JPH07247758A - Control device for automatic door - Google Patents

Abstract

PURPOSE:To provide excellent safety and reliability to an automatic door by a method wherein signals from sensors buried in a gateway undergo A/D conversion, and signals for opening and shutting the door are generated, with noises resulting from movement of the door corrected, only of components made by DC amplification out of the converted signals. CONSTITUTION:For automatic door control devices 100 and 100' for controlling a motor 9 for opening and shutting doors 2 and 4, sensors 10 and 10' buried on both sides with the doors 2 and 4 on a gateway interposed in between for functioning as capacitors and working for sensing human bodies are connected respectively to the control devices 100 and 100'. Outputs from high-frequency oscillation detectors connected to the sensors 10 and 10' are amplified, and A/D conversion for digitization is applied, and outputs on oscillation detection from the sensors are inputted precisely and correctly. Then, signals for correcting noises resulting from movement of the doors 2 and 4 are generated easily while low-frequency noises in large amplitude arising from temperature drift and the like are eliminated. Thereby, an automatic door device that is highly reliable and excellent in safety can be obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improved automatic door control device using a high frequency proximity switch.

[0002]

2. Description of the Related Art A sensor for detecting a human body is provided at the entrance of an automatic door, and this sensor is constructed as a capacitor of a high-frequency proximity switch, and when a human body approaches the sensor, it is caused by a change in the capacitance of the capacitor. A device for controlling opening / closing of an automatic door by detecting a frequency fluctuation and a decrease in sharpness is disclosed in JP-A-2-35502 and Japanese Patent Application No. 5-8677.
No. 5 publication.

In the device disclosed in Japanese Patent Laid-Open No. 2-35502,
The analog output of the high-frequency oscillation detection unit coupled to the sensing unit is compared with a reference voltage output from the DA conversion unit by an analog comparator to AD-convert the detection output, and the digitized detection output is controlled by the control unit. The human body is detected by arithmetic processing. On the other hand, in the device of Japanese Patent Application No. 5-86775, a small-amplitude human body detection signal is extracted from a high-frequency oscillation detection output mixed with large-amplitude low-frequency noise due to temperature drift or the like. A large-amplitude low-frequency noise correction device combined with a switch means has been introduced.

[0004]

As described above, in the high-frequency proximity switch in which the sensing unit is configured as the capacitor of the high-frequency oscillating unit and the human body is detected from the oscillation detection output, the amount of change in the human-body detection signal component is very small and the temperature is small. A major problem is how to remove large-amplitude low-frequency noise due to drift or the like.

However, in the device having the configuration as disclosed in Japanese Patent Laid-Open No. 35502/1990, when an attempt is made to detect a small amplitude signal, a high-precision (around 16 bits) AD conversion means or DA conversion means is required, and the apparatus is Very expensive,
There is a problem that it is difficult to maintain the accuracy of the AD conversion means or the DA conversion means constant without being affected by temperature changes. Further, in the device of the above publication, the reference voltage generation command must be constantly output from the control means to the DA conversion means, and if the detection outputs of a plurality of sensing parts are input, the load on the control means will increase significantly. There were also problems.

Further, in the apparatus having the structure as disclosed in Japanese Patent Application No. 5-86775, the relative accuracy of the AD-converted digital value can be maintained while the large amplitude low frequency noise can be corrected as expected, but the correction can be performed. When it is deviated, since the AD conversion means of low resolution is used, the absolute accuracy of the digital value cannot be obtained, and there is a problem that it takes time until the measurement result becomes stable.

By the way, the applicant of the present application has noticed the following fact while studying the oscillation detection output of the automatic door which is opened and closed by the highly accurate AD conversion means. That is, as shown in FIG. 4 (A), when the time axis is viewed in units of 1 to 2 minutes, the detection output DT has a waveform that falls rapidly as a person approaches and rises rapidly as a person exits. Detection output DT 10-20
When viewed in units of time, a temperature drift waveform that greatly exceeds the human detection level is observed as shown in FIG. Further, when the automatic door is opened and closed for a short time without an operator, the detection output as shown in FIG. 7A is obtained, but within 30 minutes, the reproducibility of this detection output waveform is good.

