KR102324989B1 - Mobile body, management server, and operating method thereof - Google Patents

Abstract

According to an embodiment, the management server collects first fingerprints measured along with the detection of the surrounding environment at a plurality of points where the surrounding environment is sensed in order to generate a SLAM map from a first moving body moving in a predetermined driving space. collect, match the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the driving space, and correspond to the fingerprint based on the matched first coordinates and the first fingerprints learning a neural network estimating coordinates on a SLAM map, receiving a second fingerprint from a second moving object moving in a driving space based on path information set on the SLAM map, and using a neural network to respond to the second fingerprint The second coordinates on the SLAM map may be estimated, and the driving information of the second moving object may be determined and transmitted based on the route information and the estimated second coordinates.

Description

A mobile body, a management server, and an operation method thereof

Embodiments relate to a mobile body, a management server, and an operating method thereof.

In order to save manpower and/or improve conveyance accuracy in a production line of a factory or a warehouse, various moving objects such as an unmanned transport vehicle and an unmanned transport robot that automatically travel along a target path have been introduced. The moving object constructs, for example, magnetic paint, induction wire, induction tape, QR code, etc. on the traveling path of the moving object, and detects the magnetic paint, tape, or QR code, etc. to determine the target path. may be induced to follow.

In such a way to induce the movement of a moving object, the process of constructing a detectable magnetic paint, an induction wire, an induction tape, and a QR code on the target path must be preceded. In addition, depending on the environment in the factory, it may be difficult to secure a space for guiding the moving object, or additional cost for securing a path may be required. In addition, additional work is required due to damage to the route change, induction paint, wire, tape, etc.

According to an embodiment, the management server can determine the driving information of the moving object more quickly and accurately by linking the 5G network-based fingerprint with the SLAM map, and deliver the driving information to the moving object at high speed through the 5G network.

According to an embodiment, the management server estimates the path information and coordinates for the moving object using a neural network that estimates the coordinates on the SLAM map corresponding to the fingerprint, thereby inducing paint, wire, tape, QR for inducing the path of the moving object. It is possible to more accurately determine the driving information of the moving object without a QR code or the like.

According to one embodiment, in response to information received through the 5G network module and IoT sensor receiving device provided in the mobile body, the management server determines the driving route by estimating the coordinates on the SLAM map in real time, so that driving information according to changes in the surrounding environment Able to respond quickly to changes in

According to an embodiment, by adding only the 5G network module and the IoT sensor receiving device, it is possible to receive more accurate route information from the management server without replacing the existing moving object.

According to an embodiment, it is possible to prevent a collision of a moving object and a safety accident due to the sudden appearance of an obstacle through 5G network-based high-speed, high-definition, and real-time image analysis.

According to an embodiment of the present disclosure, the method of operating the management server includes the first mobile device that moves in a predetermined driving space, the first measured along with the detection of the surrounding environment at a plurality of points where the surrounding environment is sensed in order to generate a SLAM map. 1 Collecting fingerprints; matching the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the driving space; learning a neural network for estimating coordinates on the SLAM map corresponding to a fingerprint based on the matched first coordinates and first fingerprints; receiving a second fingerprint from a second mobile body moving in the driving space based on route information set on the SLAM map; estimating second coordinates on the SLAM map corresponding to the second fingerprint by using the neural network; determining driving information of the second moving object based on the path information and the estimated second coordinates; and transmitting the driving information.

Collecting the first fingerprints may include collecting the first fingerprints based on propagation characteristics of a signal transmitted from at least one base station in a 5G network.

The method of operating the management server may further include setting path information on the LAM map in consideration of obstacles detected at the plurality of points based on the learning result of the neural network.

The determining of the route information may include: acquiring information on a shadow area in the driving space based on the first fingerprints; and determining the path information corresponding to the shaded area based on the neural network.

In the collecting of the first fingerprints, the first mobile body detects the surrounding environment at the plurality of points, based on a result of detecting the surrounding environment at the plurality of points. The method may further include generating the SLAM map.

