RU2814461C1 - Method for tracking driver actions during examination for driving license - Google Patents

Abstract

FIELD: vehicle control.

SUBSTANCE: invention relates to a device, a system and a method for monitoring the actions of a driver during an examination for the right to drive vehicles. Method comprises stages, at which: before beginning of examination, loading area map and pre-marked route of examination; starting the examination process; in process conducting examinations receive data from peripheral components of on-board device and evaluating the correctness of the driver's exercises, tasks or maneuvers on the examination route based on processing data received from the peripheral modules and from the navigation module of the on-board device.

EFFECT: complete control of vehicle control within a predetermined list of exercises, maneuvers and checks, high reliability and safety of both the device itself and the examination process, smaller dimensions of the device while maintaining the required level of functionality.

3 cl, 1 dwg

Description

Field of technology to which the invention relates

The present invention relates to the field of vehicle control monitoring, and more particularly to an apparatus, system and method for monitoring a driver's performance during a driving test.

State of the art

In recent years, information technologies, penetrating into many industries, contribute to the rapid development of technology and technology, improving the quality and convenience of existing services and the emergence of many new services. One of these industries is transport, in which unmanned systems are being actively introduced. However, the most advanced solutions currently have only level 4 autonomy, and they are used only in a few cities in the world. Plans to develop a fully self-driving car at Level 5 autonomy, announced by many manufacturers, date back to 2026 at the earliest. Accordingly, taking into account the time required to develop, implement and mass disseminate such technologies, we can expect that the era of completely unmanned vehicles will come no earlier than in 10 years.

Thus, there is still a need for driver training, and therefore, for conducting driving tests. It is worth noting here that despite all kinds of advances in information technology, the automation of the process of conducting such exams is either completely absent or extremely low, and the assessment of the driver candidate’s readiness is carried out subjectively by the inspector.

Although a study of the situation shows that attempts are being made around the world to develop systems to assist the inspector during the examination or those seeking to fully automate the examination system, there are no examples of successful mass practical operation of such systems. The reasons for this are that the proposed solutions are either compact, but process few events and errors, which still leads to the need for active participation of the inspector in the examination process and in making decisions on the presence or absence of errors, thereby practically negating the feasibility the use of such a system, or (for example, systems with powerful industrial on-board computers taken directly from self-driving cars) are unnecessarily complex, voluminous and expensive both in terms of the finished product and in terms of installation and maintenance, which affects the profitability of the system and readiness of relevant departments for its implementation.

Meanwhile, the analysis carried out by the authors of the present invention of the procedures for taking a practical exam for the right to drive vehicles shows that these procedures can be automated almost completely. For example, in Russia, the current examination procedure is regulated by the Administrative Regulations of the Ministry of Internal Affairs of the Russian Federation for the provision of public services for conducting examinations for the right to drive vehicles and issuing driver’s licenses. According to the regulations, the inspector is located to the right of the driver, gives instructions on the route, keeps a log of errors and monitors the safety of maneuvers, while checking 19 driving skills and monitoring 37 errors (of which 34 are for category B). Most of the errors being checked (including critical ones and important ones when assessing a candidate’s readiness to participate in road traffic) are fully or partially formalizable, that is, they can be automated. Some other operations, such as issuing a task along the route and maintaining an error log, can also be automated. In addition, according to the regulations, audio and video recording of the delivery process is required with a storage period of at least 1 month, which one way or another leads to the need for the inspector to use appropriate technical means. The development of the technical means used in the vehicle to the maximum possible automation of the process without significant additional costs would reduce the burden on inspectors and increase the transparency and objectivity of grading and decision-making on the result of the exam (pass/fail).

Turning specifically to technical solutions known from the prior art, we can note, for example, document RU 2580091 C1, which discloses an automated system for monitoring and assessing the performance of control exercises of a practical exam for obtaining the right to drive vehicles of various categories. This technical solution is an attempt to provide a fully automated examination system, when the driver candidate is alone in the car, data from sensors located in the car and the image inside the car are sent to a remote control center, which automatically monitors the progress of the exam and detects possible errors, and the dispatcher can view the position and vector of the vehicle’s movement on the map in real time and, if necessary, remotely turn off the engine. However, as mentioned above, the current level of autonomy of unmanned systems, as well as driving control systems, is such that it is not yet possible to correctly, taking into account all possible errors and circumstances and in real time, fully automatically assess the correctness of the actions of a candidate driver driving an examination vehicle in normal traffic. Therefore, it is absolutely natural that the system proposed in this document involves the use of a specialized site, which is only an imitation of the city road network with asphalt concrete pavement and control exercise zones located in such a way as to ensure their consistent implementation. In addition, the flow of data transmitted from sensors and the interior camera to a remote control center is unstable and intermittent when driving in real traffic flow, which leads to the impossibility of continuous remote control, and therefore to the impossibility of conducting an exam and obtaining an assessment in real time.

A technical solution is also known from the prior art, described in document WO 2021/045445 A1, which discloses an on-board device for conducting a driving test, which first, based on the location and orientation of the vehicle, determines in which test mode and in which zone it is located, and then, depending on the specific mode and zone, evaluates and comments on the actions of the candidate driver. At the same time, this device independently determines possible violations committed by the candidate during the test only when the car is on a special test site and it is necessary to evaluate the correctness of performing preset exercises in preset zones of this site, whereas when switching to driving mode on the road, it only provides information about the surrounding road conditions (namely, about the weather, traffic jams and traffic lights), relevant recommendations on how best to behave in such a situation, statistical messages about driving style (namely, about the average speed, the number of accelerations and braking , number and angle of steering wheel turns) and corresponding recommendations on how to improve these statistics to increase the likelihood of passing the exam. Accordingly, this solution cannot be used to automate the process of conducting an exam on the road, but is aimed primarily at preliminary training of the candidate in a driving school.

Thus, in the state of the art, there has been a need to create a compact and inexpensive on-board device that would provide the maximum possible automation of the process of conducting an examination for the right to drive vehicles and that would be simple and inexpensive to install and maintain, in order to ensure the possibility of its mass practical implementation. use.

The essence of the invention

In order to overcome at least some of the above-mentioned disadvantages of the prior art, the present invention is directed to devices, systems and methods for monitoring the control of a vehicle.

According to the present invention, a method for conducting an examination for the right to drive vehicles is proposed, performed in a system containing an on-board device for conducting an examination for the right to drive vehicles, as well as a remote server and/or a portable device of the examiner, containing the stages in which:

before the start of the exam, programs and data of the on-board device are loaded into the permanent memory of the on-board device, including a map of the area and a pre-marked route for the examination, containing pre-defined initiating points and/or sections on the route;

identifying the driver based on data entered by the user or based on recognition using peripheral components of the on-board device;

provide the driver with access to the exam using a remote server and/or the examiner’s portable device;

launch the examination process using a remote server and/or an examiner’s portable device;

during the examination process, signals and/or data from peripheral components of the on-board device are received using a wired communication module of the on-board device, wherein the peripheral components of the on-board device contain at least one camera;

storing signals and/or data from peripheral components of the on-board device, including video recordings from said at least one camera, in the permanent memory of the on-board device;

using the navigation module of the on-board device, determine the location of the vehicle on the terrain map based on the received navigation data;

using the wireless communication module of the on-board device, they exchange data on the progress of the exam with a remote server and/or with a portable device of the examiner;

determining, using the on-board device processing module, the device's approach to a predetermined initiating point or section on the route;

launch the acceleration module of the neural networks of the on-board device to process an event corresponding to a given point or section using a pre-trained neural network, wherein the initiating point or section on the route and the corresponding event are associated with checking the correctness of the execution of the exercise, task or maneuver by the candidate driver being examined in the process conducting an exam; And

using the on-board unit processing module, a remote server and/or the examiner's portable device, the driver is assessed correctly in performing exercises, tasks or maneuvers along the exam route based on processing data received from peripheral modules and from the on-board unit navigation module.

In one of the embodiments, the method additionally contains stages in which, after the end of the exam, an examination protocol is generated, containing indications of errors and circumstances of the examination, data confirming the signature of the driver and the examiner is received, and the protocol is sent to a remote server.

In one embodiment, the indications of errors and circumstances of the examination contain data on the progress of the examination, including the results of assessing the correctness of the driver's performance of exercises, tasks or maneuvers received by the on-board device, and data on the progress of the examination entered by the examiner using the examiner's portable device .

Technical result

The present invention improves the efficiency of devices, systems and methods for monitoring vehicle control. This ensures:

- full control of vehicle control within the framework of a predetermined list of exercises, maneuvers and checks;

- increasing the reliability and safety of both the device itself and the examination process;

- simplification of data processing, reduction of requirements for device performance;

- reduction of resource costs and energy consumption;

- reducing the size of the device while maintaining the required level of functionality;

- simplification of technical support for the device;

- reduction of the human factor;

- increasing the transparency of the exam, grading and the level of confidence in the exam and the examiner;

- simplification of the examination procedure from the position of the examiner;

- confirmation of the actions and decisions of the examiner.

It should be understood that not each of the embodiments may simultaneously provide all of these advantages over all known solutions from the prior art. Accordingly, some embodiments may have only some of these advantages or other advantages relative to some known solutions.

These and other advantages of the present invention will become apparent upon reading the following detailed description taken in conjunction with the accompanying drawings.

Brief description of drawings

In FIG. 1 is a block diagram of a vehicle control monitoring apparatus according to the present invention.

It should be understood that the figure is presented schematically and not to scale and is intended primarily to facilitate understanding of the present invention.

