KR102364008B1 - Accident record system and analysis method based on encrypted flight and imaging data of unmanned vehicle - Google Patents

Abstract

The accident record analysis system based on the encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention is an accident record analysis system that analyzes the cause of an accident using the accident record stored in the unmanned moving object, a navigation computer unit for outputting accident record data including one image data and operation data according to the flight of an unmanned moving object; an accident record storage unit for converting and storing the accident record data output from the operation computer unit into encrypted data; and an accident record analysis unit that decrypts the encrypted data to collect the accident record data, and analyzes the cause of the accident based on a parity check and a chi-square test using the collected accident record data.

Description

Accident record analysis system and method based on encrypted image and operation data of unmanned moving vehicle

The present invention relates to an accident record analysis system and method based on encrypted images and operation data of an unmanned moving object.

An unmanned vehicle is a vehicle that flies by remote control without a person on board or moves autonomously along a designated route. It is also used for

However, the above-described unmanned moving object may have various problems such as safety problems and security problems in the event of an accident because the movement characteristics are different from those of conventional devices.

Specifically, in the case of domestic products currently being sold, it is difficult to provide flight log data at the level of foreign manufacturers after an accident, and there is a limitation in that information leakage prevention measures to prevent sniffing and malicious attacks by external attackers are insufficient.

In addition, the unmanned moving object has a problem that there is no tracking and recovery function in case of an accident due to breakdown, climate change, external attack, etc. there is not

In other words, in the case of an unmanned moving vehicle accident, it is difficult to provide data for accurate accident cause analysis, and even if the wreckage caused by the accident is collected, there is a limit in using the data to check the record and then proceed with the post-analysis.

Embodiments of the present invention have been proposed to solve the above problems, and image data and navigation data stored in an unmanned moving object can be encrypted, stored and transmitted, and the cause of an accident can be analyzed using the encrypted data. It is intended to provide an accident record analysis system and method based on the image and operation data of an unmanned moving vehicle.

An accident record analysis system based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention is an operation that outputs accident record data including image data acquired by the unmanned object and operation data according to the flight of the unmanned object computer department; an accident record storage unit for converting and storing the accident record data output from the operation computer unit into encrypted data; and an accident record analysis unit that decrypts the accident record data by decrypting the encrypted data, and analyzes the cause of the accident based on a parity check and a chi-square test using the decrypted accident record data.

In addition, the navigation computer unit, a camera unit for acquiring image data by taking an image; GPS unit for acquiring GPS data on the flight position of the unmanned moving object; an inertial measurement unit for acquiring sensor data necessary for flight; a navigation control unit for acquiring steering data input from a user and posture data of an unmanned moving object corresponding to the steering data; and a first communication unit for transmitting the accident record data including the image data, the GPS data, the sensor data, the steering data, and the posture data.

In addition, the accident record storage unit, a second communication unit for receiving the accident record data from the first communication unit; an encryption unit that encrypts the accident record data with a certificate previously shared between the key management server and the unmanned moving object and converts it into encrypted data in a predetermined format; and a memory unit for compressing and storing the encrypted data.

In addition, the accident record storage unit extracts GPS data from the accident record data received by the second communication unit, and transmits the extracted GPS data using low-power RF data communication in a sleep & wake method. It may further include an RF communication unit.

In addition, the accident record analysis unit, a decryption unit for receiving the encryption key in a two-factor-based encryption key injection method from a key management server that stores a preset encryption key to decrypt the encrypted data to decrypt the accident record data; a dynamic characteristic analysis unit that extracts sensor data from the accident record data and acquires dynamic characteristic data that is position and posture data of an unmanned moving object according to a control input through dynamic modeling based on the extracted sensor data; a cause analysis unit for determining an accident cause based on a parity check and a chi-square test using the accident record data and the dynamic characteristic data; an accident reproduction unit for generating post-processing data through a forward-backward algorithm using the accident record data decoded by the decoding unit; and an output unit provided to display the post-processing data generated by the accident reproduction unit.

In addition, the cause analysis unit, the synchronization unit for synchronizing data; a first analysis unit that analyzes the cause of an accident through a parity check and a chi-square test; and a second analysis unit that determines the cause of the accident according to the result of the analysis of the cause of the accident.

In addition, the synchronization unit extracts GPS data, sensor data, steering data and attitude data from the accident record data, and synchronizes the extracted steering data and attitude data with the dynamic characteristic data to generate first fusion data, and The second fusion data is generated by synchronizing the steering data and the attitude data with the sensor data, and the third fusion data is generated by synchronizing the extracted steering data and the attitude data with the GPS data.

Also, the first analyzer may calculate parity values corresponding to the first fusion data, the second fusion data, and the third fusion data through a parity check.