The plurality of points may include at least one of at least one IoT sensor and at least one access point installed in the driving space.

The method of operating the management server may include: receiving in real time a driving environment image captured by at least one photographing device mounted on the second moving object; analyzing the driving environment image; generating an alarm notifying whether an object appears in the driving space based on the analysis result; and transmitting the alarm to the second moving object.

The analyzing of the driving environment image may include analyzing the driving environment image using a pre-trained deep neural network.

The generating of the alarm may include: recognizing whether an object appears in the driving space based on the analysis result; and generating an alarm according to the recognition.

the second mobile body includes a 5G network module, and the second mobile body receives information on the surrounding environment sensed at the plurality of points through the 5G network module; and monitoring the driving of the second moving object based on the information on the surrounding environment.

The method of operating the management server may include: correcting the driving information according to the monitoring result; and transmitting the modified driving information to the second mobile unit through the 5G network module.

According to an embodiment, a method of operating a moving object includes transmitting fingerprints collected as it moves in a predetermined driving space; receiving driving information determined based on coordinates and route information on a SLAM map estimated in response to the fingerprints from a pre-trained neural network; determining a driving parameter based on the driving information; and performing driving according to the driving parameters.

Receiving the driving information may include receiving the driving information from any one of a management server and a control interface.

The mobile body may include a 5G network module, and the step of transmitting the fingerprints may include transmitting the fingerprints collected as the mobile device moves through a predetermined driving space through the 5G network module.

The method of operating the moving object may include transmitting information on a surrounding environment that is changed according to the driving of the moving object; receiving driving information corrected according to a monitoring result based on the changed surrounding environment information; and changing the driving parameter according to the modified driving information. and driving the movable body according to the changed driving parameter.

The method of operating the moving object may include: capturing an image of a surrounding environment through at least one photographing device mounted on the moving object; transmitting the surrounding environment image; and receiving an alarm generated according to whether an object appears in the driving space recognized based on a result of analyzing the surrounding environment image.

According to an embodiment, the management server generates a SLAM map from a first mobile body moving in a predetermined driving space at a plurality of points where the surrounding environment is sensed and a first fingerprint measured along with the sensing of the surrounding environment. collecting (fingerprints), matching the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the travel space, the matched first coordinates and the first finger a processor for training a neural network estimating coordinates on the SLAM map corresponding to a fingerprint based on the prints; and a communication interface for receiving a second fingerprint from a second mobile body moving in the driving space based on the route information set on the SLAM map, wherein the processor uses the neural network to access the second fingerprint. The corresponding second coordinates on the SLAM map are estimated, and the driving information of the second mobile body is determined based on the route information and the estimated second coordinates.

According to an embodiment, the moving object transmits fingerprints collected as it moves in a predetermined driving space, and a driving determined based on coordinates and path information on a SLAM map estimated in response to the fingerprints in a pre-trained neural network a communication interface for receiving information; and a processor that determines a driving parameter based on the driving information and performs driving according to the driving parameter.

According to one side, the management server can determine the driving information of the moving object more quickly and accurately by linking the 5G network-based fingerprint with the SLAM map, and deliver the driving information to the moving object at high speed through the 5G network.

According to one side, the management server estimates the path information and coordinates for the moving object using a neural network that estimates the coordinates on the SLAM map corresponding to the fingerprint. It is possible to more accurately determine the driving information of the moving object without a QR code or the like.

According to one side, in response to information received through the 5G network module and the IoT sensor receiving device provided in the mobile body, the management server estimates the coordinates on the SLAM map in real time to determine the driving route, so that the driving information according to changes in the surrounding environment is obtained. Able to respond quickly to changes.

According to one side, by adding only the 5G network module and the IoT sensor receiving device, it is possible to receive more accurate route information from the management server without replacing the existing moving body.

According to one side, high-speed, high-definition, real-time video analysis based on 5G network can prevent collisions and safety accidents of moving objects due to the sudden appearance of obstacles.

