,4In the event of this loss, the lost distance information can be restored using the Weighted Moving Average (WMA) technique according to [Equation 3] below. Here, the Weighted Moving Average technique calculates a moving average by assigning weights to past distance information, assuming that the more recently measured value is more reliable. The relative positioning device (100) can restore the lost distance information by using a moving average value calculated by assigning weights to past distance information in past distance data that includes at least one past distance value.

[0091] In the example illustrated in FIG. 8, the relative positioning device (100) is Lost distance values ​​can be restored as follows.

[0092]

[0093] Here, The number of distance measurements, represents the n-1th distance value.

[0094] FIG. 9 is a configuration diagram of an artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, according to an embodiment of the present invention.

[0095] As illustrated in FIG. 9, an artificial neural network-based relative positioning device (100) for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, according to one embodiment of the present invention, includes a memory (110) and a processor (120). However, not all of the illustrated components are essential components. The artificial neural network-based relative positioning device (100) may be implemented with more components than those illustrated, or with fewer components.

[0096] Below, the specific configuration and operation of each component of the artificial neural network-based relative positioning device (100) of FIG. 9 will be described.

[0097] The memory (110) stores one or more programs related to an artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving.

[0098] The processor (120) executes one or more programs stored in memory (110). The processor (120) collects distance information between each node for a plurality of nodes forming a large group, checks for loss in the collected distance information, restores lost distance information from the collected distance information, stores distance data containing distance information that restores lost distance information from the collected distance information, trains a first artificial neural network using the stored distance data as input, and predicts the large group by estimating the relative coordinate information of each node based on the trained artificial neural network.

[0099] According to the embodiments, the distance information between each of the collected nodes may include a distance error.

[0100] According to embodiments, the processor (120) can restore the lost distance information by using an interpolation technique that estimates the lost current distance value using the past distance value of the section where the distance information was lost.

[0101] According to embodiments, the processor (120) can restore the lost distance information using linear interpolation if the node with lost distance information moves in a straight line.

[0102] According to embodiments, the processor (120) can restore the lost distance information using polynomial interpolation if the node with lost distance information moves in a curved shape.

[0103] According to embodiments, the processor (120) can restore the lost distance information using past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

[0104] According to embodiments, the processor (120) can restore the lost distance information by maintaining the final observation in the past distance data containing the at least one past distance value.

[0105] According to embodiments, the processor (120) can obtain the lost distance information by using the average value of the past distance information in the past distance data containing the at least one past distance value.

[0106] According to embodiments, the processor (120) can use the lost distance information by using a moving average value calculated by weighting the past distance information in past distance data that includes the at least one past distance value.

[0107] According to embodiments, the processor (120) can detect whether there is a loss in the collected distance information and restore the lost distance information based on a second artificial neural network trained to detect and restore the existence of lost distance information.

[0108] FIGS. 10a to 10c are drawings for explaining the operation of restoring lost distance information using an artificial neural network according to an embodiment of the present invention.

[0109] In step S210, the relative positioning device (100) receives distance information between each node for a plurality of nodes forming a large group.

[0110] In step S220, the relative positioning device (100) restores lost distance information using a second artificial neural network. Here, the second artificial neural network may be trained to detect the presence or absence of lost distance information and to perform restoration. In steps S221 and S222, the relative positioning device (100) detects the presence or absence of lost distance information in the distance information, estimates the lost distance information based on the second artificial neural network, and restores it.

[0111] Here, the second artificial neural network can be any one of the following: a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), or a Transformer-family neural network. RNNs and CNNs are representative neural network structures specialized for sequential data processing. The second artificial neural network can be a neural network suitable for handling data such as time-series data. Among the various neural networks, any network optimized for time-series data processing can be utilized. Transformer-family neural networks, which demonstrate excellent performance in time-series data processing, can also be applied as the second artificial neural network to recover lost distance information.

[0112] In step S230, the relative positioning device (100) learns a first artificial neural network using distance data containing restored distance information as input, and estimates the relative coordinate information of each node based on the learned first artificial neural network.

[0113] In step S240, the relative positioning device (100) outputs estimated coordinate information to predict the formation of all nodes.

[0114] Meanwhile, according to one embodiment of the present invention, the various embodiments described above may be implemented as software comprising instructions stored on a machine-readable storage medium (e.g., a computer). The machine may include an electronic device (e.g., electronic device (A)) according to the disclosed embodiments, which is a device capable of calling instructions stored from the storage medium and operating according to the called instructions. When instructions are executed by a processor, the processor may perform a function corresponding to the instructions directly or by using other components under the control of the processor. Instructions may include code generated or executed by a compiler or an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory" means only that the storage medium does not contain a signal and is tangible, and does not distinguish whether data is stored semi-permanently or temporarily in the storage medium.