Also, the first analyzer calculates a parity change amount corresponding to each calculated parity value, and calculates a chi-square value through a chi-square test using the parity value and the parity change amount.

In addition, the second analyzer obtains a p value, which is an index for a significance level according to the result of the chi-square test, and determines that the chi-square value is greater than or equal to the p value as a failure state, and the chi-square value When this p value is smaller than the p value, it can be determined as a normal state.

In addition, when the first chi-square value corresponding to the first fusion data is greater than or equal to the obtained p value, the second analysis unit determines that the actuator failure of the unmanned moving object is the cause of the accident, and corresponds to the second fusion data If the second chi-square value is greater than or equal to the obtained p value, the sensor failure of the unmanned vehicle is determined as the cause of the accident. If the third chi-square value corresponding to the third fusion data is greater than or equal to the obtained p value, the unmanned vehicle Algorithm failure is determined as the cause of the accident.

The accident record analysis method based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention comprises the steps of outputting accident record data including image data acquired by the unmanned vehicle and operation data according to the flight of the unmanned vehicle ; converting the output accident record data into encrypted data and storing the converted data; and decrypting the accident record data by decrypting the encrypted data, and analyzing the cause of the accident based on a parity check and a chi-square test using the decrypted accident record data.

In addition, the step of outputting the accident record data may include: acquiring image data by photographing an image; acquiring GPS data on the flight position of the unmanned mobile vehicle; acquiring sensor data necessary for flight; acquiring steering data input from a user and posture data of an unmanned moving object corresponding to the steering data; and transmitting the accident record data including the image data, the GPS data, the sensor data, the steering data, and the attitude data.

In addition, the step of converting and storing the output accident record data into encrypted data includes the steps of: receiving the accident record data; converting the accident record data into encrypted data in a predetermined format by encrypting the accident record data with a certificate previously shared between the key management server and the unmanned mobile body; and compressing and storing the encrypted data.

In addition, analyzing the cause of the accident based on the parity check and the chi-square test includes receiving an encryption key from a key management server that stores a preset encryption key in a two-factor-based encryption key injection method and decrypting the encrypted data. decoding the accident record data; extracting sensor data from the accident record data, and acquiring dynamic characteristics data, which are position and posture data of an unmanned moving object according to a control input, through dynamic modeling based on the extracted sensor data; determining an accident cause based on a parity check and a chi-square test using the accident record data and the dynamic characteristic data; generating post-processing data through a forward-backward algorithm using the decrypted accident record data; and displaying the generated post-processing data.

In addition, the step of determining the cause of the accident may include extracting GPS data, sensor data, steering data and attitude data from the accident record data, and synchronizing the extracted steering data and attitude data with the dynamic characteristic data to obtain the first fusion data. generating second fusion data by synchronizing the extracted steering data and attitude data with the sensor data, and generating third fusion data by synchronizing the extracted steering data and attitude data with the GPS data; calculating parity values corresponding to the first fusion data, the second fusion data, and the third fusion data through a parity check; calculating a parity variation corresponding to each calculated parity value, and calculating a chi-square value through a chi-square test using the parity value and the parity variation; and obtaining a p value, which is an index for a significance level according to the result of the chi-square test, and determining a failure state if the chi-square value is greater than or equal to the p value, and if the chi-square value is less than the p value and determining the normal state.

The accident record analysis system and method based on the encrypted image and operation data of the unmanned vehicle according to the embodiments of the present invention can encrypt, store and transmit image data and operation data stored in the unmanned vehicle, and use the encrypted data to By analyzing the cause of the accident, data having confidentiality and integrity can be utilized, thereby ensuring high data reliability.

1 is a schematic diagram for explaining an accident record analysis system based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention.
2 is a detailed view for explaining an accident record analysis system based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention.
3 is a view for explaining an accident record analysis unit according to an embodiment of the present invention.
4 and 5 are diagrams illustrating a process of analyzing an accident cause through a parity check and a chi-square test according to an embodiment of the present invention.
6 is a view for explaining an accident reproduction unit according to an embodiment of the present invention.
7 is a diagram illustrating a process of receiving an encryption key from a key management server according to an embodiment of the present invention.
8 is a flowchart illustrating an accident record analysis method based on an encrypted image of an unmanned moving object and operation data according to an embodiment of the present invention.

Other advantages and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only this embodiment serves to complete the disclosure of the present invention, and to obtain common knowledge in the technical field to which the present invention pertains. It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims.

