JPH0478622A - Asleep driving prevention device - Google Patents

Abstract

PURPOSE:To certainly prevent asleep driving by fuzzy-inferring from a driving state of a driver whether the driver is in the state of sleeping or not, thus correctly detecting that the driver falls in the state of sleeping and awakening the driver on the basis of this detection. CONSTITUTION:Sensors S1, S2 to detect the gripping force of a driver are pro vided on a steering wheel 1 and a shaft lever 2 which are operational members of a car. Additionally, sensors S3, S4 to detect the contact force of the back and the elbow of the driver are provided on a driver's seat 3 and an arm rest 4. Furthermore, output of each of the sensors S1-S4 is input to a fuzzy inference device 6. Thereby, publicly known fuzzy inference computation is carried out by the fuzzy inference device 6 in accordance with a previously specified fuzzy rule. Subsequently, output of the fuzzy inference device 6 is input to an alarm device 5 stored in a center console, so that an alarming to urge the driver to wake up is generated.

Description

Detailed Description of the Invention (a) Industrial Application Field The present invention is provided near the driver's seat of a vehicle, detects the driver's drowsy state and operates an alerting device such as an alarm device to prevent drowsy driving. This invention relates to a device for preventing drowsy driving.

(b) Conventional Technology Drowsy driving, in which a driver falls asleep while driving a vehicle, can cause serious accidents. For this reason, a device has been proposed that prevents the driver from falling asleep by sounding an alarm or applying vibrations to wake the driver when the driver falls asleep. In this case, the problem is how to detect that the driver has fallen into a dozing state. Conventionally, a brain wave measurement device is installed, and when the brain waves of a sleeping state are detected in the measurement results, it is determined that the driver is in a dozing state. Some devices determine that the driver has fallen into a sleep state, while others require the driver to wear special glasses and detect whether the driver has fallen into a sleep state based on the opening and closing of the eyelids.

(C) Problems to be Solved by the Invention However, with conventional drowsy driving prevention devices, it is impossible to detect brain waves in a resting state using an on-vehicle brain wave measurement device, considering that the vehicle is constantly vibrating while driving. However, it is difficult to accurately detect brain waves in a sleeping state when a person falls into a dozing state, making it difficult to put it into practical use. In addition, the method of wearing a special mask is complicated to wear, and drivers cannot be required to wear it.
Similarly, it must be said that it has little practicality.

The purpose of this invention is to detect the driving state of the driver based on the grip force on the steering wheel and the pressure applied to the seat, and to determine whether the driver is dozing off by fuzzy inference using this driving state as an input variable. To provide a drowsy driving prevention device that can accurately detect when a driver falls asleep without requiring a complicated device or requiring the wear of complicated members. .

(d) Means for Solving the Problem The drowsy driving prevention device of the present invention includes a single or plural sensors attached to an operating member and/or a driver's seat, and detects the operating state of the operating member and/or driving posture. A driving state detecting means for detecting a driving state, a fuzzy inference means for performing a fuzzy calculation using a detection result of the driving state detecting means as an input variable according to a preset fuzzy rule, and an awakening operation according to the calculation result of the fuzzy inferring means. The present invention is characterized by providing an awakening promotion means to perform the following.

fe) Effect In the present invention, the operation state and driving posture of the driver are detected by sensors attached to the operating member or the driver's seat, and the detection results are inputted as human variables to the fuzzy inference means. The fuzzy inference means performs fuzzy calculations according to preset fuzzy rules and outputs the extent of the driver's dozing state as an inference result. This inference result is input to the wake-up promotion means, and the wake-up promotion means executes a wake-up operation according to the degree of the dozing state. Therefore, whether or not the driver is dozing off can be comprehensively determined from the operating state of the operating members and the driving posture, and there is no need to install complicated devices or complicated members.

if) Embodiment FIG. 1 is a block diagram showing the configuration of a drowsy driving prevention device which is an embodiment of the present invention.