[0115] In addition, according to one embodiment of the present invention, the method according to the various embodiments described above may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed online in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or through an application store (e.g., Play Store™). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a storage medium such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0116] Additionally, according to one embodiment of the present invention, the various embodiments described above may be implemented in a recording medium readable by a computer or a similar device using software, hardware, or a combination thereof. In some cases, the embodiments described herein may be implemented as the processor itself. According to a software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein.

[0117] Meanwhile, computer instructions for performing processing operations of the device according to the various embodiments described above may be stored in a non-transitory computer-readable medium. When computer instructions stored in such a non-transitory computer-readable medium are executed by the processor of a specific device, they cause the specific device to perform processing operations in the device according to the various embodiments described above. A non-transitory computer-readable medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short moment, such as a register, cache, or memory. Specific examples of a non-transitory computer-readable medium may include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, ROMs, etc.

[0118] Additionally, each component (e.g., module or program) according to the various embodiments described above may be composed of a single or multiple entities, and some of the aforementioned sub-components may be omitted, or other sub-components may be further included in the various embodiments. Generally or additionally, some components (e.g., module or program) may be integrated into a single entity to perform the same or similar functions as those performed by each of the respective components prior to integration. The operations performed by the module, program, or other components according to the various embodiments may be executed sequentially, in parallel, iteratively, or heuristically, or at least some operations may be executed in a different order, omitted, or other operations added.

[0119] Although preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above. Various modifications are possible by those skilled in the art without departing from the essence of the invention as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present invention.

Claims

1. In a relative positioning method performed by a relative positioning device, A step of collecting distance information between each node for multiple nodes forming a large group; A step of checking for loss in the collected distance information and restoring the lost distance information from the collected distance information; A step of storing distance data containing distance information in which lost distance information is restored from the above-mentioned collected distance information; A step of training a first artificial neural network using the above-mentioned stored distance data as input; and An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, comprising the step of estimating relative coordinate information of each node based on the above-mentioned learned artificial neural network to predict the size.

2. In Paragraph 1, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where there are no fixed nodes and all nodes are moving, in which distance information between each of the collected nodes includes distance errors.

3. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, by using an interpolation technique that estimates the lost current distance value using the past distance value of the segment where distance information is lost to restore the lost distance information.

4. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, wherein the lost distance information is restored using linear interpolation when the node with lost distance information moves in a straight line.

5. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, wherein the lost distance information is restored using polynomial interpolation when the node with lost distance information moves in a curved shape.

6. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, by using past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

7. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed node exists and all nodes are moving, by maintaining a final observation value in past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

8. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, by using the average value of the past distance information from past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

9. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, by using a moving average value calculated by weighting the past distance information in past distance data that includes at least one past distance value collected at a past point in time based on the lost distance information.

10. In Paragraph 1, The step of restoring the above distance information is, An artificial neural network-based relative positioning method for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, based on a second artificial neural network trained to detect and restore the presence or absence of lost distance information, checking for loss in the collected distance information and restoring the lost distance information from the collected distance information.

11. Memory for storing one or more programs; and It includes a processor that executes one or more of the above-mentioned stored programs, and The above processor is, For multiple nodes forming a large group, collect distance information between each node, and Check for loss in the above-mentioned collected distance information, and restore the lost distance information from the above-mentioned collected distance information, Store distance data containing distance information in which lost distance information is restored from the above-mentioned collected distance information, and A first artificial neural network is trained using the above-mentioned stored distance data as input, and An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which estimates relative coordinate information of each node based on the above-mentioned learned artificial neural network to predict the formation.

12. In Paragraph 11, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where there are no fixed nodes and all nodes are moving, in which distance information between each of the collected nodes includes distance errors.

13. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which restores the lost distance information using an interpolation technique that estimates the lost current distance value using past distance values ​​of the segment where distance information is lost.

14. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, wherein the lost distance information is restored using linear interpolation when the node with lost distance information moves in a straight line.

15. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, wherein the lost distance information is restored using polynomial interpolation when the node with lost distance information moves in a curved shape.

16. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which restores said lost distance information using past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

17. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which restores said lost distance information by maintaining a final observation value in past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

18. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which restores said lost distance information by using the average value of the past distance information from past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

19. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, which restores said lost distance information using a moving average value calculated by weighting the past distance information in past distance data containing at least one past distance value collected at a past point in time based on the lost distance information.

20. In Paragraph 11, The above processor is, An artificial neural network-based relative positioning device for restoring lost distance information in an environment where no fixed nodes exist and all nodes are moving, based on a second artificial neural network trained to detect and restore the presence or absence of lost distance information, checking for loss in the collected distance information and restoring the lost distance information from the collected distance information.

Images