Even if not defined, all terms (including technical or scientific terms) used herein have the same meaning as commonly accepted by common technology in the prior art to which this invention belongs. Terms defined by general dictionaries may be construed as having the same meaning as in the related description and/or in the text of the present application, and shall not be interpreted conceptually or excessively formally, even if not expressly defined herein. won't

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, 'comprise' and/or the various conjugations of this verb, eg, 'comprising', 'comprising', 'comprising', 'comprising', etc., refer to the stated composition, ingredient, component, A step, operation and/or element does not exclude the presence or addition of one or more other compositions, components, components, steps, operations and/or elements. As used herein, the term 'and/or' refers to each of the listed components or various combinations thereof.

Meanwhile, terms such as '~ unit', '~ group', '~ block', and '~ module' used throughout this specification may mean a unit that processes at least one function or operation. For example, it can mean software, a hardware component such as an FPGA or an ASIC. However, '~ part', '~ group', '~ block', and '~ module' are not meant to be limited to software or hardware. '~ unit', '~ group', '~ block', and '~ module' may be configured to reside in an addressable storage medium or to regenerate one or more processors.

Accordingly, as an example, '~ part', '~ group', '~ block', and '~ module' are components such as software components, object-oriented software components, class components, and task components. fields, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and include variables. The functions provided within the components and '~part', '~gi', '~block', and '~module' are smaller than the number of components and '~bu', '~gi', '~block' ', '~modules' or may be further separated into additional components and '~parts', '~gi', '~blocks', and '~modules'.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings.

In addition, in describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted.

1 is a schematic diagram for explaining an accident record analysis system based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention. 2 is a detailed view for explaining an accident record analysis system based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention, and FIG. 6 is a diagram illustrating an accident reproduction unit according to an embodiment of the present invention. It is a drawing for

Referring to FIG. 1 , an accident record analysis system 1 based on an encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention includes a navigation computer unit 100, an accident record storage unit 200, and an accident record. and an analysis unit 300 .

The operation computer unit 100 may output accident record data including image data acquired by the unmanned moving object and flight data according to the flight of the unmanned moving object.

In addition, the navigation computer unit 100 is a device provided inside the unmanned moving body in a typical unmanned moving body set composed of an unmanned moving body and a remote controller for manipulating it.

Specifically, the operation computer unit 100 includes a communication unit that communicates with the control device so that the unmanned movable body can fly or move according to the control signal of the control device, and the unmanned movable body in addition to navigation data such as latitude, longitude, altitude, speed and posture of the unmanned movable body. It is possible to collect navigation data including weather data, which are factors that affect the flight or movement of an unmanned vehicle, such as wind direction, wind speed, rainfall, humidity, temperature, and illuminance of the atmosphere in contact with it.

In addition, the navigation computer unit 100 obtains data on the latitude, longitude, altitude, etc. of the vehicle from the built-in GPS, atmospheric pressure data according to the altitude with a barometer, and wind direction/wind speed data with a wind direction/wind speed sensor. Various devices such as and method can be used to measure navigation data.

The accident record storage unit 200 converts the accident record data output from the operation computer unit 100 into encrypted data and stores it.

In addition, the accident record storage unit 200 may be a normal black box, records the flight or movement of the unmanned moving object when using the unmanned moving object, safely stores the accident record in case of an accident, and through the stored data of the unmanned moving object Enable users to safely and usefully determine the cause of an accident.

The accident record analysis unit 300 decrypts the encrypted data stored in the accident record storage unit 200 to decrypt the accident record data, and uses the decrypted accident record data to cause an accident based on a parity check and a chi-square test can be analyzed.

In addition, the accident record analysis unit 300 can determine the failure state of the unmanned moving object by using an accident cause analysis algorithm through step-by-step accident classification, and more specifically, the image that is the accident record data stored in the accident record storage unit 200 . It analyzes the cause of an accident by detecting failures in actuators, sensors, and multiple algorithms required to operate an unmanned vehicle by using data and operation data.

In addition, the accident record storage unit 200 and the accident record analysis unit 300 are connected to the key management server 400 through a communication secure channel, and the key management server 400 is a secret key flow that complies with international standards. A function is applied, and an encryption key can be generated and stored.

In addition, the key management server 400 may include a hardware security module (Hardware Security Module, HSM) for strengthening physical security.

As shown in FIG. 2 , the navigation computer unit 100 includes a camera unit 110 that acquires image data by photographing an image, a GPS unit 120 that acquires GPS data for the flight position of an unmanned moving object, and a flight The inertia measurement unit 130 for acquiring necessary sensor data, the operation control unit 140 for acquiring the steering data input from the user and the posture data of the unmanned moving object corresponding to the steering data, and the image data, the GPS data, and the sensor and a first communication unit 150 for transmitting the accident record data including data, the steering data, and the attitude data.