The present invention has been made in view of the above points, and an object of the present invention is to provide a highly accurate and low cost AD conversion means utilizing voltage-time signal conversion means, and to provide a sensing unit. The detection output of is detected with high accuracy so that the movement of the door can be detected, and the effects of background noise signals and temperature drift due to the movement of the automatic door are removed, and the safety and reliability of human body detection are dramatically improved. To provide a control device for the automatic door.

[0009]

SUMMARY OF THE INVENTION The present invention relates to a control device for an automatic door, and an object of the present invention is to provide a sensing unit that is embedded in the doorway of an automatic door and acts as a capacitor to sense the human body. High frequency oscillation detection section coupled to the section, AD conversion means for amplifying and digitizing the output of the high frequency oscillation detection section, and control for controlling the opening and closing of the automatic door by processing the output of the AD conversion means In the automatic door control device including a means, a storage means readable and writable by the control means is provided, and the digital output of the high frequency oscillation detector is stored in the storage means at a predetermined cycle and stored in the storage means. The generated digital output waveform of the oscillation detection unit is time-series analyzed by the control means, and the automatic door opening / closing control signal is generated from the short-term fluctuation component of the digital output waveform. While it is accomplished by generating a noise compensation signal of the high frequency oscillator detector unit from long-term fluctuation component of the digital output waveform. Further, the above-mentioned object of the present invention is to embed the sensing unit embedded in the doorway of an automatic door and acting as a capacitor to sense a human body, the high-frequency oscillation detection unit coupled to this sensing unit, and the output of this high-frequency oscillation detection unit. In addition, a controller for an automatic door, comprising: an AD conversion means for digitizing the output; and a control means for processing the output of the AD conversion means to control the opening and closing of the automatic door. And a storage unit that is readable and writable by the control unit, and when the background data of the automatic door is recorded, the automatic door is opened and closed with no sensing object in or near the moving space of the automatic door to move the high frequency oscillation. The output data of the detection unit is measured at predetermined intervals and stored as background data in the storage means, and the high frequency oscillation is performed during automatic operation of the automatic door. And detecting the presence or absence of the sensing object by the difference data of the background data at the current position of the automatic door and the output data of the wave portion, also achieved by generating a switching control signal of the automatic door.

[0010]

According to the present invention, in the above structure, the oscillation detection output of the sensing section can be accurately and accurately inputted by the highly accurate and low cost AD conversion means. Therefore, if only the DC amplified signal component is inputted, An AC signal component can be automatically generated by arithmetic processing, which makes it easy to distinguish between a signal component and a noise signal such as temperature drift, and a noise signal component based on door movement is stored separately with door position data. Therefore, noise signal correction based on door movement can be easily executed, and a highly reliable and safe automatic door control device can be provided.

[0011]

Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a conceptual diagram of an automatic door system, in which a sensing unit 10 embedded in the doorway of the automatic doors 2 and 4 that opens and closes.
Or when the person 11 approaches 10 ', the person 11 is controlled by the control device.
The control of opening and closing the automatic doors 2 and 4 by detecting the approach of the.

FIG. 2 shows an example of the mechanical structure of the above-mentioned automatic door system. Doors 2 and 4 are mounted on a seamless (channel, base) 1 fixed on the building side through sliding rollers 3, 3 or 5, 5. And the sprocket wheels 6 and 7 are axially mounted on both ends of the seamless 1 to stretch the chain 8. Then, the upper and lower parts of the chain 8 are engaged with the door 4 or 2 and the motor control circuit 12
From the sprocket wheel 7 via the motor reduction mechanism 9
The doors 2 and 4 are opened and closed by rotationally driving. The sensing units 10 and 10 'are connected to an oscillation detection unit provided inside the control device 100 or 100', respectively, and act as capacitors whose capacitance changes when a person 11 approaches, and the detection output thereof is controlled by the control device 100 or 10 '.
Motor control signal MC1 or M
C2 is generated, and the motor control circuit 12 activates the opening and closing of the doors 2 and 4.