Detailed description

General overview of the system

The following description generally describes the procedure for conducting a driving test, but it should be understood that the principles of the present invention also apply to procedures for testing the driving skills of a driver or candidate driver other than as part of an examination conducted on behalf and under the control of official bodies and departments For example, the device, system and methods described in the present invention can be used in driving schools in the process of learning to drive or for conducting intermediate or final control tests of a candidate driver.

Accordingly, although in the following description the inspector is referred to as the supervisor, embodiments in which the controller is a driving school instructor are also possible.

Moreover, some of the principles of the present invention can be used to monitor driving skills without the participation of a supervisor. For example, the proposed device can be installed in a personal car by or at the request of a person wishing to improve his skills. In one embodiment, the presence of the proposed device may be mandatory for a vehicle that is licensed to be driven by a person with less than the threshold driving experience. For example, if a person with less than 1 year of driving experience is included in the car insurance policy, then such a car may be subject to an obligation to install the proposed device. In another embodiment, the presence of the proposed device in a vehicle may qualify for a discount on an insurance policy. In addition, the proposed device can be installed in a rental or leasing vehicle, in a taxi, or in car sharing. Also, the proposed device can be installed in service vehicles to monitor their use.

Hereinafter, in this document, a candidate driver may, for the sake of simplification of description, be called a driver, since in fact at the time of the examination or test he is behind the wheel and controls the vehicle. In addition, another person may be behind the wheel who is not directly the driver candidate who is currently undergoing a test or exam, and it is this person who will actually be the driver.

A vehicle can be considered a car, truck, bus, trolleybus or other vehicle on which the proposed device for monitoring vehicle control can be installed and used.

The vehicle in which the device according to the present invention is installed may be referred to hereinafter as a carrier vehicle for convenience of description and to distinguish it from other vehicles that may be mentioned herein.

While reference is made hereinafter to an inspector's tablet, it is generally understood to refer to a user's electronic device containing a screen and means for data entry and control of the device. Accordingly, the user's device may be a smartphone, tablet, laptop, electronic device specially designed for this purpose, or other suitable device.

A vehicle control system according to the present invention comprises a carrier vehicle, a compact vehicle control device installed in the carrier vehicle, a plurality of peripheral devices installed in or on the carrier vehicle and exchanging signals and/or data with a device for monitoring vehicle control, as well as a remote server that stores current maps and routes, as well as, if necessary, a protocol for each exam, audio and video recordings and software for updating. The system may also contain a supervisory device - for example, an inspector's tablet.

The driver candidate gets into the vehicle, an interior camera identifies him, and the device or system gives him access to the test. The inspector sits nearby, launches the exam in an application on his tablet and gives the command to the driver to start driving. While driving, audio and video recordings are made, and the entire examination process is monitored. When formalized driving errors occur, they are recorded in the protocol automatically by the device, unformalized errors are marked in the tablet by the inspector. The necessary data is sent to a remote server directly during the exam. After the exam is completed, the protocol is signed by the candidate and the inspector and sent to a remote server.

This provides a system that provides full control of the test driving experience with increased safety, reliability and reduced costs, while reducing human error, increasing test and scoring transparency and increasing confidence in the test and examiner. The examination procedure is simplified for the examiner himself, since he needs to perform fewer actions, and most of his actions and decisions are confirmed (supported) by the proposed device and system.

Device structure

Next, with reference to Fig. 1, a vehicle control device 100 according to the present invention will be described in more detail. A special case of device 100 is a device for conducting an examination for the right to drive vehicles.

The device 100 contains a central processing module 1, a neural network acceleration module 2, a peripheral processing module 3, a navigation module 4, a wired communication module 5, a wireless communication module 6, a cybersecurity module 7, a wireless update module 8, a RAM module 9, a permanent memory module 10 memory, emergency data storage module 11, power supply module 12 and diagnostics and protection module 13 from external interference.

Depending on the specific application, some of these modules may be missing, as will be explained in more detail later in this document. Arrows between modules in Fig. 1 shows their functional connections. It should be understood that the relationships shown are illustrative only and, if necessary, some of these relationships may be omitted or other relationships may be added. In FIG. 1 does not reflect the functional connections of the power supply module 12 and the diagnostics and protection module 13 from external interference, since in the preferred embodiment they are somehow connected with all other modules.

The central processing module 1 (hereinafter may also be called the first processing module) is the main computing module from the point of view of organizing the operation of the entire device and serves the needs of the operating system and, if necessary, other modules and/or applications, providing them with computing and/or processing resources. The physical implementation of the central processing unit is a central single-core or multi-core processor (CPU), die, system on a chip (SoC), chip, or the like. The central processing module 1 is connected to a neural network acceleration module 2, a peripheral processing module 3, a wireless communication module 6, a cybersecurity module 7, a wireless update module 8, a RAM module 9, a permanent memory module 10, and an emergency data storage module 11. Also, if necessary, if the conditions and requirements of a particular application allow it, the central processing module 1 can be connected to the navigation module 4 and the wire communication module 5 (not shown in Fig. 1). In this case, a connection means the presence of tracks, conductors and other physical connections by which one module is connected directly or through intermediate (auxiliary, matching, converting) elements to another module.

Neural network acceleration module 2 is a hardware neural network accelerator. During an exam or in vehicle control mode, the neural network acceleration module 2 performs processing tasks that need to be performed using artificial intelligence methods - in particular, using neural networks. This significantly reduces the load on the central processing module 1. The structure and operation of neural network acceleration module 2 is disclosed in more detail below in the corresponding section of the detailed description. The neural network acceleration module 2 is connected to the central processing module 1, and can also, if necessary, be connected to the peripheral processing module 3, wired communication module 5, RAM module 9, and permanent memory module 10.

The peripheral processing module 3 (hereinafter may also be called the second processing module) serves to organize data exchange between the central processing module 1 and the RAM module 9 on the one hand, and the navigation module 4 and the wired communication module 5 on the other side. The central processing unit 1 and the peripheral processing unit 3 may be separate devices or may be located on the same chip/chip, but in any case they are physically separated from each other. The peripheral processing module 3 may be implemented as a separate processor, a coprocessor, a controller, a chip, a circuit on a common printed circuit board with other modules of the device 100, a circuit on its own circuit board, a chip, a system on a chip (SoC), or other suitable device. or any combination of the above. The peripheral processing module 3 is connected to the central processing module 1, the navigation module 4, the wire communication module 5, the RAM module 9 and the emergency data storage module 11.

The navigation module 4 is used to determine the location of the device and the carrier vehicle based on the received navigation data. The navigation module 4 is connected to the peripheral processing module 3. The structure and operation of the navigation module 4 is described in more detail below in the corresponding section of the detailed description.

Wired communication module 5 is designed for wired communication with peripherals: cameras, radars, sensors, etc. Examples of wired communication technologies and methods used include CAN, Ethernet, fiber optic, twisted pair, asynchronous interface, Modbus, etc. The hardware implementation of the wire communication module 5 contains one or more separate chips, network cards, controllers, etc. The wire communication module 5 is connected to the peripheral processing module 3.

Wireless communication module 6 is designed for wireless communication with external devices - in particular, with a remote server, with an inspector’s tablet, with wireless peripheral components. In addition, the wireless communication module 6 can receive corrections from differential correction systems. The wireless communication module 6 may include a mobile communication module based on one or more radio access technologies, such as 2G, 3G, 4G, 5G, 6G, etc., and/or a wireless local area network module based on one or more technologies, such as WiFi, WiMax, Bluetooth, UWB, ZigBee, infrared, etc. The wireless communication module 6 is connected to the central processing module 1. The device 100 has a critical need for high-speed data transfer technologies to prevent large amounts of video and other exam data from accumulating in its memory. At the same time, when driving on the road, the coverage of both mobile communications and wireless local networks is unstable and varied, so the ability to transmit data using lower-speed data transmission technologies also allows you to increase the likelihood of successful data transmission at any given time at any given point route. Also of great importance is the ability to quickly and with minimal data loss switch from one data network to another data network, including between different technologies.

Cybersecurity module 7 is designed to protect against hacking using external programming methods (mainly via wireless communication). The cybersecurity module 7 is connected to the central processing module 1. The structure and operation of cybersecurity module 7 is discussed in more detail in the corresponding section of the detailed description.

The wireless update module 8 is designed to update the software and operating data of the device wirelessly. The wireless update module 8 is connected to the central processing module 1. The structure and operation of the wireless update module 8 are described in more detail below in the corresponding section of the detailed description.

The RAM module 9 (hereinafter may also be called the first memory module) is designed for temporary storage of data necessary for the device and its modules during the execution of current tasks with the ability to quickly access them - first of all, the central processing module 1 and the peripheral processing module 3, when they run out of their own cache memory. The operating principles and structure of RAM are known to those skilled in the art and are not the subject of the present invention and are therefore not disclosed here in more detail. In FIG. 1 shows that the RAM module 9 is connected to the central processing module 1 and the peripheral processing module 3, but it can, if necessary, be connected to other modules. The hardware implementation of the RAM module 9 may contain RAM memory, flash and other types of memory capable of providing high data exchange speeds.

The permanent memory module 10 (hereinafter may also be called the second memory module) is required for permanent storage of all programs and data, including storage of the operating system, maps, routes, exam protocol, video and audio recordings. Physically, the permanent memory module 10 may be implemented as a solid state drive such as an SSD. The most voluminous data is video recordings from cameras. Since there is a lot of data, it must be transferred as quickly as possible to the remote server and deleted from the permanent memory module 10 in order to free up space for more and more data. Accordingly, in practice, the storage of most data is not truly long-term. The size of the permanent memory module 10 is directly dependent on the peak volume of data that needs to be stored and the speed at which the conditions and requirements of a particular application allow it to be transferred. For example, if there is stable high-speed 5G or WiFi coverage along the entire route, this allows for fast data transfer and the use of a less capacious permanent memory module 10.