The camera unit 110 is an optical conversion device that outputs image data projected around an unmanned moving object, may be a camera using a CCD sensor or a CMOS sensor, and converts an optical signal into an electrical signal and outputs it.

The GPS unit 120 may be configured with a sensor capable of detecting the location and flight direction of the unmanned moving object, and may check the operation route of the unmanned moving object.

The inertia measurement unit 130 is a device that detects the operation according to the flight or movement of the unmanned moving object and the environment around the unmanned moving object. It may be configured to include an acceleration sensor, an altitude sensor capable of detecting the height of an unmanned moving object, and the like.

That is, the inertial measurement unit 130 may extract sensor data such as pressure altitude, gas posture, indicated air speed, normal acceleration, longitudinal acceleration, and lateral acceleration through a gyro sensor, an acceleration sensor, an altitude sensor, and the like.

The operation control unit 140 may be configured to control the overall operation of the unmanned moving object, and may control the operation of a driving device such as a motor according to a user's control signal received through the first communication unit 150 to allow the unmanned moving object to be moved freely. It can be controlled to fly or move in a posture.

Also, the operation control unit 140 may store a log of a signal controlled by the user.

The first communication unit 150 is configured to transmit data detected, extracted, and acquired from each device constituting the unmanned mobile body to the accident record storage unit 200, and image data and Flight data can be output and transmitted.

The accident record storage unit 200 encrypts the second communication unit 210 for receiving the accident record data from the first communication unit 150, and a certificate previously shared between the key management server 400 and the unmanned mobile body. and an encryption unit 220 that converts the encrypted data into a predetermined format and a memory unit 230 that compresses and stores the encrypted data.

The second communication unit 210 is configured to receive the accident record data output from the operation computer unit 100 and provide it to the encryption unit 220 , and may use RF data communication or LTE data communication.

The encryption unit 220 may use a conventional encryption algorithm, and may use a password verified based on the encryption module authentication system (KCMVP).

In addition, the encryption unit 220 stores the user information of the unmanned moving object and the ID of the unmanned moving object, and also stores a certificate that combines the user information of the unmanned moving object and the ID of the unmanned moving object, and stores the certificate of the key management server 400 and the mutual If authentication is successful, the received accident record data is converted into encrypted data.

The memory unit 230 stores the encrypted data in an extruded file format, and is linked with the encryption unit 220 to store the encrypted data in a TAT (Time Allocation Table) method.

In addition, since the memory unit 230 uses the TAT method, it is unnecessary to search for a usable storage space block and a confirmation procedure for connected data is unnecessary, so that it is possible to quickly write and search data in a sequential data management method.

In addition, after the amount of data to be recorded is determined, a space corresponding thereto is automatically allocated, and the space to be used is immediately changed and allocated according to a specific condition, so that the size of the data storage space can be automatically changed.

In addition, referring to FIG. 2 , the accident record storage unit 200 of the accident record analysis system 1 based on the encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention includes the second communication unit 210 It further includes an RF communication unit 240 for extracting GPS data from the received accident record data, and transmitting the extracted GPS data using low-power RF data communication in a sleep & wake method (Sleep & Wake).

Accordingly, the RF communication unit 240 can transmit the location information of the unmanned moving object in the event of an accident without restrictions in place and time, and can be linked with the receiving tracer device and application.

In this case, the receiving tracer device may be a USB type device that can be installed and used in smartphones and laptops.

In addition, the RF communication unit 240 may receive operating power from an auxiliary battery provided in the RF communication unit 240 itself when power is cut off or insufficient during an accident.

2 and 6, the accident record analysis unit 300 of the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention stores a preset encryption key A decryption unit 310 that decrypts the accident record data by receiving an encryption key from a key management server 400 using a Two Factor-based encryption key injection method to decrypt the encrypted data, extracting sensor data from the accident record data, and , a dynamic characteristic analysis unit 320 that acquires dynamic characteristic data that is the position and position data of an unmanned moving object according to a control input through dynamic modeling based on the extracted sensor data, parity check and The cause analysis unit 330 for judging based on the chi-square test, the accident reproduction unit 340 for generating post-processing data through a forward-backward algorithm using the accident record data decoded by the decoding unit 310, and the and an output unit 350 provided to display the post-processing data generated by the accident reproduction unit 340 .

The decryption unit 310 includes a terminal and a built-in memory for accessing the key management server 400, and when the key management server 400 accesses the terminal, the encryption key for the certificate from the key management server 400 can be obtained to access encrypted data.

