JPH10201727A - Apparatus for prevention of doze - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable prevention of doze for a driver of a motorcycle. SOLUTION: On this apparatus, a sensor 11 mounted at the temple part in a helmet of a driver of a motorcycle detects a brain wave of the driver and observes the body condition of the driver. The output of the sensor 11 is input into an apparatus for detecting abnormality of brain wave 12 and brain wave of the driver is observed always. The apparatus for detecting abnormality of brain wave 12 detects an abnormal brain wave as an emergency signal when doze occurs and outputs it. The detected output is treated by an emergency alarm treating unit 13 and it is sent to an alarm sound generator 14 such as a speaker or an earphone built in the helmet, the alarm sound generated in the helmet gives a warming of the doze to the driver.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drowsiness prevention apparatus which applies a deterministic nonlinear system (chaos) to prevent drowsiness while a motorcycle is running.

[0002]

2. Description of the Related Art In recent years, various devices have been developed as devices for preventing drowsiness in automobiles. In that kind of device,
In some cases, a facial expression of a driver is photographed by a CCD camera installed in a vehicle cabin, and an alert is issued when a sleeping expression (for example, a closing expression) is detected.

[0003]

However, the application of the above-mentioned doze prevention device to a motorcycle is known as a CCD.
Difficult because of the need to install a camera. Also, CCD
Even if the facial expression of the driver is photographed by the camera, it is difficult to identify whether the facial expression is dozing because the facial expression varies from person to person. In particular, in a motorcycle, it is difficult to install a CCD camera, so that the facial expression of the driver cannot be captured.

[0004] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a drowsiness prevention device capable of preventing drowsiness even for a motorcycle driver.

[0005]

SUMMARY OF THE INVENTION In order to achieve the above object, a first aspect of the present invention is to provide a sensor which is attached to a predetermined position in a helmet and detects an electroencephalogram, and an electroencephalogram detected by the sensor. Is input, an electroencephalogram abnormality detection device that detects whether there is an abnormality in the input electroencephalogram, and when there is an abnormality in the electroencephalogram observed by this electroencephalogram abnormality detection device,
An abnormal alarm processing unit for processing an output from the detection device, and an alarm sound generating unit for generating an alarm sound in the helmet based on the output from the processing unit.

According to a second aspect of the present invention, in the electroencephalogram abnormality detecting device, a time-series file in which data as an electroencephalogram is input from a data input unit also serving as a sensor, a time-series data short-term prediction unit serving also as a data file, and the time-series file When the data is supplied from, after appropriately optimizing the parameters, the parameter optimization unit that inputs the optimized parameter data to the time-series data short-term prediction unit, and the time-series data short-term prediction unit A prediction result of n steps is performed based on the obtained data and parameters, and the prediction value is supplied to the parameter optimization unit, and a prediction value file to be filed and a prediction value of the prediction value file are supplied. And a prediction result extraction unit to be executed.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing an embodiment of the present invention. In FIG. 1, reference numeral 11 denotes a sensor attached to a temple portion inside a helmet of a motorcycle driver. To detect the driver's condition. The output of the sensor 11 is input to an electroencephalogram abnormality detection device 12 whose details will be described later, where the electroencephalogram of the driver is constantly observed. The electroencephalogram abnormality detection device 12 detects an abnormal electroencephalogram at the time when dozing starts as a danger signal, and outputs it. The detection output is sent to the abnormality alarm processing unit 13.
And is sent to an alarm sound generating unit 14 such as a speaker or an earphone in the helmet to generate a sound in the helmet to notify the driver that he is dozing.

Next, a detailed block of the electroencephalogram abnormality detecting device 12 will be described with reference to FIG. In FIG. 2, reference numeral 1 denotes a short-term prediction system unit serving as a main part of the electroencephalogram abnormality detection device 12, and 2 denotes a time-series data short-term prediction unit (time-series data storage unit and prediction unit) which also serves as a data file unit. Reference numeral 3 denotes a data input unit serving also as a sensor 11 attached to a temple portion inside the helmet shown in FIG. 1. Data (encephalogram) input from the data input unit 3 is sent to the time-series file It is input to the series data short-term prediction unit 2 or the parameter optimization unit 5. When data (encephalogram) is input to the parameter optimizing unit 5, the parameter optimizing unit 5 appropriately optimizes parameters and then inputs the optimized parameter data to the time-series data short-term prediction unit 2. .