Emergency data storage module 11 is designed to restore data in the event of a power failure or exam failure. The emergency data storage module 11 is connected to the central processing module 1 and the peripheral processing module 3. The structure and operation of the emergency data storage module 11 is disclosed in more detail below in the corresponding section of the detailed description.

Power module 12 is designed to provide stable and continuous power to the remaining modules of the device. The structure and operation of the power supply module 12 are described in more detail below in the corresponding section of the detailed description.

Module 13 for diagnostics and protection against external interference is designed to monitor attempts to physically access the device. The structure and operation of the diagnostic and protection module 13 from external interference is described in more detail in the corresponding section of the detailed description.

Bus 1-3, connecting the central processing module 1 and the peripheral processing module 3, is highly loaded, since it transmits data from modules 4 and 5: satellite navigation, sensors, cameras and radars.

Bus 1-10, connecting the central processing module 1 and the permanent memory module 10, is also highly loaded, since it stores data traveling along the bus 1-3 to the permanent memory module 10, and is also transmitted in the direction of the wireless communication module 6 to the remote server/cloud.

The bus 1-6 connecting the central processing unit 1 and the wireless communication unit 6, respectively, is also highly loaded, especially at times when there is a good wireless communication channel. Precisely because the amount of data from the permanent memory module 10 is very large and high performance is needed, especially at times when there is a good wireless communication channel, the wireless communication module 6 is connected directly to the central processing module 1 rather than to the peripheral processing module 3, which would put even more stress on bus 1-3.

Thus, the present invention provides a vehicle control device that provides complete control of the vehicle through a predetermined list of exercises, maneuvers and checks with increased safety, reliability and reduced costs, while reducing human error, increasing the transparency of the examination and scoring grades and increases the level of confidence in the exam and the examiner. The examination procedure is simplified for the examiner himself, since he needs to perform fewer actions, and most of his actions and decisions are confirmed (supported) by the proposed device and system.

Neural network acceleration module

As stated above, the neural network acceleration module 2 is a hardware neural network accelerator. Namely, it can be implemented as a separate neural processor (NPU), a coprocessor, a chip, a circuit on a common printed circuit board with the rest of the modules of the device 100, a circuit based on its own printed circuit board, a chip, a system on a chip (SoC), a dedicated specialized area in within the architecture of the central processing module 1, in the form of another suitable compact device or any combination of the above.

During an exam or driving control mode, the neural network acceleration module 2, being a device optimized for neural networks, performs the processing tasks that need to be performed using artificial intelligence methods, that is, many small calculations performed on features and weights coefficients of the neural network model.

The neural network models themselves, in the preferred embodiment, are stored in the own memory of the neural network acceleration module 2 (for example, in the flash memory of a neuroprocessor) to speed up access to them, but if necessary, they can be stored in the device’s permanent memory module 10. The neural networks used in the proposed device 100 are pre-trained. Specific methods for training neural networks are known to those skilled in the art and are not the subject of the present invention and are therefore not disclosed in detail here.

The presence of neural network acceleration module 2 in the proposed device significantly reduces the load on the central processing module 1, which is completely or partially freed from resource-intensive neural network calculations and can continue processing many other processes.

In one embodiment, such processing tasks assigned to the neural network acceleration module 2 are reduced to processing the video stream in order to perform certain types of checks, such as passing a traffic light, passing a pedestrian crossing, passing a railway crossing, etc. The audio stream can also be analyzed if necessary, as will be mentioned later in this document. However, it should be understood that the proposed device 100 is not designed for fully autonomous driving, but to obtain the most compact device possible that allows you to control a predefined limited list of exercises, tasks and maneuvers, therefore, it is not intended to use neural networks as an acceleration module 2 industrial computers, which are usually installed on unmanned vehicles and have great performance, but at the same time take up a lot of space and consume a lot of energy. In one embodiment, the dimensions of the neural network acceleration module 2 do not exceed 1-DIN. The capabilities of neural network acceleration module 2 are limited and incomparable with the performance of an industrial computer, therefore the present invention does not provide for continuous analysis of the video stream.

At the same time, given that the types of checks performed by neural network acceleration module 2 fit into a small predefined list of tasks, its capabilities are sufficient to operate in a mode close to real time. In particular, if the result of a test is reported within a few seconds of the actual execution of the relevant exercise, task or maneuver, the result can be considered relatively relevant and acceptable from the point of view of the examination. For example, if 5 seconds after passing a pedestrian crossing, neural network acceleration module 2 displays a message that the driver did not allow a pedestrian to pass, who was on the edge of the pedestrian crossing facing the road, then this is the result if this event was not noticed by the driver and/or inspector immediately , is still not lost its relevance, since during this time the exam, with a high degree of probability, has not yet been completed, the vehicle has not traveled a long distance after this event, and the driver may not yet have been given new tasks by the inspector. Even if this event was noticed by the driver and/or inspector immediately, waiting for the result within an interval of up to 10-15 seconds is not critical, since it is a confirmation of what was seen. As an example illustrating these statements, we can point out that in 15 seconds a vehicle at a speed of 40 km/h will cover only 170 m. Moreover, according to the above-mentioned regulations, to conduct a practical exam for the right to drive vehicles of categories “B”, “ C" and "D" and subcategories "C1" and "D1" are given up to 30 minutes, and up to 10 minutes after completion of the exam to determine and record the results.

Despite its small size, the neural network acceleration module 2 can consume a lot of resources (in addition to power, it may also require data bus resources and memory resources, and, if there is insufficient computing power, the computing resources of the central processing module 1). Accordingly, it is impractical to keep it constantly on and it is advisable to put it in a standby mode, an idle mode, an idle mode or a similar mode in between active work, if supported by the particular selected model of the neural network acceleration module 2. Launching (preparing for operation) module 2 acceleration of neural networks may require some time. In this regard, during the exam, it is proposed to launch module 2 of acceleration of neural networks only before those moments at which its work is necessary, and for the period of work assign it the highest possible priority so that it works as quickly as possible, so that the result of its work is relevant, and it could be turned off and return to normal operation of other modules. To do this, along the route, predefined points/sections are marked on the map, during which neural network acceleration module 2 should operate - for example, intersections with traffic lights, pedestrian crossings, railway crossings, etc. While driving, the navigation module determines the coordinates of the vehicle, and when approaching the specified point/area, the neural network acceleration module 2 is launched, so that it is ready for work in time and does its part of the job. At the end of the work, it is switched back to standby mode. This ensures an increase in the overall performance of the device and a reduction in resource consumption while maintaining all the necessary functionality of the device. In addition, the fewer resources a device spends, the smaller the size of components required for its operation, that is, the more compact it can be made.

To reduce the load on the neural network acceleration module 2 and speed up its operation, the video stream may be pre-processed before being supplied to the neural network acceleration module 2. In particular, first, the image can be searched for the required objects (for example, a traffic light, a pedestrian, a train) using computer vision methods that do not use a neural network, then the image can be cropped to a region of interest containing only the found object, and so the image of the area of interest can be fed to the input of the neural network acceleration module 2. Also, if necessary, the image of the region of interest can be brought to the size (resolution) necessary for the correct operation of the neural network acceleration module 2. In some embodiments, computer vision methods that do not use neural networks work significantly faster than neural network methods, so the combination of preliminary search for a region of interest without the participation of neural networks and further analysis of only the region of interest using neural networks can further optimize the operation of the device, increase its performance and reduce resource consumption .

It was indicated above that during the operation of neural network acceleration module 2, it is given the highest possible priority. Meanwhile, when conducting an exam, due to the need for video recording, the process of saving the video recording to the permanent memory module 10 may be continuous and mandatory, and one of the most resource-intensive processes is also sending the video recording to a remote server. Since the permanent memory module 10 and the central processing module 1 are not heavily loaded with neural network processing tasks during the operation of the neural network acceleration module 2 (in a specific non-limiting example, their load with such tasks is about 3-5%), they can perform not only saving video recordings, but and, if possible, sending it to a remote server. Moreover, a wireless connection with a remote server while driving on the road may be unstable, therefore, in order to reduce the required volume of the permanent memory module 10, it is necessary to make maximum use of any data transfer opportunity, while the unreasonable omission of such an opportunity directly affects both the dimensions of the device and its functionality and the ability to conduct the exam.

In addition, in one embodiment, the processes of navigation (or localization, determining location, and optionally direction and time) and collecting data from the vehicle (peripherals, sensors, CAN bus, etc.) can be more critical and important during the exam, therefore, to increase the reliability and safety of the exam, the priority in using device resources for these processes may be higher than that of neural network acceleration module 2.

Navigation module

The navigation module is configured to determine the location of a given vehicle based on the received navigation data. In addition, the navigation module, based on the received navigation data, can determine the direction of movement of the vehicle (for example, azimuth) and synchronize the time.

In one embodiment, navigation data is ready-made data on the coordinates of the vehicle, time and/or direction of movement of the vehicle, received by the navigation module from external devices. For example, such data may be received from a Global Navigation Satellite System (GNSS) receiver located external to the device. The GNSS receiver, in turn, receives satellite navigation signals from the GNSS antenna located on the vehicle. The GNSS antenna is preferably located in the area of maximum visibility of satellites - for example, on the roof of a vehicle or on the windshield, under it, etc. GLONASS, GPS, Beidou, Galileo, QZSS or any other global navigation satellite system or a combination thereof can be used as GNSS. This embodiment makes it possible to simplify the device, reduce its dimensions and, if necessary, save space for other modules in the device body or in the slot allocated for the device in the car.