The dynamic characteristic analysis unit 320 generates a virtual dynamics modeling for an unmanned moving object, but as a rotor-based dynamics equation, the virtual dynamics modeling is generated using the relational expressions of potential energy, translational kinetic energy, and rotational kinetic energy, and the motor By adding redundancy based on the virtual dynamics modeling of the structure, it is possible to create a virtual dynamics modeling for the unmanned moving body of the quad-rotor structure.

In addition, the dynamic characteristic analysis unit 320 may output the dynamic characteristic data of the unmanned moving object as output data based on the virtual dynamics modeling by using the sensor data of the accident record data as input data.

The cause analysis unit 330 determines the cause of the accident through the analysis of the cause of the accident through the stepwise classification of the accident and the result according to the analysis of the cause of the accident.

Also, the cause analysis unit 330 may be a distributed Kalman filter in which a probability-based filter such as a parity check and a chi-square test is fused.

First, the cause analysis unit 330 detects a failure of components constituting an unmanned mobile body using a parity check, sets a parity vector to be sensitive to failure, and detects a failure by detecting a change in a parity value.

That is, after the cause analysis unit 330 synchronizes the data units of one or more selected data among the sensor data, posture data, steering data, and GPS data included in the accident record data and the dynamic characteristic data output from the dynamic characteristic analysis unit 320 , , a parity check may be performed to obtain a parity value.

In this case, the parity value of the data related to the failure is the largest, and in the case of the data not related to the failure, the parity value may appear as '0'.

In addition, the cause analysis unit 330 may perform an accident cause analysis through a chi-square test using the parity value obtained by the parity check, but obtain a chi-square value obtained by the chi-square test.

In addition, the cause analysis unit 330 compares the obtained chi-square value with the p-value, which is an indicator of how significant the result of the chi-square test is, detects failures of components constituting the unmanned vehicle, and determines the cause of the accident do.

The accident reproduction unit 340 is a normal forward-backward smoother, and acquires accurate post-processing data. The forward-backward smoother may calculate a hidden state variable through a forward-backward algorithm.

As shown in FIG. 6 , the accident reproduction unit 340 may include a forward filter and a backward filter, and accident record data including image data, sensor data, steering data, posture data, and GPS data of an unmanned moving object; Based on the dynamic characteristics data of the unmanned moving vehicle, post-processing data for accident reproduction can be generated and output.

Thereafter, the output unit 350 may display an accident reproduction image based on the post-processing data and visualize it to the user.

3 is a diagram for explaining an accident record analysis unit according to an embodiment of the present invention, and FIGS. 4 and 5 are a parity check and a chi-square test-based Kalman filter according to an embodiment of the present invention to analyze the cause of an accident It is a diagram showing the process.

3, 4 and 5 , the cause analysis unit 330 includes a synchronization unit 331 for synchronizing data, and a first analysis unit 332 for performing accident cause analysis through a parity check and a chi-square test. , it may include a second analysis unit 333 that determines the cause of the accident according to the result of the analysis of the cause of the accident.

The synchronization unit 331 generates fusion data by fusing one or more selected data among the sensor data, posture data, steering data, and GPS data included in the accident record data of the unmanned vehicle and the dynamic characteristic data output from the dynamic characteristic analysis unit 320 . can do.

Specifically, as shown in FIG. 4 , the synchronization unit 331 extracts GPS data, sensor data, steering data, and attitude data from the accident record data, and synchronizes the extracted steering data and attitude data with the dynamic characteristics data. to generate the first fusion data.

Also, the synchronization unit 331 extracts steering data, attitude data, and sensor data from among the accident record data, and generates second fusion data by synchronizing the extracted steering data, attitude data, and sensor data.

Also, the synchronization unit 331 may extract steering data, attitude data, and GPS data from among the accident record data, and may generate third fusion data by synchronizing the extracted steering data, attitude data, and GPS data.

The first analyzer 332 calculates parity values corresponding to the first fusion data, the second fusion data, and the third fusion data through a parity check.

For example, as shown in FIGS. 4 and 5 , the first analyzer 332 may include N local filters (N is a positive integer), and the first fusion data The first parity value may be calculated through the first local filter using . Similarly, a second parity value is calculated through a second local filter using the second fusion data, and a third parity value is calculated through a third local filter using the third fusion data.

Thereafter, the first analyzer 332 may calculate a parity variation corresponding to each calculated parity value, and calculate a chi-square value through a chi-square test using the parity value and the parity variation.

For example, the first analyzer 332 calculates a first parity variation corresponding to the first parity value through the first local filter, and uses the calculated first parity value and the first parity variation to perform a chi-square test , and a first chi-square value is calculated as the result.

Similarly, the first analyzer 332 calculates a second parity variation corresponding to the second parity value through the second local filter, and performs a chi-square test using the calculated second parity value and the second parity variation. and a second chi-square value is calculated as the result.