1 is a flowchart illustrating a method of operating a management server according to an embodiment;
FIG. 2 is a view for explaining a method for a mobile body to collect fingerprints according to an exemplary embodiment;
3 is a view for explaining a process of generating an alarm through image analysis in a management server according to an embodiment;
4 is a flowchart illustrating a method of operating a moving object according to an embodiment;
5 is a diagram for explaining components of a mobile body and a control flow between the components according to an embodiment;
6 is a perspective view of a movable body according to an embodiment;
7 is a configuration diagram of a movable body according to an embodiment;
8 is a block diagram of a management server according to an embodiment;

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Like reference numerals in each figure indicate like elements.

Various modifications may be made to the embodiments described below. It should be understood that the embodiments described below are not intended to limit the embodiments, and include all modifications, equivalents or substitutes thereto.

Terms used in the examples are only used to describe specific examples, and are not intended to limit the examples. The singular expression includes the plural expression unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, in the description with reference to the accompanying drawings, the same components are assigned the same reference numerals regardless of the reference numerals, and the overlapping description thereof will be omitted. In the description of the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

1 is a flowchart illustrating a method of operating a management server according to an exemplary embodiment. Referring to FIG. 1 , the management server according to an embodiment of the present invention generates a SLAM map from a first moving body moving in a predetermined driving space at a plurality of points where the surrounding environment is sensed and measured together with the sensing of the surrounding environment. First fingerprints are collected (110). The 'first moving body' may correspond to a moving body that moves to collect information for estimating coordinates and determining route information prior to actual driving in a predetermined driving space, such as a factory, cargo loading dock, parking lot, etc. have. The first mobile body includes, for example, an automatic guided vehicle (AGV), an unmanned transport robot, an unmanned vehicle (UMV), as well as an autonomous vehicle, a transport vehicle driven by a driver, and the like. can be understood The first movable body may be the same as or different from the second movable body to be described later.

Here, 'collecting the first fingerprints measured together with the sensing of the surrounding environment at a plurality of points where the surrounding environment is sensed to generate a SLAM map' refers to the result of sensing the surrounding environment at a plurality of points. This means that the operation of generating the SLAM map based on the basis and the operation of collecting the first fingerprints at a plurality of points are performed together. In other words, the management server generates a SLAM map corresponding to the location of the first mobile body by the SLAM technique based on the result of detecting the surrounding environment at a plurality of points, and interworks with the SLAM map corresponding to the plurality of points on the SLAM map. The first fingerprints may be collected at the coordinates. The plurality of points may include, for example, a signal transmitter such as at least one IoT sensor and at least one access point installed in the driving space.

SLAM (Simultaneous Localization and Mapping) technique is a simultaneous localization and mapping technique. According to the SLAM technique, a moving object, such as a robot, moves in an arbitrary space and searches the surroundings, while estimating a map and a current location of the space.

The fingerprint(s) may correspond to location information used in a fingerprint technique. The fingerprint technique measures the signal strength transmitted from a signal transmitter (eg, an access point, an IoT sensor, or an IoT device, etc.) in advance and stores it in a database, for example, in the form of the WiFi radio map 210 of FIG. 2 . After storing as , when the value of the signal strength received from the moving object is transmitted, the location information corresponding to the corresponding signal strength value is called from the database and provided to the moving object. In this case, the signal transmitted from the signal transmitter may be, for example, any one of an RF signal, an LTE signal, and a 5G signal.

Hereinafter, 'first fingerprint(s)' refers to the fingerprint(s) collected by the first moving object, and the second fingerprint(s) refers to the fingerprint(s) collected by the second moving object. can mean In addition, the first fingerprint(s) and the second fingerprint(s) may be, for example, any one of an RF fingerprint, an LTE fingerprint, and a 5G fingerprint. A method of collecting a fingerprint by a moving object according to an exemplary embodiment will be described in more detail with reference to FIG. 2 below.

The management server matches the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the driving space ( 120 ). The management server according to an embodiment collects first fingerprints for a driving space in which the second moving object needs to travel, and matches the SLAM map of the driving space with the first fingerprints, thereby providing path information for future driving of the second moving object. and determining the coordinates on the SLAM map to enable autonomous driving of the second mobile body.