In one embodiment, the navigation module may itself comprise a GNSS receiver configured to receive satellite navigation signals from the GNSS antenna. This option makes it possible to produce a device that is less dependent on external components, and vice versa, which imposes fewer restrictions on the characteristics of external components suitable for use, since the manufacturer independently debug the operation of the GNSS receiver and, if necessary, updates the corresponding software, which also improves the reliability of the device.

In one embodiment, more GNSS antennas and receivers may be used—for example, 2 antennas and 2 corresponding GNSS receivers. All or part of the GNSS receivers can be built into the navigation module or external. In this case, the antennas must be spaced at a certain distance from each other - for example, 0.5 m, 1 m, 2 m. In this case, the GNSS signals received by the antennas will be slightly different, which will allow the antenna coordinates to be distinguished. Knowing the exact position of each antenna relative to the vehicle, it is possible to determine with increased accuracy the direction of movement and position of the vehicle on the road, taking into account its dimensions, and monitor related parameters and events. For example, due to this, it is possible to track the location of the wheels of an examination vehicle within a traffic lane with acceptable accuracy without the use of resource-intensive calculations based on computer vision methods.

The navigation module can determine location, direction and exact time additionally based on corrections received from differential correction systems such as Real Time Kinematic technology or the like. This makes it possible to achieve positioning accuracy of 3-5 cm or better with a synchronization accuracy of 5 ms or better. It should be noted that corrections are received not through the GNSS receiver, but through an alternative means of communication - for example, through a wireless communication module (in particular, through a mobile communication module using radio access technology such as 2G, 3G, 4G, 5G, 6G etc., or via wireless LAN module using technology such as WiFi, WiMax, Bluetooth, UWB, ZigBee, infrared, etc.).

The driving test is an extremely responsible procedure, so any possible malfunctions in the operation of the device should be completely avoided or at least minimized. One of these failures that directly affects the examination process can be GNSS signal failure, which can occur as a result of loss of the GNSS signal due to poor satellite visibility conditions or as a result of spoofing, that is, the replacement of a true GNSS signal with a fake signal with other data. Due to a GNSS signal failure, the GNSS receiver may either cease to determine coordinates, direction, and time altogether, or may begin to produce erroneous readings. Because of this, the device may, for example, not detect in time the approach to the initiation point of the neural network acceleration module, and it will not be able to process the corresponding event. In addition, the journal/log/exam protocol may be disrupted. In this regard, the navigation module may provide an option to check the accuracy of the GNSS receiver readings.

When two or more antenna diversity are used, the actual GNSS signals at each antenna will vary slightly. Accordingly, if it is determined that identical data has been received from at least two GNSS antennas or receivers (depending on the implementation), then this indicates the presence of a spoofing attack, and such data is incorrect. Additionally, if the discrepancies between data from different GNSS antennas or receivers are too large (exceed a preset threshold), or no or zero signals are received from any GNSS antenna or receiver, this may indicate a GNSS signal failure or failure. antenna or GNSS receiver.

If the device or vehicle contains a built-in clock, its readings can be received and compared with those of the GNSS receiver(s). A difference in readings greater than a preset threshold indicates a GNSS signal failure or a problem with the antenna, GNSS receiver, or watch itself. In this case, it is necessary to compare the clock readings with its previous readings, and the GNSS receiver readings with its previous readings. If it is determined that the clock continues to run correctly (for example, the difference with its previous readings is greater than 0 and falls within a predefined interval, and the difference in the GNSS receiver readings does not fall within), then their readings can be used further for the exam until The correct operation of the GNSS receiver will resume. If necessary, the built-in clock can be synchronized from time to time (for example, periodically or when certain events occur) based on correct readings from the GNSS receiver.

If the device or vehicle does not contain a built-in clock, or if the built-in clock does not work or does not work correctly, or if additional time verification data is needed, then time data may also be supplied to the navigation module from the inspector's tablet. In addition, coordinates and direction data can be sent from the inspector's tablet to the navigation module if the inspector's tablet performs such measurements.

Additionally, if the device or vehicle contains an acceleration sensor (eg, an accelerometer) and an angular velocity sensor (eg, a gyroscope), then their readings may be compared with those of the GNSS receiver(s). For example, if the accelerometer and gyroscope clearly indicate that the device (and therefore the vehicle in which the device is embedded) is moving, but the GNSS receiver believes that the device is standing still, then the GNSS receiver readings may be considered incorrect.

Moreover, the navigation module may comprise an inertial navigation module configured to determine the location and direction of movement of the device based on signals received from at least one acceleration sensor and at least one angular velocity sensor. The inertial navigation module is autonomous and does not depend on the GNSS signal, therefore, from the moment it is determined that the GNSS signal is incorrect, the readings from the inertial navigation module can be used for further navigation. Like the GNSS receiver, the inertial navigation module can receive corrections using technologies other than satellite navigation - for example, using cell towers, WiFi hotspots, UWB networks, etc. Once the GNSS signal is restored, the navigation module can return to using the GNSS receiver readings. The inertial navigation module can serve not only as a replacement in case of GNSS signal failure, but also as a supplement or refinement for GNSS data. Depending on the requirements and conditions of a particular application, the correct GNSS receiver readings may weigh more heavily than the inertial navigation module readings. As with the GNSS receiver, the inertial navigation module can be located outside the device in the vehicle, and the navigation module then receives ready-made navigation data from the inertial navigation module.

This ensures that the correct examination procedure can be carried out even in the event of a GNSS signal failure, increasing the reliability of the device.

Periphery

According to the exam regulations, audio and video recording of the actions of the driver (driver candidate) during the exam may be required. Accordingly, the vehicle may be equipped with at least one interior camera facing the interior of the vehicle and capturing at least the area in which the driver is located. For example, such a camera can be installed in the upper part of the cabin in front of the driver. In addition to recording the driver's actions, this camera can be used to identify the driver. Specific identification methods are known to those skilled in the art and are not the subject of the present invention and are therefore not described in detail here.

Additionally, at least one interior camera may be installed in the vehicle, directed into the interior of the passenger compartment and capturing at least the area in which the passenger is located. In particular, such a camera may be necessary to record the actions of the examiner. In another embodiment, such a camera may be needed to monitor the presence or absence of others in the vehicle in the back seat and to record their activities.

In addition, the vehicle may be equipped with a set of exterior cameras containing at least two cameras directed outside the passenger compartment and capturing at least an area of the surrounding space near the vehicle - for example, the area in front of the vehicle, to the left of the vehicle, to the right of vehicle, behind the vehicle.

At least a portion of the set of exterior cameras may be positioned to provide a stereo image. A point cloud can be obtained from a stereo image, key points can be located, etc. The stereo image may be processed by the device to search for objects and events in front of the driver, to determine how the driver reacts to those events or objects, and/or to initiate execution of predefined device scenarios. To obtain a stereo image, the cameras must be positioned in such a way that at least part of their viewing angles intersect with each other (overlap). Accordingly, a stereo image appears in the area where viewing angles intersect.

The specific location of the cameras depends on the characteristics of the cameras and on the range and viewing angles within which two-dimensional or stereo images need to be obtained. For example, in one embodiment, there are 2 cameras spaced apart from each other, aimed at the area in front of the vehicle to obtain a stereo image in this area, and 2 or 3 wide-angle cameras, aimed at the remaining sides of the vehicle (to the sides and rear) and designed to obtain a flat image.

In another embodiment, there are 2 or more cameras located close to each other - for example, touching each other or being at a short distance. To increase the viewing angle, the cameras can be positioned at an angle to each other. The proximity of the cameras simplifies their installation and cable routing. Also, closely spaced cameras can be attached to a single rigid mounting device (such as a base or bracket), which not only further simplifies their installation in the vehicle, but also simplifies post-installation calibration. In addition, the fastening means may be in the form of a hood, may contain a hood, or may be configured to attach a hood. A lens hood may be necessary to protect against glare and glare. You can use one hood for multiple cameras, or one hood for each camera. Some cameras may not have a hood. In one non-limiting example, a set of exterior cameras containing three cameras facing the area in front of the vehicle and having a single lens hood may be mounted in the interior mirror area on a single bracket, and an interior camera camera may be attached to the same bracket, facing the area driver's location and does not have a hood.

In one embodiment, the video stream from each camera may be continuously recorded in the device's memory. In another embodiment, the video stream is continuously recorded only from a part of the cameras - for example, only from the internal one, while the video stream from the remaining cameras may not be recorded at all or recorded selectively - for example, periodically or in connection with the occurrence of predetermined events. Accordingly, some cameras may not operate continuously, but only periodically or in connection with the occurrence of predetermined events. For example, if a vehicle's location on the map determines that it is approaching a pedestrian crossing, a set of exterior cameras can be activated to monitor and record the process of crossing the pedestrian crossing. In another example, if a signal from at least one of the radars indicates dangerous proximity to an external object, one or more of the external cameras responsible for monitoring the area in which that radar is located may be activated and recording from that camera may be made. .

