RU2539004C1 - Method for controlling control circuit logoff by live operator - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: EEG frequency response is pre-determined by a period analysis in a highly-watchful live operator being involved in active visual-motor activities and in actual activities. Theta-, alpha- and gamma-wave count per second is measured. The derived values are compared. If the theta-, alpha- and gamma-wave count per second is stated to fall outside the values specific for high watchfulness, an EEG analyser converts a sequence of the values into a sensory signal and feeds a biofeedback-like warning signal automatically to state the control circuit logoff by the live operator.

EFFECT: method improves the accuracy of the control measurements enabling feeding the sensory signal in due time that is ensured by pre-determining a combination of highly-watchful EEG theta-, alpha- and gamma-wave frequencies.

1 dwg, 2 ex

Description

The invention relates to medicine, in particular to occupational health, and to ergonomics and can be used to control the exit of a human operator from the control loop.

Maintaining high efficiency, vigilance and error-free activity of operators in human-equipment systems is an extremely urgent task due to the fact that, according to statistics, from 30 to 80 percent of industrial accidents and disasters, traffic accidents and traffic accidents occur through the fault of the "human factor". Moreover, the most common cause of such incidents is a decrease, distraction, or loss of visual attention, that is, episodes (including short-term) of an involuntary and unconscious exit of a human operator from the control loop of a dynamic object.

Known methods and technical devices for monitoring and maintaining vigilance, based on the periodic (every 15-90 seconds) giving the operator (driver, driver) a test signal to which he must quickly respond with a specific action (by pressing a button, turning the handle, etc. P.). In the absence of such a response, the device gives an alarm and / or changes the parameters of a dynamic object (for example, turns on the braking mode) [L. Ventsevich Locomotive devices to ensure the safety of train traffic and decryption of information data of their work. - M .: Route, 2006. - 328 p .; WO 00/55000, 09/21/2000]. The disadvantages of these methods are the relatively low speed (at least 20 seconds), that is, the inability to determine short-term (of the order of units of seconds) outputs of the human operator from the control loop, as well as additional information load on the operator, which, in addition, distracts him from his main activity .

There is another group of methods and devices for monitoring and maintaining vigilance, based on continuous registration (monitoring) and automatic measurement of various physiological parameters of a human operator directly in the process. When the parameters of these indicators go beyond some experimentally determined limits, the operator is given a signal warning him of the onset of a “dangerous” functional state, and also (optionally) the parameters of the controlled dynamic object may change [Rail Safety and Standards Board's (RSSB) overview response to the report by Quintec entitled 'Driver vigilance devices: systems review' 03 T024 Quintec 22 RPT Final Report Issue 01. 30th August 2002, Railway Safety, GB]. These methods do not create additional and distracting loads on the operator. However, the accuracy, speed and noise immunity of these methods fundamentally depends on the nature of the recorded physiological indicators.

So, when using parameters of skin-galvanic reactions and / or electric skin resistance as an informative indicator, at least 15-20 seconds are required to determine the onset of a “dangerous” drowsiness of the operator. Therefore, short-term (of the order of units of seconds) episodes of the exit of a human operator from the control loop by methods based on registration of electric skin indices are not determined. Moreover, the state of “fluctuating vigilance” of the operator is regarded by such systems as the state of active activity. In addition, the parameters of electrodermal activity vary significantly not only in connection with a change in the functional state of the operator, but also depending on external conditions (primarily temperature and humidity) and on the nature of the operator's activity (frequency and intensity of control actions) [patent RU 2025731, 12/30/1994].

When used as an informative indicator of the level of vigilance of the parameters of the oculomotor system (closing the eyes, blinking frequency, speed of closing / opening of the eyelids, gaze direction and frequency of its switching) [patent RU 2423070, 04/10/2009; patent RU 2444275, 03/10/2012] significant limitations of the methods are: the operator wearing glasses (especially sun glasses), head turns, posture changes, vibration associated with vehicle features and the state of the road surface. The speed of these methods (due to the temporal characteristics of the recorded indicators) also does not allow the determination of short-term (of the order of units of seconds) outputs of the human operator from the control loop.

The best time resolution among physiological indicators (of the order of tens of milliseconds) is possessed by the parameters of the bioelectrical activity of the brain - electroencephalograms (EEG). In addition, it has been convincingly proved that the amplitude and frequency parameters of the EEG most reliably characterize the functional state of a person in the sleep-wake continuum, and some of them are directly related to visual attention. Until recently, the use of EEG as an informative indicator of the level of vigilance of a human operator in real-life conditions was difficult due to the low noise immunity of both recording equipment and traditional methods of EEG analysis. With the development of technology, some of these limitations have been overcome and several methods are now known for determining the state of a human operator (usually in terms of vigilance-drowsiness) by EEG parameters [patent WO 0044580, 03.08.2000; Program emulators on PC measuring devices; SmartCap Resource Internet, publ. 06/15/2010 http://www.radiomaster.net/news/view/699/].

