JPH10250450A - Detecting method for existence or absence and posture of occupant in automobile - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce using quantity of a data storage area and to easily obtain seat image distance distribution data by arranging a plurality of pairs of optical sensor arrays in the diagonal front upper part of a seat so that at least two linear view fields include a seat backrest part and finding the position and the inclination of the seat from a distance value for the seat backrest end part. SOLUTION: An occupant sensor 1 is arranged in the diagonally front upper part of a seat 4 so as to set view fields R1, R2, and then, a seat backrest end part is observed. Firstly, from the installation coordinate position S of the sensor 1, coordinates of points P1, Q1 are found on the basis of a seat end part distance in a straight line SLP1 in the seat end part, and then, an inclination angle of the seat 4 is found from an inclination of the straight line SLP1. In this way, a cross section coordinate R of a vehicle body floor face is found, while a coordinate position DS is found from a seat length DL. Subsequently, a straight line SLP2 is found from the straight line SLP1 and thickness D or the seat backrest part, and a straight line SLP3 is found from width W of the seat backrest part. In this way, points P2, Q2, P3, Q3 are found, and a seat image distance distribution can be formed with ease.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting the presence / absence and posture of an occupant of a vehicle using a linear optical sensor array comprising a plurality of optical sensor elements, and more particularly to a method for controlling deployment of an airbag. The present invention relates to a suitable method for detecting the presence / absence of a vehicle occupant and its posture.

[0002]

2. Description of the Related Art The mounting rate of airbags has been increasing in recent years, and airbags are becoming standard equipment regardless of the type of vehicle. However, if the child is standing in front of the seat,
It has been reported that sitting small women can cause casualties. Therefore, various deployment control of the airbag for avoiding such a casualty accident has been proposed. Among them, a conventional method in which a sensor is provided at a ceiling position in an automobile to measure a distance from the sensor to an occupant (ranging) and to detect the presence or absence and posture of the occupant from the distance distribution will be described below.

FIG. 5 is a schematic view for explaining such a conventional example, wherein 1 is an occupant sensor, 2 is an occupant, 3 is an automobile,
Indicates a seat, respectively. This means that the occupant sensor 1
In this case, for example, four linear inspection areas (fields of view) R1, R2, R3 for the occupant 2 are formed.
By setting R4, occupant sensor outputs of a plurality of parts are obtained for each area, and the outputs are processed by a processing device (not shown).
By measuring the distance of each part of the occupant and calculating the difference from the seat image distance (distance from the occupant sensor when the occupant is not present) set by the processing device, not only the presence or absence of the occupant but also its posture Is to judge.

FIG. 6 shows an example of an occupant sensor. Here, the occupant sensor 1 is configured by integrating a multi-stage light receiving IC 10 formed by providing four pairs of optical sensor arrays and imaging lenses 21 and 22. The number of the optical sensor array pairs may be any, but here, four inspection regions or fields of view R1,
In order to set R2, R3, and R4, a four-stage configuration is used. In addition, the code | symbol 5 of FIG. 6 shows an auxiliary light source, and code | symbol V shows the direction of the visual field center of a sensor.

First, the principle of distance measurement will be described with reference to FIG. Now, the center of the imaging lenses 21 and 22 is set to the origin O
, The horizontal axis X and the vertical axis Y are set, and the coordinates of the imaging positions L ₁ and R ₁ are respectively (−a _L −B / 2, −f) and (a _R + B
/ 2, -f). Also, the center point O of the imaging lens 21
_L is the coordinate (-B / 2, 0), the coordinates of the center point O _R of the imaging lens 22 is a (B / 2, 0), the object (subject)
Assuming that the coordinates of the point M are (-x, y), the coordinates of the intersection N of the X axis with the perpendicular drawn from the point M to the X axis are (-x, 0),
The position L of the perpendicular line lowered from the point _OL to the optical sensor array 12
₀ coordinates (-B / 2, f), the coordinate position R ₀ of perpendicular dropped from the point O _R to the optical sensor array 11 (B / 2,
f).