To record the sound environment inside the vehicle and/or near the vehicle, at least one microphone inside the vehicle (internal microphone) and/or at least one microphone outside the vehicle (external microphone) can be used. The microphone can be directional or omnidirectional. The use of directional microphones can allow audio streams from different directions to be isolated - for example, audio stream from the driver's area and audio stream from the passenger's area. For ease of installation, the microphone can be built into the camera. In this case, some cameras may be equipped with an omnidirectional microphone, some cameras may be equipped with a directional microphone, and some cameras may not have a microphone. If the camera and microphone are separate elements, they may have a single mounting hardware to simplify installation. For example, one omnidirectional internal microphone can be mounted on the same single bracket to which the internal camera and external cameras are attached.

In one embodiment, the audio environment may be analyzed to identify predetermined events - for example, the likelihood of a conflict, the sound of impacts or glass breaking, etc. may be determined. Another option may detect key words spoken by the driver or passenger, or simply the presence of speech.

As with cameras, the audio stream from each microphone can be continuously recorded into the device's memory. In another embodiment, the audio stream from only a portion of the microphones is continuously recorded - for example, only from the internal microphones, while the audio stream from the remaining microphones may not be recorded at all or may be recorded selectively - for example, periodically or in connection with the occurrence of predetermined events. Accordingly, some microphones may not operate continuously, but only periodically or in connection with the occurrence of predetermined events. For example, recording of the internal microphone may only be enabled at moments when it is determined that the driver or passenger has said the keyword "Record" or some other keyword that initiates recording. In another example, only those periods when there is speech inside the cabin may be recorded. In yet another example, if a signal from at least one of the radars indicates dangerous proximity to an external object, an external omnidirectional microphone or a directional microphone responsible for monitoring the area in which that radar is located may be turned on and recorded from it. microphone.

Synchronization of audio streams and video streams can be performed using precise time readings received from the navigation module.

At least one radar may be installed on the vehicle. Radars are designed to measure the distance (distance) to objects near a vehicle. The range over which radars must operate depends on the requirements of the particular application. By way of non-limiting examples, radars can be used to determine the distance to other vehicles located or moving near a given vehicle, lateral clearance, distance to a pedestrian, animal or other objects located or moving on or near a road/roadway. In one embodiment, it may be sufficient for the purposes of the present invention to have one radar at the front of the vehicle to monitor objects in front of the vehicle and one radar at the rear of the vehicle to monitor objects behind the vehicle.

A vehicle can be equipped with many sensors that provide information about other units/systems of the vehicle and driver actions that can be directly measured or tracked without the use of artificial intelligence methods - for example, driving speed, degree of pedal pressure (for this purpose, the pedals may have pads with sensor), whether the seat belt is fastened, whether the engine is running, etc. Data from the sensors enters the device and is processed by the corresponding modules to monitor vehicle control and/or detect predefined events. For example, if the average speed of the vehicle over a predetermined period of time or the instantaneous speed of the vehicle exceeds a predetermined permitted threshold, a control error may be determined indicating that the driver is not sufficiently skilled in driving at the maximum permitted speed. An indication of the error can then be recorded in the examination report.

If the vehicle is equipped with standard peripheral units (cameras, microphones, radars, sensors), the characteristics of which are suitable for use with the proposed device for monitoring vehicle control, the device can be connected to these units and use them to perform its functions. For example, if the vehicle already contains an interior microphone, it may be connected to the vehicle control device to capture the audio environment inside the cabin. In another non-limiting example, if the vehicle already contains radars for a parking assist system, data from those radars may also be used by the device to monitor vehicle control. This allows you to simplify the installation of the device and reduce the cost of organizing its operation.

I/O and Device Control

The proposed device may contain control and data input means to facilitate interaction with it. For example, any buttons, keys or other control and input means can be installed on the device body to control the power supply of the device, to control its modes, etc. In another embodiment, the controls and input elements may be located on the instrument panel, center console, or other positions from which device control and data input may be conveniently performed. In yet another embodiment, controls and inputs may be provided on a graphical user interface through which the device can be interacted with wirelessly.

In one embodiment, the device may contain only one physical power on/off button on the chassis and no more physical controls and inputs either on the chassis or remotely, and all other control and input may be performed through a graphical user interface . This reduces the likelihood of incorrect influence on the device and on the course of the examination, which increases the reliability and safety of the proposed device and examination procedure.

The graphical user interface can be accessed through an inspector's tablet, through a remote server, through a technician's equipment, etc. Device features and controls may vary from user to user depending on their access level and connection method. For example, a technician, connecting to a device through a specialized communication protocol, may have the highest level of access and therefore can control the internal settings of the device and its components, while an inspector may have less control over the device. This increases the reliability and safety of the proposed device. In addition, some settings may not be available in a particular device mode - for example, update functions may not be available in exam mode. This increases the reliability and safety of the proposed device and examination procedure.

Entering data into the device may be necessary during the exam - for example, an inspector, through the inspector’s tablet, can enter text, voice, photo or video notes regarding the completion of exercises and tasks or regarding other circumstances of the exam, can mark errors or correctly completed exercises and tasks , can, if necessary, adjust the route, etc. Before the exam, the inspector can enter his data, data about the upcoming exam, and data about the driver candidate into the device. After the exam, the proctor can enter exam results into the device. It should be noted that there may also be other reasons for entering data into the device, and although for the sake of simplicity of description they are not specified here in more detail, they should fall within the scope and principles of the present invention.

The proposed device may also contain data output means. As in the case of data input means, they can be implemented in the form of elements on the body (for example, a power indicator), remote elements (for example, an indicator on the dashboard, a display on the center console, etc.). In yet another embodiment, output elements may be displayed on a graphical user interface for wireless user interaction with the device. For example, an inspector on an inspector's tablet can view current information about the exam process, assessment of exercises and assignments, status of the device and its peripheral components, map, route and current location, candidate, etc. through a graphical user interface. A technician through his equipment, connecting to the device directly or through a remote server, can view the internal settings of the device modules and its peripheral components.

Access levels and the list of functions available for management for different users can be set by users who have the appropriate access level. Information about users, their access levels and the list of management functions available to them can be stored on a remote server and/or in the device's memory. Accordingly, during user authorization, the device can check information from the built-in memory and/or information from a remote server. The case where successful authorization requires the presence of a corresponding user both in the device memory and on the remote server is the most secure.

In one embodiment, only users from a predefined list of users that is stored in the device's memory may have maximum access to manual control of the device. The fewer users have such access, the higher the security of the device.

Remote server

Depending on the requirements of a specific application, a remote server can be either concentrated (centralized) or distributed, including implemented in the cloud.

The remote server stores the protocol of each exam, the required audio and video recordings (as mentioned above, according to the regulations, not all continuous recordings may be required). It also stores current routes, maps and software of the device itself and, if necessary, peripheral units to ensure the possibility of installing and updating software, including over-the-air (OTA) updates.

The remote server can store a database of candidates admitted to the exam. The inspector can use this database to ensure that the candidate who boards the vehicle is eligible to take the exam on that day. In addition, driver identification can be performed using an interior camera aimed at the driver. The image from this camera is sent to the server, the server recognizes the driver’s face from the image and compares it with the database of candidates admitted to the exam. If the match is successful, a notification is sent to the proctor's device and/or tablet and the exam can begin. If the match fails, a notification is sent to the inspector's device and/or tablet, and actions may be taken to prevent the test from being conducted with an unauthorized person in the driver's seat - for example, the test record may be blocked, or one or more devices in the vehicle may be physically blocked. , with the help of which the vehicle can start and/or continue moving.

Wireless update module

The software of the device proposed in the present invention is complex and needs to be constantly updated to ensure stable and reliable operation, to add functionality, to correct errors, to reduce resource consumption, etc. In addition, maps and exam routes may change quite often.

Meanwhile, since the proposed device in some embodiments can be used to conduct an official driving test, it must be protected from possible fraudulent and other actions aimed at falsifying the results of the test or interfering with the test process. Accordingly, in order to prevent unauthorized physical access by attackers to the device, it can be installed in the vehicle in such a way that there is no unimpeded physical access to it. It may also not have external connectors through which you can connect to it, since this is also a potential security threat (for example, if you provide the ability to connect a flash drive from inside a car, this can be a source of threat - both intentional and unintentional , since flash drives are often infected without the owner’s knowledge). All of this may require a technician with the appropriate level of access to the device each time an update needs to be made.

However, given the possible frequency of updates and the possible scale of use of the proposed device, it would be difficult to manually update it everywhere and on time. Frequent physical intervention on the part of a technician, in turn, can lead to a decrease in the level of protection of the device, since at some point the human factor may appear, and the technician may accidentally or intentionally take actions that reduce protection - for example, incorrectly closing the case. This can also lead to accelerated wear of the device body and the fasteners with which the device body is installed in the vehicle.

To solve these problems, the proposed device may include an over-the-air update module configured to update software and operating data of the device wirelessly. Software here refers to any software elements and modules that may be necessary for the device and its components to operate - for example, firmware elements for controlling the operation of the central processing module, software elements of the operating system, elements of the cybersecurity module, artificial intelligence models, etc. .d. Operating data here refers to any data that may be necessary for the device and its components to operate - for example, maps, routes, databases, data sets, etc.

Over-the-air update may also be called over-the-air update or OTA (over-the-air) update. To perform the update, the wireless update module communicates (exchanges data) with a remote server.

It is possible to implement both a hardware-software implementation of a wireless update module (for example, in the form of a separate chip or as a physically dedicated area in a chip or processor), or a purely software implementation, in which the software of the wireless update module uses functions and resources (of computing) for its operation. , memory, data transfer) other hardware modules. In general, the operations performed by the wireless update module can be divided into data storage, processing and transmission operations.

To perform storage operations, the wireless update module may contain its own hardware memory or use the resources of a permanent memory module and/or a RAM module.