Also, the first analyzer 332 calculates a third parity variation corresponding to the third parity value through the third local filter, and performs a chi-square test using the calculated third parity value and the third parity variation. and a third chi-square value can be calculated from the result.

The second analysis unit 333 obtains a p value, which is an index for a significance level according to the result of the chi-square test, and determines that the chi-square value is greater than or equal to the p value as a failure state, and the chi-square value If it is smaller than the p value, it is determined as a normal state.

For example, the second analysis unit 333 obtains a first p value determined by the significance level of the chi-square distribution table obtained after performing the chi-square test through the first local filter, and the first chi-square value is the second If it is greater than or less than the 1 p value, it is determined that the first fusion data contains data of the configuration related to the failure, and if the first chi-square value is less than the first p value, the configuration related to the failure is included in the first fusion data. data is not included.

Similarly, the second analyzer 333 obtains the second p value through the second local filter, and when the second chi-square value is greater than or smaller than the second p value, the second fusion data contains It is determined that data is included, and if the second chi-square value is less than the second p value, it is determined that the second fusion data does not include data of a configuration related to a failure.

In addition, the second analysis unit 333 obtains a third p value through the third local filter, and when the third chi-square value is greater than or smaller than the third p value, the third fusion data contains a configuration related to a failure. It may be determined that data is included, and if the second chi-square value is less than the second p value, it may be determined that the second fusion data does not include data of a configuration related to a failure.

More specifically, when the first chi-square value corresponding to the first fusion data is greater than or equal to the obtained p value, the second analysis unit 333 determines that the actuator failure of the unmanned moving object is the cause of the accident, and the second fusion If the second chi-square value corresponding to the data is greater than or equal to the obtained p value, the sensor failure of the unmanned moving object is determined as the cause of the accident, and the third chi-square value corresponding to the third fusion data is greater than the obtained p value , it is determined that the algorithm failure of the unmanned vehicle is the cause of the accident.

7 is a diagram illustrating a process of receiving an encryption key from a key management server according to an embodiment of the present invention.

The accident record analysis system 1 based on the encrypted image and operation data of an unmanned moving object according to an embodiment of the present invention may share a certificate used to encrypt the accident record data with the key management server 400 . However, the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object does not have a decryption key for the certificate, so even if an external malicious attacker attacks the system and acquires administrator rights, the accident record stored after being encrypted with the certificate The recorded data cannot be decrypted.

In addition, the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object receives an encryption key from the key management server 400 to decrypt the accident record data, and the key management server 400 provides the security key The generation unit generates an encryption key capable of decrypting the certificate, stores the encryption key in the security key storage unit, and the stored encryption key is transmitted through the main communication channel formed between the security key management unit and the security key injector 500 . can be transmitted.

The security key injector 500 may be a USB security token including a wireless communication module (not shown) capable of wireless communication, and may receive and store an encryption key from the key management server 400 through a main communication channel. .

After that, the security key injector 500 uses the two-factor-based encryption key injection method that simultaneously utilizes the main communication channel using the wireless communication module and the additional communication channel using USB to encrypt the image and operation of the unmanned mobile body. It is transmitted to the accident record analysis system (1) based on the data.

Accordingly, the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention is any of the components for the flight or movement of the unmanned moving object in the event of an accident of the unmanned moving object It is possible to check whether a malfunction has occurred in the device, so it is possible to analyze the cause of the accident.

In addition, the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention reproduces the accident pattern of the unmanned moving object by using the accident record data and dynamic characteristic data of the unmanned moving object. can be visualized.

In addition, the accident record analysis system 1 based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention can encrypt and store the accident record data of the unmanned moving object even if the unmanned moving object falls due to an emergency accident, Secondary damage can be prevented by a malicious attacker rather than the user himself.

8 is a flowchart illustrating an accident record analysis method based on an encrypted image of an unmanned moving object and operation data according to an embodiment of the present invention.

The accident record analysis method based on the encrypted image and operation data of the unmanned moving object according to an embodiment of the present invention comprises the steps of outputting accident record data including image data acquired by the unmanned vehicle and operation data according to the flight of the unmanned vehicle (S100), converting the output accident record data into encrypted data and storing it (S200) and decrypting the encrypted data to decrypt the accident record data, and using the decrypted accident record data to perform a parity check and chi-square test It includes a step (S300) of analyzing the cause of the accident based on the

The step S100 includes: acquiring image data by photographing an image (S110), acquiring GPS data for a flight position of an unmanned moving object (S120), acquiring sensor data necessary for flight (S130), a user Step (S140) of obtaining the inputted steering data and the attitude data of the unmanned moving object corresponding to the steering data, and the accident record including the image data, the GPS data, the sensor data, the steering data, and the attitude data It may include transmitting data (S150).