The management server learns a neural network for estimating coordinates on the SLAM map corresponding to the fingerprint based on the matched first coordinates and the first fingerprints ( 130 ). In this case, when a fingerprint is input, the neural network may be a neural network previously trained to output coordinates on the SLAM map corresponding to the input fingerprint. The neural network may include, for example, a deep neural network (DNN) or a recurrent neural network (RNN). The management server may set path information on the SLAM map in consideration of obstacles detected at a plurality of points based on the learning result of the neural network. The management server may obtain information on the shaded area in the driving space based on the first fingerprints, and determine path information corresponding to the shaded area based on the neural network.

The management server receives the second fingerprint from the second mobile body moving in the driving space based on the route information set on the SLAM map ( 140 ). In this case, the second mobile body may include a 5G network module and an IoT sensor receiving device.

The management server estimates the second coordinates on the SLAM map corresponding to the second fingerprint by using the neural network learned through step 130 (step 150).

The management server determines driving information of the second moving object based on the route information and the estimated second coordinates ( 160 ).

The management server transmits driving information (170).

According to an embodiment, the management server may receive information on the surrounding environment sensed by the second mobile body at a plurality of points in the driving space through the 5G network module. The management server may monitor the driving of the second moving object based on the information on the surrounding environment. The management server may correct the driving information according to the monitoring result, and transmit the corrected driving information to the second mobile unit through the 5G network module.

The management server according to an embodiment may be, for example, a distributed server or a cloud server. In this case, the management server may include, for example, a control interface such as fleet management software. According to an embodiment, the control interface may be installed separately from the management server, such as a factory management server or a distributed server of a workplace.

FIG. 2 is a diagram for describing a method of collecting a fingerprint by a moving object according to an exemplary embodiment. Referring to FIG. 2 , a database 210 created by first fingerprints collected from a first moving body at a plurality of points is shown.

The management server may measure the signal strength transmitted from a plurality of signal transmitters (eg, an access point, an IoT sensor, or an IoT device, etc.) located in the driving space of the moving object and store it in the database 210 . In this case, the signal transmitted from the signal transmitter may be, for example, an RF signal, an LTE signal, or a 5G signal.

Thereafter, when the measurement value (eg, the value of the signal strength received by the moving object) is transmitted online from the moving object in the driving space, the management server receives position information estimated in response to the corresponding signal strength value through pattern matching. It can be provided as a mobile object by calling it from the database. In this case, the pattern matching may be performed through, for example, a Nearest Neighboring (K-NN) classifier, a Support Vector Machine (SVM), or a Neural Network (NN).

In this case, the database 210 may include, for example, location information represented by (x, y) and signal strength information received from each signal transmitter at each location. A value (eg, -62, -88, -79, etc.) displayed as received signal information corresponding to the (x1, y1) position in the database 210 is, for example, (x1, y1) AP1 at the position, It may indicate the strength of a signal received from AP2 and AP3.

According to the embodiment, when the predetermined driving space is a limited place, such as a factory, a small 5G receiver installed for a smart factory in the factory, etc., as a signal transmitter, is used as a signal transmitter to collect fingerprints. may be

3 is a diagram for explaining a process of generating an alarm through image analysis in a management server according to an embodiment. Referring to FIG. 3 , the management server according to an exemplary embodiment may receive a driving environment image captured through at least one photographing device mounted on the second moving body in real time ( 310 ). In this case, the driving environment image captured by the second moving object may be transmitted to the management server at high speed by the 5G network as a real-time image of a Full-HD level, for example.

The management server may analyze the driving environment image using, for example, a deep neural network (DNN) classifier ( 320 ). In this case, the deep neural network (DNN) classifier may be trained in advance to classify other moving objects, people, animals, and other obstacles included in the driving environment image.

The management server may recognize whether a new object appears in the traveling space of the moving object through image analysis ( 330 ).