In the system having the mechanical structure as shown in FIG. 2, the hardware structure of the sensor signal processing devices 100, 100 'of the sensing units 10, 10' will be described as an embodiment of the present invention with reference to FIG. An impedance converter 23, an in-phase current blocking choke coil 24, and a feeder 30 such as a coaxial cable are connected in series between the two sensing plates 21 and 22 forming the sensing unit 10 and the oscillation detection unit 120. Has been done. The oscillation output of the oscillation detection unit 120 (frequency is about several hundred KHz to several tens of MHz) is detected by the detection unit composed of the diodes D1, D2, etc., and the detection output DT is input to the AD conversion means 140 which will be described later and the voltage- It is converted into a time signal VT, and this signal VT is inputted into a control means composed of a microprocessor (hereinafter, abbreviated as MPU) 160 and the like, arithmetically processed and converted into a digital signal. That is, the time signal VT is AD-converted into a digital signal by the timer 170 and the software timing processing unit 172 provided inside the MPU 160, and is also stored in the storage unit 150 such as a battery-backed RAM or flash memory. 150
When the MPU 160 time-sequentially analyzes the digital signal stored in the MPU 160 and detects the approach or exit of the human body, the MPU 16
The motor control signal MC1 is generated via the output terminal 02 of 0, the relay 166, etc., and the motor control circuit 1 shown in FIG.
It is designed to be output to 2nd grade.

The structure of the AD conversion means 140 described above will be described in more detail. The operational amplifier 142 and the capacitor C1.
A sawtooth-shaped reference voltage Vr is generated by a voltage integrator circuit configured with 0 (this capacitor is preferable to have as little leakage current as possible), and the reference voltage Vr and the detection output DT are generated.
And DC are amplified and compared by the operational amplifier 148, and the output VT is the external interrupt terminal IN of the MPU 160.
It is designed to be connected to T. Then, MPU1
A timer 170 and a software timing processing unit 172 are provided inside the 60, and the output terminal 0 of the MPU 160 is
The time from resetting the capacitor C10 via the reset signal CNT outputted from 1 and the switch means 146 to the fall of the interrupt signal VT is measured.

Further, M provided in the control device 100 described above.
The PU 160 is connected to an output level adjusting means 162 and a varicap voltage adjusting means 164, each of which is constituted by a DA converting means or the like for automatically controlling and stabilizing the operation of the oscillation detecting section 120, and adjusting the varicap voltage. The output AS2 of the means 164 is changed to adjust the capacitance of the varicap VC to stabilize the oscillation frequency at the center of the oscillating unit, and the output AS1 of the output level adjusting means 162 is changed to cause the temperature drift of the detection output DT. And so on.

In such a structure, the operation is shown in FIG.
Referring to the flowchart of FIG. 6 and the time chart of FIG. 6, first, when the power is turned on or the tuning of the oscillator is largely deviated, as described in JP-A-62-82189, the oscillation detector In order to tune the high frequency oscillation frequency of 120, for example, the MPU 160 outputs the output level setting signal AS1 via the output level adjusting means 162 so that the detection output DT is set to the middle of all the amplitudes, and the MPU 160 outputs the varicap voltage. The sawtooth waveform data similar to the conventional one is sequentially output to the adjusting means 164, the analog sawtooth signal AS2 is input to the varicap VCD, and the detection output DT thereof is AD conversion means 14.
When it is read into the MPU 160 via 0 to check the tuning state and a predetermined detection output level is obtained, the sawtooth waveform data DIA10 at that time is latched inside the MPU 160 to stop the output of the sawtooth signal AS2, and Cap V
The waveform data DIA10 is continuously output to the varicap voltage adjusting means 164 so that a constant voltage is applied to the CD (step S2).

Thus, when the tuning adjustment of the high frequency oscillation frequency is completed, the oscillation detection output DT is digitized by the AD conversion means 140 every predetermined period and stored in the MPU 160 as time series data (step S4). This digitizing process will be described in more detail with reference to the block diagram of FIG. 3 and the time chart of FIG.
When the reset signal CNT is output from the MPU 160 in No. 1, the switch means 146 composed of an analog switch or the like is short-circuited, and the reference voltage generating capacitor C10.
Is reset. Next, when the reset signal CNT is released at time t2 in FIG. 6A, a sawtooth reference voltage Vr as shown in FIG. 6C is generated with a time constant determined by the resistor R4 and the capacitor C10. , This reference voltage V
r is compared with the oscillating detection output DT by the operational amplifier 148, and at the time t by the software timing processing unit 172.
From 2 on, the time counting process of the timer 170 is started.