The time-series data short-term prediction section 2 performs a prediction result n steps ahead based on input data (eg, brain waves) and parameters. The prediction value is sent to the parameter optimization unit 5 and the prediction result extraction unit 7 after being sent to the prediction value file 6. The parameter optimizing unit 5 optimizes parameters based on the input time-series data in addition to the time-series data.

[0010] The prediction result extraction unit 7
The prediction result is extracted from the above and input to the abnormality alarm processing unit 13. The abnormality alarm processing unit 13 compares the predicted value with the actually measured value, determines whether or not an abnormality has occurred based on a predetermined criterion, and causes the speaker in the helmet to sound an alarm sound indicating dozing through the alarm sound generation unit 14.

Next, the operation of the short-term prediction system unit 1 will be described. First, through the time series file 4, the time series data y (t), y (t−τ), y (t−2τ), which are brain waves, are sent from the data input unit 3 also serving as the sensor 11 to the time series data short-term prediction unit 2. (Τ
Is a time delay). In the time-series data short-term prediction unit 2 reconstructs the n-dimensional reconstructed state space R ⁿ based on time-series data is input, obtains a predicted value after s steps by reproducing the attractor of the observation system , And compares the predicted value with the actual observed value after s steps. FIG. 3 is an explanatory diagram showing a relationship between a predicted value and an observed value.

As shown in FIG. 3, when an abnormality is determined, an upper limit value and a lower limit value are set based on a predicted value, and an area between the upper limit value and the lower limit value is determined by an observation system. Is a normal area determined to be normal. Other areas are regarded as abnormal areas.

For example, a predicted value after s steps at time ta is obtained. This predicted value is indicated by point a in FIG.
The normal region set from point a is indicated by the solid line at time ta-s + s in FIG. Since the actual observation value at t−a + s falls within the normal region, the observation value at this point is determined to be in an abnormal state.

On the other hand, the predicted value after s steps obtained at time t is shown by a point b in FIG. 3, and a normal region set from the point b is shown by a solid line portion at time t + s in FIG.
Since the actual observation value at t + s falls within the abnormal region, the observation value at this point is determined to be in an abnormal state.

The time-series data short-term prediction section 2 embeds the normal time series to reconstruct the strange attractor, and also records the embedding parameters at this time. When a system perturbation occurs, the prediction error increases. When the prediction error exceeds the set threshold, the parameter optimizing unit 5 causes the parameter to be re-optimized. By checking the change of “delay” among the parameters obtained as a result, it can be seen that “abnormality” is beginning. However, the width of the perturbation that becomes “abnormal” / the “delay” of the embedding parameter differs depending on the system to be applied.

In the short-term chaos prediction method of the prediction section, a time series is embedded to reconstruct a strange attractor.
By checking the shape and density of the attractor, you can grasp the state of the system. Taking the time series of brain waves as an example, the shape and density of attractors differ depending on the psychological state.

[0017]

As described above, according to the present invention,
It is possible to prevent the driver from falling asleep during the running of the motorcycle, thereby obtaining an advantage that an accident due to the falling asleep can be prevented.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram showing details of an electroencephalogram abnormality detection device.

FIG. 3 is an explanatory diagram showing a relationship between a predicted value and an observed value.

[Explanation of symbols]

 11: Helmet sensor 12: EEG abnormality detection device 13: Abnormal alarm processing unit 14: Alarm sound generation unit

Claims (2)

[Claims] 1. A sensor attached to a predetermined position in a helmet and detecting an electroencephalogram, an electroencephalogram abnormality detection device to which an electroencephalogram detected by the sensor is input and observing whether or not the input electroencephalogram is abnormal. An abnormal alarm processing unit that processes an output from the detection device when an electroencephalogram observed by the electroencephalogram abnormality detection device has an abnormality; and an alarm sound generation that generates an alarm sound in the helmet based on an output from the processing unit. And a drowsiness prevention device comprising: 2. An electroencephalogram abnormality detecting device comprising: a time-series file to which brain wave data is input from a data input unit also serving as a sensor; a time-series data short-term prediction unit serving also as a data file; A parameter optimizing unit for inputting data of the optimized parameter to the time-series data short-term prediction unit after appropriately optimizing parameters when supplied, and the time-series data short-term prediction unit The prediction result of n steps is performed based on the parameters and the parameters, and the prediction value is supplied to the parameter optimizing unit, and the prediction value file to be filed and the prediction value to which the prediction value of the prediction value file is supplied The drowsiness prevention device according to claim 1, further comprising a result extraction unit.