The management server may generate an alarm notifying whether an object appears in the driving space based on the analysis result (eg, appearance of a new object) ( 340 ). The management server may transmit the generated alarm to the second mobile body. Upon receiving the alarm, the second moving object may immediately stop driving according to the type of the alarm, or may express an alarm for an object that suddenly appears in the driving space through various alarm sounds, voice warnings, or flashing lights. The type of the alarm may have various forms, such as, for example, generation of an obstruction due to a failure of another moving object, or a sudden appearance of a person or animal.

The processes described above with reference to FIG. 3 may be performed by the management server, or may be performed by a control interface separate from the management server according to an embodiment.

In an embodiment, real-time performance in the 5G network may be guaranteed by performing the above-described process through mobile edge computing. Mobile edge computing is a technology that alleviates congestion in the mobile core network and creates new local services by applying distributed cloud computing technology to wireless base stations and deploying various services and caching contents close to user terminals. With mobile edge computing, traffic and service computing can be moved from a centralized cloud to the edge of the network and closer to the customer. At this time, instead of sending all data to the cloud for processing, data can be analyzed, processed, and stored at the edge of the distributed network. By collecting and processing data close to the customer in this way, latency can be reduced and real-time performance in high-bandwidth applications can be provided.

4 is a flowchart illustrating a method of operating a moving body according to an exemplary embodiment. Referring to FIG. 4 , a moving object according to an embodiment transmits fingerprints collected as it moves through a predetermined driving space ( 410 ). The mobile body may include, for example, a 5G network module and an IoT sensor receiver. The mobile body may transmit fingerprints collected as it moves through a predetermined driving space through the 5G network module. The moving object may transmit the fingerprints, for example, to the management server or to the control interface. In this case, the movable body may be, for example, the second movable body. The fingerprints may be second fingerprints collected from the second moving body.

The moving object receives the driving information determined based on the coordinates and route information on the SLAM map estimated in response to the fingerprints from the pre-trained neural network ( 420 ). The moving object may receive driving information from, for example, any one of a management server and a control interface.

The moving object determines a driving parameter based on the driving information ( 430 ). The driving parameters include, for example, a motor control parameter for controlling a front wheel steering motor and a rear wheel driving motor of the moving body, a gear ratio control parameter of the moving body, a wheel motion control parameter for controlling the wheel rotation speed and rotation angle of the moving body, etc. can do.

The moving object is driven according to the driving parameter ( 440 ).

According to an embodiment, the moving object may transmit information about the surrounding environment that is changed according to the driving of the moving object to the control interface or the management server. The moving object may receive the modified driving information from the control interface or the management server according to the monitoring result based on the changed information on the surrounding environment. The moving object may change a driving parameter according to the modified driving information, and may drive a driving unit of the moving object according to the changed driving parameter.

5 is a diagram for explaining components of a moving body and a control flow between the components according to an embodiment. Referring to FIG. 5 , a mobile body according to an embodiment may include a communication unit 510 , a sensor unit 520 , a control unit 530 , and a driving unit 540 .

The communication unit 510 may receive the SLAM map and driving information from the management server or the control interface and transmit it to the sensor unit 520 . The communication unit 510 may include, for example, a 5G network module and/or an IoT sensor receiving device.

The sensor unit 520 modifies the driving information received through the communication unit 510 according to the movement information of the moving object, based on the surrounding environment information sensed through various sensors included in the sensor unit 520 , or a new driving method. information can be generated.

Based on the driving information received from the sensor unit 520 , the controller 530 may determine a driving parameter for controlling the driving of wheels, gears, steering devices, and/or engines of the moving object. The controller 530 may correspond to, for example, the main controller shown in FIG. 6 .

The driving unit 540 may drive each driving engine of the moving body according to driving parameters. In addition, the driving unit 540 transmits, for example, wheel movement information of the moving object, such as wheel rotation speed and rotation angle, to the sensor unit 520 so that the sensor unit 520 corrects the driving information or receives new driving information. can be made to create

The moving object 500 according to an exemplary embodiment generates new driving information based on the surrounding environment information sensed by the sensor unit 520 and transmits it to the controller 530 and the driving unit 540 to continuously autonomously drive the moving object 500 . Alternatively, unmanned driving may be enabled.