Subsequently, the reference voltage Vr rises with a constant gradient, and reaches the time point t3 shown in FIG. 6C, and the oscillation detection output D
The level dtφ of T becomes equal to the reference voltage Vr, the output VT of the operational amplifier 148 falls, and the MPU 160 generates an external interrupt INT. As a result, the software timing processing unit 172 can stop the timing processing of the timer 170 and convert the analog oscillation detection output DT into a digital value.
Here, if the frequency of the timer 170 is set to about 10 MHz, 16-bit precision AD conversion can be easily realized within several tens of msec. Above time points t1 to t4
By repeating the above process, the oscillation detection output DT is digitized into a time signal in this cycle, and M is used as time series data.
The data is sequentially stored in the storage unit 150 via the PU 160.

The thus-digitized oscillation detection output is time-series analyzed by the MPU 160 to detect background noise signals and human approach / exit signals (step S).
6). An example of such time-series data analysis processing is introduced in, for example, Japanese Patent Laid-Open No. 2-35502. This will be described with reference to FIG. 4. The DC detection output DT has a time axis of 1
When viewed in a unit of ~ 2 minutes, the waveform rapidly falls with the approach of a person, and rises rapidly with the exit of the person. When the DC detection output DT is viewed in the unit of 10 to 20 hours,
As shown in (B), a temperature drift waveform that greatly exceeds the human detection level is observed. Therefore, as shown in FIG. 4C, it is preferable to perform a process of stably extracting the human body detection signals P1, P2, ..., P4 of small amplitude from the detection output in which the large amplitude low frequency noise is mixed. The time change ΔDTS / t due to the approach / exit of the human body and the time change ΔDTN1 / t with respect to the drift of the temperature and the like generally have the following relationship.

[0020]

## EQU1 ## ΔDTS / t> ΔDTN1 / t In the present application, this temporal change is referred to as a short-time change time width (ti
-Tj) and the time width for long-term change (ti-tk) are used for evaluation. From the detection output DTi sampled at the time point ti, the temporal change component ΔD
Equations 2 and 3 show the process of calculating T1 and ΔDT2.

[0021]

## EQU00002 ## .DELTA.DT1 = (DTi-DTj) / (ti-tj)

[0022]

## EQU00003 ## .DELTA.DT2 = (DTi-DTk) / (ti-tk)

[0023]

(Ti-tj) <(ti-tk) where DTi is the detection output at time ti, DTj is the detection output at time tj, and DTk is the detection output at time tk. The determination condition of is as follows.

[0024]

[Expression 5] ΔDT1 <-DTH1

[0025]

[Mathematical formula-see original document] ΔDT1> DTH2 where DTH1 and DTH2 are judgment values for human body detection

[0026]

[Expression 7] ΔDT2 <-DTH3

[0027]

[Delta] DT2> DTH4 However, DTH3 and DTH4 are background noise determination values. That is, when the MPU 160 obtains the digital data input value DTi at time ti, the time change at this time is calculated by Equations 2 and 3. When the signal change ΔDT1 in a short time (ti-tj) satisfies the condition of Expression 5, it is determined that a human body or the like has approached (step S8), and an automatic door open signal is output from the MPU 160 (step S10). When the signal change ΔDT1 for a short time satisfies the condition of Expression 6, it is determined that the human body or the like has exited (step S8), and the MPU 160 outputs the automatic door closing signal (step S).
10).

On the other hand, a long-term (ti-tk) signal change Δ
When DT2 satisfies the condition of Expression 7 or Expression 8, it is determined that the background noise signal has changed (step S12), and the correction output is output from the MPU 160 to the output level adjusting unit 162 in the direction of canceling the signal change. (Step S1
4).