A handle 1 and a shift lever 2, which are operation members of a vehicle.
Each is equipped with a sensor S and S2 that detect the grip force of the driver. Further, the seat 3 and armrest 4 of the driver's seat are each equipped with sensors S3 and S4 that detect contact forces of the back and elbows of the driver. The sensor 9'SI, s2 detects the handle position and the operating state of the shift lever 2, and the sensor S3. S4 detects the driving posture.

The outputs of these sensors S, ~S4 are input to the fuzzy inference device 6. This fuzzy inference device 6 executes a known fuzzy inference operation according to predetermined fuzzy rules, and may be constructed of either an analog circuit or a digital circuit. The output of the fuzzy inference device 6 is input to an alarm device 5 housed in the center console. The warning device 5 generates a warning to urge the driver to wake up in response to the input from the fuzzy inference device 6.

2 and 3 are diagrams showing membership functions of the antecedent part and the consequent part in the fuzzy inference device 6.

According to the empirical rule, the fuzzy inference device 6 calculates, for example, if (x+=NL or X2=NL or x,=
NL or x4. =PL) then y, =PL if (x, =NM or X2=NM or x, =
NM or X4=PM) then y, two PM if (x+=NS Or xz=NS or X3=
Fuzzy rules such as NS or x4=PS) then y,=PS are set in advance. In each of these fuzzy rules, the input variables x1 to x of the antecedent part
The degree of affiliation and label of 4 are determined from the membership function shown in Figure 2, the fuzzy rule corresponding to the combination of labels of the input variables in the antecedent part is selected, and the fuzzy rule is selected in the membership function of the consequent part. The label of the consequent part is truncated at the minimum value of the layer to which the antecedent part belongs. This head truncation of the membership function of the consequent part is executed for all of the selected fuzzy rules, and the center position of the truncated consequent membership function is output to the alarm device 5 as the calculation result Y of the fuzzy inference. Ru.

The above fuzzy rule determines the grip force on the steering wheel 1 and the shift lever 2 as the driver's drowsiness increases.
This setting is based on the assumption that the gripping force on the handlebar 1 will decrease, the pressure on the seat 3 will decrease, and the pressure on the armrest 4 will increase.
When the gripping force of 2 and the pressure on the sheet 3 decrease,
An alarm is generated from the alarm device 5 in accordance with the lowering state. Further, as the pressing force on the armrest 4 increases, an alarm is generated with a strength corresponding to the degree of increase.

In this embodiment, an alarm device that sounds an alarm is used as the awakening promoting means, but other devices such as a ventilation device and a seat vibration device can also be used. Further, a plurality of awakening promoting means may be provided, and the awakening promoting device operated may be changed depending on the degree of the inference result of the fuzzy inference device.

(g) Effects of the Invention According to the present invention, it is possible to determine whether or not the driver is dozing off based on the driver's driving condition, so there is no need for a complicated device for measuring brain waves. It is possible to accurately detect when a driver has fallen asleep without requiring the driver to wear special glasses, etc. Based on this detection, the wakefulness promoting means is activated to wake the driver, thereby reliably preventing dozing off while driving. There are advantages that can be achieved.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a drowsy driving prevention device that is an embodiment of the present invention, and FIGS. 2 and 3 show members of the antecedent part and the consequent part, respectively, of the fuzzy inference means of the drowsy driving prevention device. FIG. 3 is a diagram showing a ship function. 1-handle, 2-shift lever seat, 4-armrest, 1-alarm device (awakening promotion means), 1-fuzzy inference device, 1-S4-sensors (driving state detection means).

Claims (1)

[Claims] (1) A driving state detection means that includes a single or multiple sensors attached to the operating member and/or the driver's seat and detects the operating state and/or driving posture of the operating member, and according to preset fuzzy rules. A dozing device characterized by comprising: fuzzy inference means for performing fuzzy calculation using the detection result of the driving state detection means as an input variable; and awakening promotion means for performing an awakening action according to the calculation result of the fuzzy inference means. Driving prevention device.