6 is a perspective view of a movable body according to an exemplary embodiment. Referring to FIG. 6 , a mobile body 600 according to an embodiment may further include a 5G network module 610 and an IoT sensor receiving device 630 in addition to the general structure of the mobile body. Since the structure and operation of a general mobile body are well known, the 5G network module 610 and the IoT sensor receiving device 630 will be described below.

The 5G network module 610 may be mounted on the mobile unit 600 and transmit fingerprints collected as it moves through a predetermined driving space to a control interface and/or a management server. In addition, the 5G network module 610 may receive driving information determined based on the coordinates and route information on the SLAM map estimated in response to the fingerprints transmitted through the 5G network module 610 from the pre-trained neural network. .

The 5G network module 610 transmits fingerprints to, for example, the control interface and/or the management server through the RS-232 port 240 shown in FIG. 7 below, or from the control interface and/or the management server. You can receive driving information.

The IoT sensor receiving device 630 may receive surrounding environment information sensed through various IoT sensors. The IoT sensor receiving device 630 may receive surrounding environment information from various IoT sensors, for example, through the I/O port 730 shown in FIG. 7 below.

According to an embodiment, the management server estimates the coordinates on the SLAM map in real time in response to information received through the 5G network module 610 and the IoT sensor receiving device 630 provided in the mobile unit 600 to determine the driving route. By making this decision, it is possible to quickly respond to changes in driving information according to changes in the surrounding environment. In addition, by adding only the 5G network module 610 and the IoT sensor receiving device 630, it is possible to receive more accurate route information from the management server without replacing the existing moving body.

The mobile unit 600 according to an embodiment may transmit the driving situation to the management server or the control interface through the 5G network module 610 mounted thereon. The management server or the control interface may continuously check whether the mobile device 600 is traveling on a correct path based on the information received from the mobile device 600 . The management server or control interface may precisely control the driving of the mobile unit 600 by transmitting driving information in real time through the 5G network module 610 .

7 is a configuration diagram of a movable body according to an embodiment. Referring to FIG. 7 , a moving body 700 according to an embodiment includes an analog/digital converter 710 , a counter 720 , and an I/O port 730 . ), and an RS-232 port 740 .

The A/D converter 710 may adjust the speed of the moving object 700 by controlling the front wheel motor according to the driving parameter determined by the controller of the moving object 700 . In this case, the A/D converter 710 may adjust the driving speed of the front wheel motor of the movable body 700 by adjusting the value of the variable resistor.

The counter 720 may measure the driving state of the moving object 700 by measuring the rotation speed and/or the rotation angle of the rear wheel motor.

The A/D converter 710 and the counter 720 may be included in, for example, the controller 530 of the moving object described above with reference to FIG. 5 . Alternatively, according to an embodiment, the A/D converter 710 may be included in the control unit 530 of the above-described moving body, and the counter 720 may be included in the driving unit 540 of the above-described moving body.

The I/O port 730 may receive a detection result from an Ultra Sonic Sensor and/or an IoT sensor. The detection result of the ultra sonic sensor and/or the detection result of the IoT sensor may be used to detect an obstacle for parking the moving object 700 .

The RS-232 port 740 may receive an image of the surrounding environment from the CCD camera through 1:1 serial communication. The aforementioned 5G network module 610 may transmit and receive various information through the RS-232 port 740 . The mobile unit 700 may recognize the surrounding environment through image recognition based on the surrounding environment image received through the RS-232 port 740 and/or information transmitted to the 5G network module 610 .

8 is a block diagram of a management server according to an embodiment. Referring to FIG. 8 , the management server 800 according to an embodiment includes a processor 810 and a communication interface 830 . The base station 800 may further include a memory 850 . The processor 810 , the communication interface 830 , and the memory 850 may communicate with each other via the communication bus 805 .