Here, a _L represents the distance between points L ₀ and L _1, and a _R represents the distance between points R ₀ and R ₁ . At this time, △ MO _L N, △ O _L L ₀ L ₁ , △ MO _R
Since N and △ O _R R ₁ R ₀ are similar, (−x + B / 2) f = a _L · y (1) (x + B / 2) f = a _R · y (2) I do. When x is deleted from the equations (1) and (2), y = B ・ f / (a _L + a _R ) (3), and the image forming position L ₁ and the point L of the left optical sensor array 12 are obtained.
_If the distance a _L to ₀ and the distance a _R between the imaging positions R ₁ and R ₀ of the right optical sensor array 11 are known, the distance y to the object can be obtained from the above equation (3). .

Next, a method of setting a seat image will be described with reference to a flowchart of FIG. This setting is performed only once at the time of shipment. That is, in step S1 in FIG. 8, the tilt of the seat back (the seat tilt Θ) is initialized. This seat inclination is indicated by Θ in FIG. In step S2, the position of the frontmost surface of the seat (seat position) is initialized. The seat position is indicated by the reference symbol DS in FIG. Step S
At 3, the distance of the seat image at a certain seat position and inclination is measured, and at step S4, the distance distribution of the seat image is stored in the data storage area of the sensor. FIG. 10 shows an example of the distance distribution of the seat image.
In steps S5 and S6, the processes in steps S3 and S4 are executed a plurality of times, and a distance distribution pattern of the seat image for each seat position and inclination is created and stored. According to this flowchart, the seat is started from the position of Θ = Θmin, Ds = min, and the Ds value and Θ value are sequentially changed to obtain the seat distance distribution, thereby obtaining the entire seat distance distribution data.

Here, the concept of obtaining a discrete seat distance distribution as shown in FIG. 10 will be described below. Left of n optical sensor element as shown in FIG. 11, each A ₁ ~ the quantization data of the right photosensor arrays 11 and 12
A _n, and B ₁ ~B _n. A distance index (a _L + a _R ) from the front of the sensor to an object at a predetermined angle (in FIG. 7, an angle formed by an axis connecting the Y axis, the object coordinate M, and the origin O).
In order to obtain the data, observation areas (windows) W1 to Wm _{+ 1} of a predetermined size are set for the data, and (m + 1) sets of subsets C _{0 which} are alternately shifted by one sensor unit (1 bit) are set. ＭCm, and an evaluation function f (C ₀ ) 〜f (C
m) is calculated and the combination C _k that minimizes the evaluation function value is _obtained , whereby the degree of displacement between the left and right images can be determined from the value of the subscript k, and a distance index proportional to (a _L + a _R ) can be obtained. It is known. Therefore, if such processing is performed as shown in FIG. 12 for 1 to n fields of view while sequentially shifting, for example, one bit at a time, n discrete distance data can be obtained.

FIG. 13 shows an occupant discriminating process. In order to extract occupant information from the measured distance distribution, a process for obtaining a difference from the empty seat information obtained by the process shown in FIG. 8 is executed. That is, in step S7, for example, the occupant image as shown in FIG. 14 is measured. FIG. 15 shows an example of the occupant image distribution at that time. In step S8, for example, the distance distribution of the seat image as shown in FIG. 10, which is stored as a result of the processing described with reference to FIG. 8, is read from the data storage area. In step S9, a difference between the occupant image distance distribution and the seat image distance distribution is obtained, and a difference distance distribution is created. FIG. 16 shows an example of the difference distance distribution. Here, the number of measurement points (distance measurement points) indicating a distance value equal to or smaller than the threshold value TH is counted. At this time, as is clear from FIG. 16, the distance value takes a discrete value, which indicates that the above-described counting can be performed.