The closest analogue of the present invention is a method for controlling human fatigue during visual tracking by recording the amplitude-frequency characteristics of the alpha-rhythm waves of the electroencephalogram of a human operator in the course of activity and determining the first signs of visual fatigue when registering three successive alpha-rhythm waves with an amplitude range from - 3 dB to +6 dB [SU 1090341 dated 05/07/1984].

However, this method has insufficient noise immunity, in addition, the appearance of fragments of the alpha rhythm can be associated not only with operator fatigue, which can produce “false alarms”.

The technical result of our proposed method of controlling the exit of the human operator from the control loop is to increase the accuracy of determining involuntary episodes of the exit of the human operator from the control loop of a dynamic object directly in the process.

The technical result is achieved by the fact that to monitor the exit of the human operator from the control loop, continuous recording and automatic analysis of his electroencephalogram directly in the course of activity is carried out in real time, while preliminary analysis of the frequency characteristics of the electroencephalogram is carried out experimentally in real time by periodometric analysis with determination the number of waves of theta, alpha and gamma activities of the brain per second in a high bite state identity of the human operator on the model of visual-motor activity, then the registration and analysis of the electroencephalogram of the operator is carried out directly in the process of real activity, with automatic comparative analysis of the values of the number of theta, alpha and gamma waves per second and when a combination of the values of the number of theta, alpha and gamma waves per second beyond the values characteristic of the state of high alertness of the human operator during active visual-motor activity, the EEG analyzer converts this data into sensor ny (acoustic) signal controlling automatic output state of the human operator of the control loop according to the type of biofeedback.

The method is as follows:

Example 1. In laboratory conditions, sensors (standard EEG electrodes) are installed on the subject’s head and connected to the input device of a hardware-software complex for recording and analyzing the electrical activity of the brain (digital electroencephalograph). Then, the subject is asked to perform a task that is a simplified model of operator activity, namely, to begin compensatory visual-motor tracking of the shift on the display screen of a point - a “target” controlled by a program-simulator of a dynamic tracking object, holding the “target” within the image of the “frame” using the manipulator (“mouse” or joystick). The shift of the "target" according to a random law is carried out by the program-simulator of the tracking object. When the “target” goes beyond the “framework”, the tracking object simulator program gives an error signal whose amplitude is proportional to the tracking error value. At the same time, the electroencephalogram is continuously recorded and automatically in real time (on-line) the frequency characteristics of theta, alpha and gamma-activity of the brain of the subject are analyzed using periodometric analysis according to the previously described algorithm [Chayanov N.V., Iznak A.F. An electronic device for automatically recognizing certain rhythmic patterns of an electroencephalogram. Biol. Sciences, 1979. - No. 3. - S. 93-98].

Theta activity is a bioelectric component representing potential fluctuations with a frequency of 4 to 8 Hz (with oscillation periods from 125 ms to 250 ms distributed randomly). Alpha activity is a bioelectric component representing potential fluctuations with a frequency of 8 to 12.5 Hz (with oscillation periods from 80 ms to 125 ms distributed randomly). Gamma activity is a bioelectric component representing potential fluctuations with a frequency lying in a band above 40 Hz (with oscillation periods of less than 25 ms distributed randomly). The ratios of theta, alpha, and gamma activities reflect certain functional states of the brain.

Automatically, in the state of high alertness of the human operator, during active visual-motor activity, the periods of electroencephalogram waves are measured every second and the number of theta, alpha and gamma waves is determined. It has been experimentally determined that the state of high alertness of the test operator during active visual-motor activity corresponds to a combination of less than two theta waves, less than three alpha waves and more than ten gamma waves per second. When the values of the number of theta, alpha and gamma waves per second go beyond these limits, the EEG analyzer generates a signal about the exit of the human operator from the control loop by the type of biological feedback.

The electroencephalogram of the test subject, the signal of the detector of the state of the test operator and the error signal are recorded on the hard disk of the hardware-software complex.

The graph (Fig. 1) illustrating the results of the experiment shows an example of the characteristic relationships between the functional state of the operator, determined by the combination of the number of waves of theta, alpha and gamma rhythms of the bioelectrical activity of the brain per second, and the quality of the activity - a signal of tracking error ( to simplify, this graph shows the deviation of the "target" only vertically).

In this case, for about 6 seconds from the start of the operator’s state detector, the number of theta waves ranged from 2 to 6, the number of alpha waves ranged from 3 to 10, and the number of gamma waves ranged from 2 up to 8 in each of 6 seconds, which led to the appearance of an operator exit signal from the control loop. At the end of the episode of the operator’s exit from the control loop, no more than one theta wave, no more than two alpha waves and from 15 to 30 gamma waves were determined in the EEG of the test operator.