The processor 810 collects first fingerprints measured along with sensing of the surrounding environment at a plurality of points where the surrounding environment is sensed to generate a SLAM map from the first moving body moving in a predetermined driving space. The processor 810 matches the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the driving space. The processor 810 learns a neural network for estimating coordinates on the SLAM map corresponding to the fingerprint based on the matched first coordinates and the first fingerprints. In addition, the processor 810 estimates the second coordinates on the SLAM map corresponding to the second fingerprint by using the neural network. The processor 810 determines driving information of the second moving object based on the path information and the estimated second coordinates.

The communication interface 830 receives the second fingerprint from the second mobile body moving in the driving space based on the route information set on the SLAM map.

The memory 850 includes path information set on the SLAM map generated by the processor 810 , the second coordinates of the SLAM map company corresponding to the second fingerprint estimated by the processor 810 , and the second determined by the processor 810 . At least one of driving information of the moving object may be stored.

Also, the processor 810 may perform at least one method described above with reference to FIGS. 1 to 7 or an algorithm corresponding to the at least one method. The processor 810 may be a hardware-implemented data processing device having a circuit having a physical structure for executing desired operations. For example, desired operations may include code or instructions included in a program. For example, a data processing device implemented as hardware includes a microprocessor, a central processing unit, a processor core, a multi-core processor, and a multiprocessor. , an Application-Specific Integrated Circuit (ASIC), and a Field Programmable Gate Array (FPGA).

The processor 810 may execute a program and control the management server 800 . The program code executed by the processor 810 may be stored in the memory 850 .

The memory 850 may store various pieces of information generated in the process of the above-described processor 810 . In addition, the memory 850 may store various data and programs. The memory 850 may include a volatile memory or a non-volatile memory. The memory 850 may include a mass storage medium such as a hard disk to store various data.

The method according to an embodiment may be implemented in the form of program instructions 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, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes 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 present invention, and vice versa.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited to the above embodiments, and various modifications and variations from these descriptions are provided by those skilled in the art to which the present invention pertains. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the following claims as well as the claims and equivalents.

800: management server
805: communication bus
810: processor
830: communication interface
850: memory

Claims (19)