In step S10 of FIG. 13, the ratio A of the distance measurement points equal to or less than the above threshold value to the total number of distance measurement points in the visual field area is calculated. The processes in steps S8 to S10 are performed for all seat images. In step S11, the case where the ratio A is minimum is selected, and seat image data to be used is determined. That is, the measured distance distribution data,
Calculate the difference from the stored empty seat distance distribution data,
The one with the smallest difference is selected as the distance distribution data of the seat at the measurement point.

[0011]

As described above, conventionally, an enormous data storage area is required to previously store all the patterns of the distance distribution in the linear visual field area by changing the seat position and the inclination. There is a problem that a lot of processing is required to determine which seat image distance distribution data to use in determining the seat distance distribution data. Therefore, an object of the present invention is to reduce the use amount of a data storage area and easily obtain seat image distance distribution data.
The object is to reduce the processing time.

[0012]

SUMMARY OF THE INVENTION In order to solve such a problem, according to the present invention, a plurality of pairs of linear optical sensor arrays each composed of a plurality of optical sensor elements are provided to form an image of a vehicle occupant. The distance distribution in at least one linear visual field region of the occupant image is measured to detect the presence / absence and posture of the vehicle occupant. At least two of the rectilinear fields of view are arranged so as to include the seat back portion, and by calculating the seat position and the inclination from the distance value of the end portion of the seat back, the distance distribution of the arbitrary seat position and the inclination can be calculated. .

That is, since the shape of an automobile seat is relatively simple, the fact that the seat shape can be approximated as a combination of simple planes is used. In other words, it is only necessary to solve several equations from the position and inclination of the seat to create the seat image distance distribution data, thereby shortening the processing time. Further, in solving this equation, only a few constants need to be stored in the data storage area, so that it is not necessary to use almost any memory capacity, and it is possible to reduce the data storage area usage.

[0014]

FIG. 1 is a flowchart for explaining an embodiment of the present invention, and FIG. 2 is an explanatory diagram for facilitating understanding of FIG. Here, the occupant sensor 2 is attached to the upper part of the seat diagonally forward as shown in FIG. 2 so that the end of the seat back can always be observed and seat image distance distribution data can be created. In this case, two fields of view (R1, R2) are set for the backrest. In FIG. 1, the sensor pairs corresponding to the two visual fields R1 and R2 are referred to as a sensor 1 and a sensor 2, respectively.

In step S7, as described with reference to FIG. 13, a distance distribution as shown in FIG. 15 is obtained. In step S12, the seat end distance is obtained. The distance between the seat ends is from the mounting coordinate position S of the sensor 1 shown in FIG.
The distance to Q1 is shown, and the end of the seat is shown in FIG.
As shown. Here, to find the seat edge distance,
Find the part where the amount of change of the distance value shown in FIG.
Among the parts giving the maximum change, the smaller distance value is used. This is because the mounting position of the occupant sensor 1 is located at the upper left side of the seat.

In step S13, the respective seat distances obtained in step S12 are converted into the elevation angle information θ1 of the sensor as shown in FIG. 3 and the position of the imaged distance in the field of view of the seat end distance shown in FIG. From the information, the coordinates of the points P1 and Q1 are obtained. In FIG. 3, symbol ω indicates the sensor rotation angle, and V indicates the sensor center direction. In step S14, a straight line equation SLP1 is calculated from the coordinates of the points P1 and Q1, and in step S15, the inclination angle 傾 き of the seat is obtained from the inclination of the straight line. In step 16, the equation of the straight line SLP1 and the floor of the vehicle body is solved, and the coordinates R of intersection of the floor of the vehicle body are calculated. FIG. 2 or FIG.
Since the seat length DL is a constant as shown in FIG. 7, the seat position DS can be easily obtained from this value in step 17.