The graph also shows that the signal of the operator’s state detector (exiting the control loop) is ahead of the tracking error signal (in this episode by about 2 seconds).

Moreover, the graph illustrates that during the episode of the operator’s exit from the control loop, his visual-motor coordination is impaired, as a result of which the operator’s control effect after the end of such an episode, which is designed to correct the tracking error that he suddenly realized, is not only disproportionate, but also incorrect direction. As a result, the tracking error at first sharply increases and only then becomes equal to zero. With regard to driving, this would mean that the driver drove for some time (in this episode about 6 seconds), ignoring and not responding to changes in traffic conditions, and then turned the steering wheel abruptly in the wrong direction, or braked sharply, or instead of braking pedals pressed the gas pedal.

Example 2. Under real driving conditions, an experienced driver operator puts on a hat (a protective helmet or a tennis ribbon), in which sensors (EEG electrodes) are mounted, connected to the input of a portable ("Holter" type) hardware-software complex for recording and analysis of theta, alpha and gamma activities of the brain. Preliminary analysis of the frequency characteristics of the operator’s electroencephalogram is carried out experimentally in real time by the method of periodometric analysis to determine the number of waves of theta, alpha and gamma-activity of the brain per second in the state of its high alertness on the model of visual-motor activity, then registration and automatic periodometric analysis electroencephalograms of the operator are carried out directly in the process of real activity - for example, in this case, when managing vtomobilem. The values of the number of waves of theta, alpha and gamma activities that go beyond the above experimentally determined limits are converted into a sensor (acoustic) signal, which automatically controls the state of the driver’s output from the control loop by the type of biological feedback. To do this, a miniature loudspeaker mounted in the headgear is used, through which a signal from the operator’s state detector is sent in the form of a sound tone, warning the driver about a dangerous exit from the car control loop.

According to the driver’s report, after several hours (4-5 hours) of continuous driving, both in urban traffic and on the intercity highway, warning sound signals of the operator’s state detector were issued only in situations when the driver unconscious left the control loop. This manifested itself, for example, in the form of a dangerous approach to a vehicle in front, in situations of distraction from driving a car (for example, reading a roadside advertisement, the landscape surrounding the road with the mental representation of missing details, tuning the radio, working with a cell phone or navigator), as well as with short-term “blinding” of the driver with bright light (during the day - when leaving the tunnel, at night - with the headlights of oncoming cars). In all these cases, the values of theta, alpha, and gamma waves per second went beyond the values characteristic of the state of high alertness of the driver during active hand-eye activity, and the driver noted short-term episodes of exit from the vehicle control loop, which indicates the absence of “false alarms”. Long exits from the control loop, dangerous from the point of view of the occurrence of an accident, were not noted, and, according to the driver, this was facilitated by warning signals from the operator's EEG-detector.

Thus, the proposed method allows you to determine the episodes of the involuntary exit of a human operator from the control loop of a dynamic object with high speed (no more than 1 second) and even before the appearance of a real activity error. It is important to note that such episodes are far from always associated with states of drowsiness or drowsiness.

The method is based on modern, experimentally proven scientific data on the functional value of the bioelectrical activity of the brain and the principles of processing visual sensory information [Guselnikov V.I., Iznak A.F. Rhythmic activity in sensory systems. M .: publishing house of Moscow State University, 1983]. As informative parameters, the results of automatic (in real time) analysis of the biopotentials of the human operator’s brain are used. To maintain a high level of vigilance of the human operator, the technique of acoustic EEG-mediated biological feedback is used.

The practical implementation of the method at the current level of biomedical electronic equipment and information and communication technologies is not a problem. When using the so-called "active" ("dry") EEG electrodes, the method becomes non-invasive, since it does not require the use of contact gels or liquids.

The method can be used in technical systems for determining and maintaining a high level of attention, vigilance and operability of operators controlling various vehicles (pilots, drivers, drivers), as well as operators of "critical" technologies (for example, dispatchers of transport and energy systems).

Claims (1)

A method of controlling the exit of a human operator from the control loop, including continuous recording and automatic analysis of its electroencephalogram directly in the process of real activity, characterized in that preliminary analysis of the frequency characteristics of the electroencephalogram is carried out experimentally in real time by periodometric analysis with the determination of the number of theta, alpha waves - and gamma-activity of the brain per second in a state of high alertness of the human operator with active motor-motor activity, then registration and automatic periodometric analysis of the electroencephalogram of the operator is carried out directly in the process of real activity, with automatic comparative analysis of the values of the number of theta, alpha and gamma waves per second, and when the values of the number of theta, alpha and gamma -waves per second beyond the limits of numbers characteristic of the state of high alertness of a human operator during active hand-eye activity, the EEG analyzer converts a combination of these numbers into weed signal, and supplies to the automatic warning signal for biofeedback type of output state of the human operator from the control circuit.

Images