The step S200 includes the steps of receiving the accident record data (S210), encrypting the accident record data with a certificate previously shared between the key management server and the unmanned mobile body, and converting the accident record data into encrypted data in a certain format (S220) and the and compressing and storing the encrypted data (S230).

In step S300, receiving an encryption key from a key management server that stores a preset encryption key in a two-factor-based encryption key injection method, decrypting the encrypted data to decrypt the accident record data (S310), the accident record Extracting sensor data from the data, and obtaining dynamic characteristics data, which is the position and posture data of an unmanned moving object according to a control input, through dynamic modeling based on the extracted sensor data (S320), the accident record data and the dynamic characteristics data Step (S330) of determining the cause of the accident based on the parity check and chi-square test using the It may include displaying the processing data (S350).

In step S330, GPS data, sensor data, steering data, and attitude data are extracted from the accident record data, and first fusion data is generated by synchronizing the extracted steering data and attitude data with the dynamic characteristic data, and the extracted steering Synchronizing data and attitude data with the sensor data to generate second fusion data, and synchronizing the extracted steering data and attitude data with the GPS data to generate third fusion data (S331); Calculating each parity value corresponding to the first fusion data, the second fusion data, and the third fusion data (S332), calculating a parity variation corresponding to each calculated parity value, the parity value and Calculating a chi-square value through a chi-square test using the parity variation ( S333 ) and obtaining a p-value that is an index for a significance level according to the result of the chi-square test, wherein the chi-square value is the p value The method further includes a step (S334) of determining a failure state if it is greater than or equal to , and determining a normal state if the chi-square value is less than the p value (S334).

The above detailed description is illustrative of the present invention.

In addition, the above description shows and describes preferred embodiments of the present invention, and the present invention can be used in various other combinations, modifications, and environments. That is, changes or modifications are possible within the scope of the concept of the invention disclosed herein, the scope equivalent to the written disclosure, and/or within the scope of skill or knowledge in the art. The written embodiment describes the best state for implementing the technical idea of the present invention, and various changes required in specific application fields and uses of the present invention are possible. Therefore, the detailed description of the present invention is not intended to limit the present invention to the disclosed embodiments. Also, the appended claims should be construed as including other embodiments.

100: operation computer unit
200: accident record storage unit
300: accident record analysis unit
400: key management server
500: security key injector

Claims (16)