To perform processing operations, the wireless update module may contain its own hardware processor or use the resources of a central processing module and/or an auxiliary processing module.

To perform data transfer operations, the wireless update module may contain its own hardware transceiver or may communicate with a remote server through a wireless communication module (in particular, through a mobile communication module using radio access technology such as 2G, 3G, 4G, 5G, 6G, etc. .p., or via a wireless LAN module using technology such as WiFi, WiMax, Bluetooth, UWB, ZigBee, infrared, etc.).

The update is performed outside the exam procedure, since, firstly, it is very resource-intensive, and there is no way to spend additional resources on extraneous processes that are not included in this procedure, and secondly, an attempt to update directly during the exam can lead to reboot or failure of any operations and modules, due to which the exam may fail. The wireless connection when driving on the road can be unstable and intermittent, which can lead to download errors and the need to download again, so it is preferable to update under conditions of a stable and high-quality wireless connection - for example, when the vehicle is parked in a garage or in a parking lot where high-speed Wi-Fi or 4G network.

In one embodiment, the update may be performed at the request of a user - for example, a responsible official of the department or organization that owns the vehicle. In another embodiment, the update may be performed at the command of a technician who provides technical support for the device. For security purposes, restrictions may be placed on exactly how (by what data transfer protocols) you can connect to the device management interface. For example, if the device can only be controlled via a direct Wi-Fi or Bluetooth connection, this would require the physical presence of the person who intends to control it in the immediate vicinity of the device/vehicle, thereby avoiding remote control of the device via the Internet . In yet another embodiment, the update may be performed automatically, that is, initiated by the device itself or by a remote server. In this case, the device can poll the server for updates, or the server itself can inform the device that there are updates for it. If necessary, the current configuration parameters of the device (for example, versions and settings of the software installed on it and versions of operating data) can be saved on a remote server. In another embodiment, instead of specifying the version of each application, element, and module, the remote server may store a device configuration identifier—for example, a hash of a configuration description file.

The update may contain many different files and/or instructions (commands). Files and/or instructions may be packaged in packages or archives. In particular, an update may contain control instructions and/or metadata necessary to organize the update procedure itself, patches that do not require a reboot of the corresponding component/module, files that require a reboot of the corresponding component/module, global updates that require a reboot of the entire device, etc. d.

To enhance security, the update may contain files and/or instructions designed to verify that unmodified licensed (or authorized) components/modules/data are installed. If it is detected that modified licensed components/modules/data or unlicensed components/modules/data are installed, this may indicate that an unauthorized attempt has been made to tamper with the device software and/or files are corrupted. In this case, you may receive a notification and the update may not install. In another embodiment, an attempt may be made to replace the modified/unlicensed/damaged files along with the update with the correct licensed ones.

To improve reliability, an update may contain files and/or instructions designed to help you verify that the update is appropriate for your device model and current software configuration. If an update is found to be completely or partially unsuitable, it may either not be installed at all, or only those components/modules/data that are suitable may be downloaded and updated.

To reduce the load on the remote server, on the data network and on the device itself, as well as to avoid wastage of network resources, the update may contain files and/or instructions designed to check whether the downloaded and/or installed files are already on this device, included into the current update package. Accordingly, files that are already on this device are not downloaded.

To improve reliability and to reduce the memory resources required for the operation of the device, the update may contain files and/or instructions designed to test which files should be deleted during or after the update.

Module for diagnostics and protection against external interference

As stated above, the proposed device can be installed in a vehicle in such a way that it does not have unobstructed physical access. However, even a discreet or hard-to-reach installation cannot guarantee that attackers will not attempt to carry out unauthorized actions against the device, its peripherals or the vehicle in which it is installed (carrying vehicle). In this regard, to increase security, the device may contain a diagnostic and protection module against external interference.

The module for diagnostics and protection against external interference is configured to diagnose at least the internal modules of the device, and also, if necessary, to diagnose peripheral components and the carrier vehicle.

In particular, diagnostics and protection against external interference consists of checking the functionality and correct response to interrogating signals from the diagnostic and protection module against external interference. For example, if the forward radar does not respond to interrogation signals at all, or if its responses differ from what they should be according to the diagnostic protocol, it may be concluded that the forward radar is not in proper condition. The reason for this may be that it is missing (for example, deliberately removed or dropped by itself), damaged (for example, intentionally by an intruder, accidentally as a result of a collision, or spontaneously due to aging or environmental factors), unauthorized replacement, or there is no connection with it (for example, , wires are damaged or disconnected) or it has been tampered with externally.

In addition, diagnostics and protection against external interference may include monitoring signals from sensors responsible for the safety of the device. One of these sensors can be located in the body of the device itself and respond to the opening of the case. For example, it can be implemented as a volume sensor, magnetic sensor, reed sensor, Hall sensor, etc. It can also monitor the disconnection/attachment of wires, unauthorized replacement of peripherals, etc.

Additionally, diagnostics and protection from external interference may include monitoring images from cameras - in particular, it may be detected that the view of the camera is partially or completely blocked or that the image from the camera is fraudulent.

Additionally, diagnostics and protection from external interference may include monitoring sound from microphones - in particular, unnatural noise or a fake audio stream can be detected.

The module for diagnostics and protection against external interference contains its own standby power, which is necessary to maintain operation for some time if the external power is turned off. Standby power can be implemented on the basis of a separate microcontroller. Standby power can be supplied from the device power module and/or, if necessary, can contain a separate battery. Battery capacity depends on the requirements of the specific application.

The module for diagnostics and protection against external interference may contain a separate chip (for example, a microcontroller) for monitoring and processing signals, or physically or logically allocated resources (computing, memory, data transfer) of other hardware modules can be used for these purposes.

If the diagnostic and protection module against external tampering detects that any component (module, sensor, peripheral component) is in an improper state or that an external tampering attempt has been made, a notification of the corresponding event is generated. The notification can be stored in the memory of the diagnostic and protection module from external interference and/or device, can be displayed on the inspector’s tablet, on the screen of the vehicle’s on-board computer, on a separately displayed indicator, and can be sent to a remote server. An audio notification in the form of a voice message or alarm may also be displayed. The specific type of notification that should be displayed at a given time may depend on what mode the device is in. For example, if the vehicle is stationary and locked and the device is in idle (standby) mode, an audible alarm may be output and a message may be transmitted to a remote server. If an exam is being conducted, the vehicle is in motion, and the inspector is in the vehicle, then the sound, for example, may not be output, but a notification may be received on the inspector's tablet, and the inspector can then decide whether this event is grounds for termination or interruption exam or you can continue it. In another embodiment, events from a predefined list of events or any events may automatically cause the exam to be terminated or paused. For example, if both the permanent memory module and the emergency storage module are not operating at the same time, information about the further progress of the exam cannot be stored in memory, so it may be advisable to pause it. If the internal camera stops working or is partially or completely blocked, further recording of the exam will not be recorded correctly, so it may be advisable to pause the exam. If the problem is subsequently identified (for example, the chamber is no longer blocked), the exam can continue.

The presence of a diagnostic module and protection from external interference in the proposed device increases the reliability and safety of both the device itself and the examination process.

Cyber Security Module

The proposed device has voluminous and complex software. In addition, using wireless technologies, it can be controlled externally, updated, and transmit and receive data. Accordingly, attackers may attempt to interfere with the exam procedure, gain access through the device to data in the device’s memory and/or on the server (including maps, routes, personal data, operating algorithms), etc.

To avoid such threats, the device may contain a cybersecurity module configured to protect the device from hacking by programming techniques that could allow control of the device to be gained or fraud to be committed on or through the device. The cybersecurity module monitors authorization processes, compliance with encryption protocols, the security and reliability of encryption keys, the reliability of software updates, and also monitors input data for threats.

It is possible to implement both a hardware and software implementation of a cybersecurity module (for example, in the form of a separate chip or in the form of a physically allocated area in a chip or processor), as well as a purely software implementation, in which the software of the cybersecurity module uses functions and resources (computations, memory) for its operation , data transfer) other hardware modules.

If the cybersecurity module detects that a particular software process or data element is a potential source of threat and/or is unreliable, an event notification can be generated. The notification can be stored in the memory of the cybersecurity module and/or device, can be displayed on the inspector’s tablet, on the screen of the vehicle’s on-board computer, on a separately displayed indicator, and can be sent to a remote server. An audio notification in the form of a voice message or alarm may also be displayed. Upon receiving notification of a cybersecurity threat event, a user (for example, an on-site inspector or a remote technician) can instruct the cybersecurity module on how to handle the threat (for example, try to resolve it or ignore it), or, with the appropriate level of access, can independently access processes and data and process the threat.

The specific type of notification that should be displayed at a given time may depend on the threat level and what mode the device is in. For example, if the vehicle is stationary and locked, the device is in over-the-air update mode, and a low-level threat is detected by the cybersecurity module, a message can be sent to a remote server so that a technician can check the status of the device if necessary. If an exam is being conducted, the vehicle is in motion, and the inspector is in the vehicle, then the sound, for example, may not be output, but a notification may be received on the inspector's tablet, and the inspector can then decide whether this event is grounds for termination or interruption exam or you can continue it. In another embodiment, events from a predefined list of events or any events may automatically cause the exam to be terminated or paused. For example, if during an exam a maximum level threat is identified that can lead or is already leading or has led to incorrect operation of the persistent memory module and the emergency data storage module (for example, log files as a result of the attack were encrypted with an unknown key), then information about the exam progress will not be will be able to be properly stored, so it may be advisable to pause it. If the threat is subsequently determined to have been resolved (for example, the processes and data at the source of the threat are removed or corrected, and data containing information about the exam's progress is restored), the exam can continue.