collecting first fingerprints measured along with the sensing of the surrounding environment at a plurality of points where the surrounding environment is sensed to generate a SLAM map from a first moving body moving in a predetermined driving space;
matching the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in the driving space;
learning a neural network for estimating coordinates on the SLAM map corresponding to a fingerprint based on the matched first coordinates and first fingerprints;
receiving a second fingerprint from a second mobile body moving in the driving space based on route information set on the SLAM map;
estimating second coordinates on the SLAM map by pattern matching signal intensity values corresponding to the second fingerprint using the neural network;
determining driving information of the second moving object based on the path information and the estimated second coordinates;
receiving a driving environment image captured by at least one photographing device mounted on the second moving object in real time;
analyzing the driving environment image;
generating an alarm notifying whether an object appears in the driving space based on the analysis result; and
transmitting the driving information and the alarm to the second moving object;
Including, the method of operation of the management server. According to claim 1,
The step of collecting the first fingerprints
collecting the first fingerprints based on propagation characteristics of a signal transmitted from at least one base station in a 5G network
Including, the method of operation of the management server. According to claim 1,
setting path information on the SLAM map in consideration of obstacles detected at the plurality of points based on the learning result of the neural network
Further comprising, the method of operation of the management server. 4. The method of claim 3,
The step of determining the route information
obtaining information on a shaded area in the driving space based on the first fingerprints; and
determining path information corresponding to the shaded area based on the neural network
Including, the method of operation of the management server. According to claim 1,
The step of collecting the first fingerprints
generating the SLAM map corresponding to the location of the first mobile body by the SLAM (Simultaneous Localization And Mapping) technique based on a result of the first mobile body sensing the surrounding environment at the plurality of points;
Further comprising, the method of operation of the management server. According to claim 1,
The plurality of points
A method of operating a management server comprising at least one of at least one IoT sensor and at least one access point installed in the driving space. delete According to claim 1,
The step of analyzing the driving environment image is
Analyzing the driving environment image by a pre-trained deep neural network
Including, the method of operation of the management server. According to claim 1,
The step of generating the alarm is
recognizing whether an object appears in the driving space based on the analysis result; and
generating an alarm according to the recognition
Including, the method of operation of the management server. According to claim 1,
The second mobile includes a 5G network module,
receiving information on the surrounding environment sensed by the second mobile unit at the plurality of points through the 5G network module; and
monitoring the driving of the second mobile body based on the information on the surrounding environment;
Further comprising, the method of operation of the management server. 11. The method of claim 10,
correcting the driving information according to the monitoring result; and
transmitting the modified driving information to the second mobile unit through the 5G network module;
Further comprising, the method of operation of the management server. transmitting fingerprints collected as the user moves in a predetermined driving space;
receiving driving information determined based on coordinates and route information on a SLAM map estimated in response to the fingerprints from a pre-trained neural network;
determining a driving parameter based on the driving information;
performing driving according to the driving parameters;
transmitting a driving environment image captured by at least one photographing device in real time; and
Receiving an alarm generated according to whether an object appears in the driving space recognized based on the analysis result of the driving environment image
including,
Coordinates on the SLAM map are estimated by pattern matching signal strength values corresponding to the fingerprints using the neural network,
The path information is set in consideration of obstacles sensed at a plurality of points in the driving space based on a learning result of the neural network,
How the moving object works. 13. The method of claim 12,
Receiving the driving information includes:
Receiving the driving information from any one of a management server and a control interface
Including, a method of operating a moving object. 13. The method of claim 12,
The mobile body includes a 5G network module,
The step of transmitting the fingerprints
Transmitting the fingerprints collected by moving the predetermined driving space through the 5G network module
Including, a method of operating a moving object. 13. The method of claim 12,
transmitting information on a surrounding environment that is changed according to the driving of the moving object;
receiving driving information corrected according to a monitoring result based on the changed surrounding environment information; and
changing the driving parameter according to the modified driving information; and
driving the movable body according to the changed driving parameter
Further comprising, a method of operating a mobile body. 13. The method of claim 12,
The step of transmitting the driving environment image in real time is
photographing an image of the surrounding environment through at least one photographing device mounted on the moving object; and
Transmitting the surrounding environment image
Including, a method of operating a moving object. A computer program stored in a computer-readable recording medium in combination with hardware to execute the method of any one of claims 1 to 6 and 8 to 16. In order to generate a SLAM map from a first moving body moving in a predetermined driving space, first fingerprints measured with the sensing of the surrounding environment are collected at a plurality of points where the surrounding environment is sensed, and the driving is performed. Match the first fingerprints with a plurality of first coordinates on the SLAM map corresponding to the plurality of points in space, and based on the matched first coordinates and first fingerprints, a processor for learning a neural network estimating coordinates on the corresponding SLAM map; and
A communication interface for receiving a second fingerprint from a second mobile body moving in the driving space based on the route information set on the SLAM map and a driving environment image captured through at least one photographing device mounted on the second mobile body
including,
the processor
Using the neural network, by pattern matching signal intensity values corresponding to the second fingerprint, the second coordinates on the SLAM map are estimated, and based on the path information and the estimated second coordinates, the second determining the driving information of the moving object, analyzing the driving environment image, and generating an alarm indicating whether an object appears in the driving space based on the analysis result;
The communication interface is
transmitting the determined driving information and the generated alarm to the second moving object;
management server. It transmits fingerprints collected as it moves in a predetermined driving space, receives driving information determined based on coordinates and path information on a SLAM map estimated in response to the fingerprints in a pre-trained neural network, and at least one a communication interface that transmits a driving environment image captured by a photographing device in real time and receives an alarm generated according to whether an object appears in the driving space recognized based on an analysis result of the driving environment image; and
A processor that determines a driving parameter based on the driving information and performs driving according to the driving parameter
including,
Coordinates on the SLAM map are estimated by pattern matching signal strength values corresponding to the fingerprints using the neural network,
The path information is set in consideration of obstacles detected at a plurality of points in the driving space based on a learning result of the neural network,
moving body.

Images