In an accident record analysis system that analyzes the cause of an accident using the accident record stored in an unmanned moving object,
a navigation computer unit for outputting accident record data including image data acquired by the unmanned vehicle and operation data according to the flight of the unmanned vehicle;
an accident record storage unit for converting and storing the accident record data output from the operation computer unit into encrypted data; and
An accident record analysis unit that decrypts the accident record data by decrypting the encrypted data, and analyzes the cause of the accident based on a parity check and a chi-square test using the decrypted accident record data,
The operation computer unit:
a camera unit to acquire image data by photographing an image;
GPS unit for acquiring GPS data on the flight position of the unmanned moving object;
an inertial measurement unit for acquiring sensor data necessary for flight;
a navigation control unit for acquiring steering data input from a user and posture data of an unmanned moving object corresponding to the steering data; and
A first communication unit for transmitting the accident record data including the image data, the GPS data, the sensor data, the steering data, and the posture data,
The accident record analysis unit:
a decryption unit that receives an encryption key from a key management server storing a preset encryption key in a two-factor-based encryption key injection method, decrypts the encrypted data, and decrypts the accident record data;
a dynamic characteristic analysis unit that extracts sensor data from the accident record data and acquires dynamic characteristic data that is position and posture data of an unmanned moving object according to a control input through dynamic modeling based on the extracted sensor data;
a cause analysis unit for determining an accident cause based on a parity check and a chi-square test using the accident record data and the dynamic characteristic data;
an accident reproduction unit for generating post-processing data through a forward-backward algorithm using the accident record data decoded by the decoding unit; and
Accident record analysis system including an output unit provided to display the post-processing data generated by the accident reproduction unit.
delete The method of claim 1,
The accident record storage unit,
a second communication unit for receiving the accident record data from the first communication unit;
an encryption unit that encrypts the accident record data with a certificate previously shared between the key management server and the unmanned moving object and converts it into encrypted data in a predetermined format; and
Accident record analysis system comprising a memory unit for compressing and storing the encrypted data.
4. The method of claim 3,
The accident record storage unit,
Accident record further comprising an RF communication unit that extracts GPS data from the accident record data received by the second communication unit and transmits the extracted GPS data using low-power RF data communication in a sleep & wake method (Sleep & Wake) analysis system.
delete 5. The method of claim 4,
The cause analysis unit,
a synchronization unit for synchronizing data;
a first analysis unit that analyzes the cause of an accident through a parity check and a chi-square test; and
Accident record analysis system including a second analysis unit for determining the cause of the accident according to the result of the analysis of the cause of the accident.
7. The method of claim 6,
The synchronization unit,
GPS data, sensor data, steering data, and attitude data are extracted from the accident record data, and first fusion data is generated by synchronizing the extracted steering data and attitude data with the dynamic characteristic data, and the extracted steering data and attitude data are An accident record analysis system for generating second fusion data by synchronizing the sensor data, and generating third fusion data by synchronizing the GPS data with the extracted steering data and posture data.
8. The method of claim 7,
The first analysis unit,
An accident record analysis system for calculating parity values corresponding to the first fusion data, the second fusion data, and the third fusion data through a parity check.
9. The method of claim 8,
The first analysis unit,
An accident record analysis system for calculating a parity variation corresponding to each calculated parity value, and calculating a chi-square value through a chi-square test using the parity value and the parity variation.
10. The method of claim 9,
The second analysis unit,
A p value, which is an index for the significance level according to the result of the chi-square test, is obtained, and when the chi-square value is greater than or equal to the p value, it is determined as a failure state, and when the chi-square value is less than the p value, it is normal Accident record analysis system that judges the state.
11. The method of claim 10,
The second analysis unit,
If the first chi-square value corresponding to the first fusion data is greater than or equal to the obtained p value, actuator failure of the unmanned moving vehicle is determined as the cause of the accident, and the second chi-square value corresponding to the second fusion data is obtained If it is greater than or equal to the p value, the sensor failure of the unmanned vehicle is determined as the cause of the accident, and if the third chi-square value corresponding to the third fusion data is greater than or equal to the obtained p value, the algorithm failure of the unmanned vehicle is determined as the cause of the accident Accident record analysis system.
In the accident record analysis method for analyzing the cause of an accident using the accident record stored in an unmanned moving object,
outputting accident record data including image data acquired by the unmanned moving object and flight data according to the flight of the unmanned moving object;
converting the output accident record data into encrypted data and storing the converted data; and
Decrypting the encrypted data to decrypt the accident record data, and analyzing the cause of the accident based on a parity check and a chi-square test using the decrypted accident record data,
The step of outputting the accident record data is:
acquiring image data by photographing an image;
acquiring GPS data on the flight position of the unmanned mobile vehicle;
acquiring sensor data necessary for flight;
acquiring steering data input from a user and posture data of an unmanned moving object corresponding to the steering data; and
Transmitting the accident record data including the image data, the GPS data, the sensor data, the steering data, and the attitude data,
The step of analyzing the cause of the accident based on the parity check and the chi-square test is:
receiving an encryption key from a key management server storing a preset encryption key in a two-factor-based encryption key injection method, decrypting the encrypted data, and decrypting the accident record data;
extracting sensor data from the accident record data, and acquiring dynamic characteristics data, which are position and posture data of an unmanned moving object according to a control input, through dynamic modeling based on the extracted sensor data;
determining an accident cause based on a parity check and a chi-square test using the accident record data and the dynamic characteristic data;
generating post-processing data through a forward-backward algorithm using the decrypted accident record data; and
Accident record analysis method comprising the step of displaying the generated post-processing data.
delete 13. The method of claim 12,
The step of converting the output accident record data into encrypted data and storing it,
receiving the accident record data;
converting the accident record data into encrypted data in a predetermined format by encrypting the accident record data with a certificate previously shared between the key management server and the unmanned mobile body; and
Accident record analysis method comprising the step of compressing and storing the encrypted data.
delete 15. The method of claim 14,
Determining the cause of the accident is
GPS data, sensor data, steering data, and attitude data are extracted from the accident record data, and first fusion data is generated by synchronizing the extracted steering data and attitude data with the dynamic characteristic data, and the extracted steering data and attitude data are generating second fusion data by synchronizing the sensor data, and generating third fusion data by synchronizing the GPS data with the extracted steering data and posture data;
calculating parity values corresponding to the first fusion data, the second fusion data, and the third fusion data through a parity check;
calculating a parity variation corresponding to each calculated parity value, and calculating a chi-square value through a chi-square test using the parity value and the parity variation; and
A p value, which is an index for the significance level according to the result of the chi-square test, is obtained, and when the chi-square value is greater than or equal to the p value, it is determined as a failure state, and when the chi-square value is less than the p value, it is normal Accident record analysis method comprising the step of judging the state.

Images