The presence of a cybersecurity module in the proposed device increases the reliability and security of both the device itself and the examination process.

If the device has a wireless update module, then the presence of a cybersecurity module becomes especially important, since during the update, essentially, the main task of the device is to receive data, and with them and with the help of them, the principles and algorithms of the device’s operation are controlled or adjusted, then There are changes being made to the device software itself. And since the data received can be dangerous, it is important to carry out cybersecurity controls. In turn, the wireless update module itself allows you to protect the device from the most simple and difficult to control threats of physical access to the device software. Accordingly, the presence in the proposed device of both a wireless update module and a cybersecurity module together provide a significant increase in the reliability and security of both the device itself and the examination process, and at the same time simplify the maintenance of the device, allowing it to be maintained in proper condition.

The combination of a cybersecurity module, a wireless update module, a diagnostic module and protection from external interference in the device provides the highest level of reliability and security of both the device itself and the examination process, allowing you to control both physical and software methods of influence.

Emergency storage module

Emergency data storage module 11 is designed to restore data in the event of a power failure or other emergency situations that could result in the examination process being interrupted. Because video and other data are streamed, processed, and transmitted during the exam, the files being processed can be relatively large in size, and failures or abnormal situations may result in the loss or corruption of these files.

The emergency data storage module 11 is synchronized with the permanent memory module 10 and stores in real time critical data about the process (including the exam protocol), as well as information about which file the permanent memory module 10 is currently working with, and the status of this file. The hardware implementation of the emergency data storage module 11 is a separate chip of conditionally non-erasable memory - for example, a flash memory bank containing its own controller, which does not allow erasing operations from inside the vehicle and from the outside.

Emergency data storage module 11 provides additional protection for both the inspector and the examinee, performing a function similar to a “black box”. If necessary, the data recorded on this module can help restore all or part of the data of the permanent memory module 10 if it was lost or damaged for any reason, or at least determine the circumstances under which this happened.

Thus, the presence of an emergency data storage module in the proposed device increases the reliability and safety of both the device itself and the examination process.

Power module

Power module 12 is necessary due to the fact that the vehicle’s own on-board network is extremely unstable and is subject to serious voltage surges and sags, especially when starting the engine and all standard vehicle systems. These processes can take quite a long time, causing the proposed device 100 to be constantly rebooted, which can negatively affect its performance. In addition, in the on-board network, a long-term drop in supply voltage is possible due to insufficient battery charge or too high current consumption, which can sharply reduce the performance of the proposed device 100, leading to the impossibility of its normal operation and, consequently, to a violation of the exam procedure.

The power module 12 contains a power controller designed to control the power in the on-board network of the vehicle and supply power from a backup power source to the modules of the device 100, if the power in the on-board network is lower than required for the correct conduct of the exam, or smoothing/cutting off surges, if the power in the on-board network is higher than permissible.

In the preferred embodiment, power module 12 includes a 1-DIN maximum supercapacitor as a backup power source so that device 100 remains fairly compact while still being able to house all modules in a single housing for increased reliability and safety. Taking into account the required performance of the device 100 and the range of tasks it performs, the capacity of a supercapacitor of this size may be sufficient to maintain the operation of the device 100 for up to 1 minute even in the complete absence of power in the on-board network (for example, in the event of an accident), which may allow the device 100 (or at least to the emergency data storage module) correctly terminate all processes and save data before shutting down. In the case of small voltage drops, the duration of maintaining performance increases significantly.

In another embodiment, backup power may be provided from an additional battery. In this case, the battery is located outside the device, and the power module 12 within the body of the device 100 contains only a controller and a power bus.

The presence of a power module in the proposed device increases the reliability and safety of both the device itself and the examination process.

operating system

The proposed device contains a wide variety of hardware and hardware-software modules. During the life cycle of a device, many different programs run, involving many different aspects of working with data. Accordingly, to organize all processes in the device, an operating system is installed that is capable of servicing the device, taking into account all its peripherals and available functionality. In one non-limiting example, a Linux OS may be used.

Typical exam scenario

Next, a typical exam scenario, which is the main function of the proposed device 100, will be described.

A map of the area where the exam will be held is drawn up in advance. The card format can vary depending on the requirements of the specific application. In one non-limiting example, the maps are in HD map format.

The map shows traffic lanes, coordinates and contents of signs, traffic lights, pedestrian crossings and other road elements that must be taken into account when drawing up routes and when conducting the exam. Required areas are assigned appropriate restrictions. Coordinates are usually clarified manually by the operator. However, in one embodiment, it is possible to use the proposed device 100 and its functionality to determine coordinates and assist in marking maps.

Exam routes are generated by the user/system operator or automatically by the system, taking into account the need to test the maximum possible or predetermined threshold number of skills and errors (or points of regulations) during the time allotted for the exam, and, if necessary, taking into account the minimum repetition of exercises, tasks and maneuvers. Optionally, other factors can be taken into account, such as the average traffic congestion in a given area at this time of day and day of the week, the average speed of traffic in this area, the number of accidents in this area over the last period of time, predicted weather conditions, the quality of the road surface, the results of previous exams in a given area, the presence of visible road markings, etc. With regard to visible road markings, it should be noted that if the test is conducted on a road where the markings are not visible, the lane keeping requirements may be reduced for candidates. The appropriate configuration of the device 100 for the upcoming exam is performed by the user or operator. If necessary, the generated route (or several route options) is approved by responsible persons - for example, signed with an electronic signature.

The approved route (or set of routes) is automatically loaded into device 100 via wireless update module 8 or manually by an inspector via a flash drive. Also, maps can be updated in device 100 before the exam and coordinates can be corrected.

The driver candidate gets into the vehicle, an interior camera identifies him, and the device or system gives him access to the test. The inspector sits nearby, launches the exam in an application on his tablet and gives the command to the driver to start driving.

While driving, audio and video recordings are made, and the entire exam process is monitored using the modules described above. When formalized driving errors occur, they are recorded in the protocol automatically by the device, unformalized errors are marked in the tablet by the inspector. When approaching the initiating points, module 2 of acceleration of neural networks is launched in priority mode and then processed. The necessary data is sent to a remote server directly during the exam.

After the exam is completed, the protocol is signed by the candidate and the inspector and sent to a remote server.

Frame

If possible, all blocks of the proposed device are enclosed in a single housing, if the requirements of a specific implementation allow this.

In one embodiment, the body of the proposed device is 2-DIN in size, so as to increase the speed and ease of installation in a vehicle - in particular, on the center console. The state of the art has not identified such compact, fully functional secure devices for conducting the practical part of the exam for the right to drive vehicles.

In other embodiments, the body of the proposed device may have different dimensions and/or be installed in other positions - for example, in a difficult-to-reach position under the hood of a car near the center console.

If necessary, control and/or input/output means can be located on the device body - for example, a display, keyboard, speaker, connector for a USB drive. This implementation improves usability. Controls and/or input/output can be protected from accidental or malicious actuation/touching - for example, by using a transparent or opaque casing.

In other embodiments, to enhance reliability and security, no controls and/or I/O controls may be located on the chassis, or only a power button and power indicator may be located.

Way

Next, the vehicle control monitoring method according to the present invention will be described. A special case of this method is the method of conducting an exam for the right to drive vehicles.

It should be understood that this method corresponds to the above-described functions performed by each of the blocks in the device 100, and if any information is not disclosed in relation to the method, but is disclosed in relation to the device, and vice versa, it does not imply that in the device or method this the function or step cannot be performed, but is done only to avoid cluttering the description with repetition of details.

The method for controlling the control of a vehicle as a whole contains a stage at which a map and a pre-marked route for the vehicle are loaded into the device memory.

Additionally, the method includes identifying the driver based on data entered by the user or based on recognition by peripheral components of the device.

Additionally, the method contains a step at which the correctness of the task is assessed based on the processing of data received by the device from peripheral modules.

Additionally, the method includes the step of determining the location of a given vehicle based on the received navigation data.

Additionally, the method includes a step in which the vehicle's approach to the initiating point or section on the route is determined and the neural network acceleration module is launched to process the event corresponding to this point or section using a pre-trained neural network.

Additionally, the method includes a step where data on the progress of the exam is stored in the permanent memory module of the device and data on the progress of the exam is transmitted to a remote server.

Additionally, the method includes a stage where, synchronously with the permanent memory module, data about the exam process and information about which file the permanent memory module is currently working with, and the state of this file are stored in real time, using an emergency data storage module containing conditionally indelible memory.

Additionally, the method includes updating the software and operating data of the device wirelessly from a remote server.

Additionally, the method contains a stage at which software processes and/or data elements are monitored for unreliability or the presence of threats to protect the device from hacking using programming methods using a cybersecurity module.

Additionally, the method contains a stage in which, using a diagnostic module and protection against external interference, the internal modules and/or peripheral modules of the device are monitored to determine their operability and the correctness of the output signals and/or data, and the signals from sensors responsible for the safety of the device are monitored.

Additionally, the method contains a stage in which, using the power module, the power in the on-board network of the vehicle is controlled and power is supplied from a backup power source to the device modules if the power in the on-board network is lower than required for the correct conduct of the exam, or smoothing out power surges if the power in the on-board network is higher than permissible.