In step S18, the straight lines SLP1 and SL
A straight line SLP2 is determined from P2 being parallel and the constant thickness D of the seat back portion. In step S19, the straight line SLP1 and SLP3 are parallel and the constant width W of the seat back portion being a constant. A straight line SLP3 is obtained. In step S20, the points P2, Q2, P3, and Q3 are obtained by solving the equations of the straight lines SLP2 and SLP3 obtained in steps S18 and S19 and the planes 1 and 2 that are the viewing planes of the sensors 1 and 2. At that time, a straight line SP
From the distance values of SQ2, SQ2, SP3, and SQ3, it is determined at which position in the visual field distance of the sensor 1 the image is formed. Therefore, in step S21, the sensor 1, the sensor 1,
The seat image distance distribution of the sensor 2 can be created. FIG. 4 shows an example of the seat image distance distribution. Note that the fields of view R3, R
Regarding No. 4, the mounting mode of the sensor is determined in advance so as to show the same distance distribution regardless of the seat position and the inclination.

As described above, according to the present invention, the seat image distance distribution can be easily created by solving a simple equation several times, so that the processing time can be shortened. It is only necessary to store a few pieces of information regarding the data storage area, and the amount of data storage area used can be greatly reduced.

[0019]

According to the present invention, when creating seat data necessary for recognizing the presence / absence and posture of an occupant, it is only necessary to apply a simple seat image distance distribution creating algorithm, thereby shortening the processing time. Since only a few pieces of information about the seat and the sensor need be stored, there is an advantage that the amount of data storage area used can be significantly reduced.

[Brief description of the drawings]

FIG. 1 is a flowchart for explaining an embodiment of the present invention.

FIG. 2 is an explanatory diagram for facilitating the description of FIG. 1;

FIG. 3 is an explanatory diagram of an elevation angle, a rotation angle, and a center direction of a sensor.

FIG. 4 is an explanatory diagram showing an example of a seat image distance distribution created by the processing of FIG. 1;

FIG. 5 is a schematic diagram showing a conventional example.

FIG. 6 is an explanatory diagram illustrating a relationship between a sensor used in FIG. 5 and a visual field.

FIG. 7 is an explanatory diagram of a distance measuring principle.

FIG. 8 is a flowchart showing a conventional process of creating a seat image distance distribution.

FIG. 9 is an explanatory diagram of various constants relating to a seat.

10 is an explanatory diagram of an example of a seat image distance distribution obtained by the processing of FIG. 8;

FIG. 11 is an explanatory diagram of a correlation calculation based on the principle of distance measurement.

FIG. 12 is a diagram illustrating the principle of multi-point distance measurement.

FIG. 13 is a flowchart showing a conventional process of setting (determining) a seat image distance distribution.

FIG. 14 is an explanatory diagram of various constants relating to a seat when an occupant is present.

FIG. 15 is an explanatory diagram of an example of a distance distribution for each field of view of an occupant.

FIG. 16 is an explanatory diagram of a distance distribution of a difference between an occupant image and a seat image.

[Explanation of symbols]

1 ... occupant sensor, 2 ... occupant, 3 ... automobile, 4 ... seat, 5
... Auxiliary light source, 10 ... Multistage light receiving IC, 11, 12 ... Optical sensor array, 21,22 ... Imaging lens.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FI G01V 8/20 G01V 9/04 Q

Claims (1)

[Claims] An image of an occupant of an automobile is formed by providing a plurality of pairs of linear optical sensor arrays composed of a plurality of optical sensor elements, and a distance distribution of the occupant image in at least one linear visual field region is measured. In detecting the presence / absence and posture of a vehicle occupant, the plurality of pairs of optical sensor arrays are arranged diagonally forward and in the upper part of the seat such that at least two of the linear visual fields corresponding to the respective optical sensor arrays include the seat back. A method for detecting the presence / absence and posture of a vehicle occupant, wherein a distance distribution of an arbitrary seat position and inclination can be calculated by obtaining a seat position and inclination from a distance value of an end portion of a seat backrest.