The proposed method for monitoring vehicle control provides full control of vehicle control within the framework of a predefined list of exercises, maneuvers and checks with increased safety, reliability and reduced costs, while reducing the human factor, increasing the transparency of the exam and grading and increasing the level of confidence in the exam and to the examiner. The examination procedure is simplified for the examiner himself, since he needs to perform fewer actions, and most of his actions and decisions are confirmed (supported) by the proposed method.

Application

The devices, systems and methods according to the present invention can be used to control the control of a vehicle in general and to conduct the practical part of the driving test in particular.

Additional implementation features

The wireless communication module and the wired communication module may include a transmitter and a receiver. The transmitter and receiver can physically be one unit or different units.

The various illustrative blocks and modules described in connection with the disclosure herein may be implemented or executed by a general purpose processor, digital signal processor (DSP), application specific integrated circuit (ASIC), field programmable gate array (FPGA), or other programmable logic device (PLD), discrete logic gate or transistor logic, discrete hardware components, or any combination of the foregoing, designed to perform the functions described herein. The general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices (eg, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors plus a DSP core, or any other similar configuration).

Some blocks or modules, individually or collectively, may constitute, for example, a computer, and include a processor that is configured to call and execute computer programs from memory to perform method steps or functions of the blocks or modules in accordance with embodiments of the present invention. According to embodiments, the device may further include memory. The processor may call and execute computer programs from memory to perform the method. Memory may be a separate device, independent of the processor, or may be integrated into the processor. The memory may store code, instructions, commands and/or data for execution on a set of one or more processors of the described device. Codes, instructions, or commands may direct a processor to perform steps of a method or device function.

The functions described herein may be implemented in hardware, software executed by one or more processors, firmware, or any combination thereof. The hardware and software that implement the functions may also be physically located in different locations, including according to a distribution such that portions of the functions are implemented in different physical locations, that is, distributed processing or distributed computing may be performed.

If the volume of data is large, multi-threaded data processing can be performed, which in a simple representation can be expressed in the fact that the entire set of data to be processed is divided into a set of subsets, and each processor core performs processing in relation to the subset of data assigned to it.

The above memory may be volatile or non-volatile memory, or may include both volatile and non-volatile memory. One skilled in the art will also understand that when talking about memory and the storage of data, programs, codes, instructions, commands, etc., it is meant to be a machine-readable (or computer-readable, processor-readable) storage medium. Computer readable media includes both non-transitory computer storage media and communication media, which includes any transmission medium that facilitates movement of a computer program or portion thereof from one location to another. A non-transitory computer-readable storage medium may be any available medium that can be used to carry or store desired program code media in the form of instructions or data structures, and that can be accessed by a computer, processor, or other general purpose or special purpose processing device. appointments.

By way of example, and not limitation, computer-readable media may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electronically erasable programmable read-only memory (EEPROM), flash memory , random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM) , Enhanced Speed Synchronous Dynamic Random Access Memory (ESDRAM), Synchronous Line DRAM (SLDRAM) and Direct Access Random Access Memory (DR RAM), register, cache memory, semiconductor memories, magnetic media such as internal hard drives and removable drives, magneto-optical media and optical media such as CD-ROMs and digital versatile disks (DVDs), as well as any other storage media known in the art.

The information and signals described herein may be represented using any of a variety of different technologies. For example, the data, instructions, commands, information, signals, bits, symbols and chips that may be exemplified in the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combinations of the above.

At least one of the steps in the method or blocks/modules in the device may use an artificial intelligence (AI) model to perform corresponding operations. The AI-related function may be performed through a processor and non-volatile and/or volatile memory.

The processor may include one or more processors. At the same time, one or more processors may be a general purpose processor such as a central processing unit (CPU), an application processor (AP), etc., a graphics processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU) ), and/or a specialized AI processor such as a neural processing unit (NPU).

One or more processors control the processing of input data according to a predefined operating rule or artificial intelligence (AI) model stored in non-volatile memory and/or volatile memory. A predefined operating rule or artificial intelligence model can be obtained through training. In this case, the processor can perform a pre-processing operation to convert the data into a form suitable for use as input to an artificial intelligence model.

“Learned” means that by applying a learning algorithm to a set of training data, a predefined operating rule or AI model with a desired characteristic is created. The training may be performed on the device itself on which the AI is running according to the embodiment, and/or may be implemented through a separate server/system.

An artificial intelligence model may include many layers of neural network. Each of the plurality of layers of the neural network includes a plurality of weights and performs an operation for that layer by computing with respect to the computation result of the previous layer and the plurality of weights.

Examples of neural networks include, but are not limited to, convolutional neural network (CNN), deep neural network (DNN), recurrent neural network (RNN), restricted Boltzmann machine (RBM), deep belief network (DBN), bidirectional recurrent deep neural network (BRDNN) ), generative adversarial networks (GANs), and deep Q-networks.

A learning algorithm is a method of training a predefined target device (such as an NPU-based neural network) using a set of training data to invoke, enable, or control the target device to perform a determination or prediction. Examples of learning algorithms include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

It should be understood that although terms such as "first", "second", "third" and the like may be used herein to describe various elements, components, regions, layers and/or sections, these elements, components , areas, layers and/or sections should not be limited to these terms. These terms are used only to distinguish one element, component, area, layer or section from another element, component, area, layer or section. Thus, a first element, component, region, layer or section may be referred to as a second element, component, region, layer or section without departing from the scope of the present invention. As used herein, the term “and/or” includes any and all combinations of one or more of the respective items listed. Elements referred to in the singular do not exclude the plurality of elements unless specifically stated otherwise.

The functionality of an element specified in the description or claims as a single element may be implemented in practice by means of several components of the device, and conversely, the functionality of elements specified in the description or claims as several separate elements may be realized in practice by means of a single component.

In one embodiment, the elements/blocks/modules of the proposed device are located in a common housing, can be placed on the same frame/structure/printed circuit board/chip and are connected to each other structurally through installation (assembly) operations and functionally through communication lines. The mentioned lines or communication channels, unless otherwise indicated, are typical communication lines known to specialists, the material implementation of which does not require creative efforts. The communication line can be a wire, a set of wires, a bus, a track, a wireless communication line (inductive, radio frequency, infrared, ultrasonic, etc.). Communication protocols over communication lines are known to those skilled in the art and are not disclosed separately.

The functional connection of elements should be understood as a connection that ensures the correct interaction of these elements with each other and the implementation of one or another functionality of the elements. Particular examples of functional communication may be communication with the ability to exchange information, communication with the ability to transmit electric current, communication with the ability to transmit mechanical motion, communication with the ability to transmit light, sound, electromagnetic or mechanical vibrations, etc. The specific type of functional connection is determined by the nature of the interaction of the mentioned elements, and, unless otherwise indicated, is provided by widely known means, using principles widely known in the art.

The design of the elements of the proposed device is known to those skilled in the art and is not described separately in this document unless otherwise indicated. The elements of the device can be made of any suitable material. These component parts can be manufactured using known methods including, by way of example only, machining, investment casting, and crystal growth. The assembly, connection and other operations in accordance with the above description are also within the knowledge of a person skilled in the art and, therefore, will not be explained in more detail here.

Although exemplary embodiments have been described in detail and shown in the accompanying drawings, it is to be understood that such embodiments are merely illustrative and are not intended to limit the present invention, and that the invention should not be limited to the specific arrangements and structures shown and described, since Various other modifications and embodiments of the invention may be apparent to one skilled in the art based on the information set forth in the specification and knowledge of the prior art without departing from the spirit and scope of the present invention.

Claims (14)

1. A method for tracking the actions of a driver during an exam for the right to drive vehicles, performed in a system containing an on-board device for conducting an exam for the right to drive vehicles, as well as a remote server and/or a portable device of the examiner, containing the stages at which: before the start of the exam, programs and data of the on-board device are loaded into the permanent memory of the on-board device, including a map of the area and a pre-marked route for the examination, containing pre-defined initiating points and/or sections on the route; identifying the driver based on data entered by the user or based on recognition using peripheral components of the on-board device; provide the driver with access to the exam using a remote server and/or the examiner’s portable device; launch the examination process using a remote server and/or an examiner’s portable device; during the examination process, signals and/or data from peripheral components of the on-board device are received using a wired communication module of the on-board device, wherein the peripheral components of the on-board device contain at least one camera; storing signals and/or data from peripheral components of the on-board device, including video recordings from said at least one camera, in the permanent memory of the on-board device; using the navigation module of the on-board device, determine the location of the vehicle on the terrain map based on the received navigation data; using the wireless communication module of the on-board device, they exchange data on the progress of the exam with a remote server and/or with a portable device of the examiner; determining, using the on-board device processing module, the device's approach to a predetermined initiating point or section on the route; launch the acceleration module of the neural networks of the on-board device to process an event corresponding to a given point or section using a pre-trained neural network, wherein the initiating point or section on the route and the corresponding event are associated with checking the correctness of the execution of the exercise, task or maneuver by the candidate driver being examined in the process conducting an exam; And using the on-board unit processing module, a remote server and/or the examiner's portable device, an assessment is made of the driver's correct performance of exercises, tasks or maneuvers along the exam route based on processing data received from peripheral modules and from the on-board unit navigation module. 2. The method according to claim 1, additionally containing the stages in which, after the end of the exam, a protocol for the examination is formed, containing indications of errors and circumstances of the examination, data confirming the signature of the driver and the examiner is received, and the protocol is sent to a remote server. 3. The method according to claim 2, in which the indications of errors and circumstances of the examination contain data on the progress of the examination, including the results of assessing the correctness of the driver’s performance of exercises, tasks or maneuvers received by the on-board device, and data on the progress of the examination entered by the examiner using the examiner's handheld device.