The above process is repeated every predetermined period to control the opening and closing of the automatic door. In the above calculation, if the data at time ti is smoothed by moving average with (m + 1) pieces of data of ti, ti-1, ... You can, but A
When the number of bits of D conversion is small (for example, 8 bits), the human body detection signal level is buried in the background noise, and it is not possible to reliably detect the approach and exit of the human body. By inputting a signal with a high precision of 16 bits or more as in the present invention, it becomes possible to distinguish the human body detection signal component from the noise component, and it is possible to safely and reliably execute the automatic door opening / closing control. It was

FIG. 8 shown in correspondence with FIG. 3 is an example of another embodiment of the present invention. The devices with the same reference numerals perform the same functions, and the door position measuring means 18 is provided.
0 is combined with the MPU 160, background noise due to door movement is learned in advance and stored in the storage means 150, and background noise due to door movement is automatically removed from the oscillation detection output. . In the control device 100a having such a configuration, the door position measuring means 180 and the MPU 1
The relationship between the memory 60 and the storage means 150 will be described in more detail. As shown in FIG. 9, the oscillation detection outputs which are the outputs of the high frequency proximity switches 1 to n are AD-converted and stored in the storage means 150. Inside the automatic door moving sections d1, d2, ..., Dm, the oscillation detection output DT is AD-converted while distinguishing the moving direction of the door according to the door opening movement and the door closing movement. It is designed to be stored in the memory location. In addition to the encoder mounted on the motor, the door position measuring means 180 may replace the moving distance of the automatic door with a time signal and substitute the elapsed time from the door opening start signal or the elapsed time from the door closing start signal. It is possible (the time interval in this case is, for example, in the unit of several 10 msec). Further, since the background noise processing based on the door movement described above is effective as a safety measure particularly when the automatic door is closed, the background data may be stored in the storage unit 150 only during the closing operation of the automatic door. Furthermore, in the fully closed position and the fully open position of the automatic door,
If you reset the timing counter for position measurement,
The accumulated error with the passage of time can be reset.

In such a configuration, the operation is shown in FIG.
Referring to the time chart of FIG. 10 and the flow chart of FIG. 10, first, the varicap voltage adjusting means 164 outputs a sawtooth wave, and the detection output DT of the oscillation detection unit 120
The varicap capacitance is changed so as to obtain a predetermined output level (step S20).

When the tuning adjustment of the high frequency oscillation frequency is completed, the MPU 160 outputs the automatic door opening signal MC1 to open and close the automatic doors 2 and 4 in an unmanned state, and then, as shown in FIG.
The process moves to the background data recording / learning operation based on the opening / closing movement of the automatic door as shown in (step S22). In this mode, for example, the detection output DT is AD-converted in units of several tens of msec, and the time signal (or movement distance signal) is stored in the storage unit 150.
The background noise data is sequentially stored in accordance with. The accuracy of the background data can be improved by automatically executing the unmanned opening / closing operation of the automatic door a plurality of times and storing the average value as the background noise data.

Thus, when the background data learning operation based on the opening / closing movement of the automatic door is completed, the automatic operation mode is entered and the opening / closing control of the automatic door is performed as follows. First,
The oscillation detection output DT is AD-converted at a predetermined cycle (cycle from time t1 to time t4 in FIG. 6A), and the door position measuring means 18 is operated.
It is stored in the MPU 160 and the storage unit 150 together with the door position data output from 0 (step S24). Next, the background noise level based on the door position is stored in the storage unit 15.
The present door position data of 0 is read out, and the detection output signal with door movement correction is generated by the MPU 160 by the difference calculation (step S26 and FIG. 7E).

Subsequently, the detection signal output with the door movement correction is subjected to short-term fluctuation processing by the equation 2 to check whether or not there is a rapid short-term fluctuation (step S28). In this case, when a rapid falling waveform is detected, it is determined that a human body or the like is approaching, and a door opening signal is output, while when a rapid rising waveform is detected, a person or the like exits. It is determined that the door is closed and a door closing signal is output after a predetermined waiting time (step S30).

If a rapid short-term fluctuation is not detected, the door movement correction detection output signal is subjected to long-term fluctuation processing by the equation 3 to check whether or not there is a long-term fluctuation (step S32). In this case, when a long-term fluctuation is detected, a correction signal is output to the detection output level adjusting means 162 in a direction to reduce this fluctuation, and the MPU 160 controls so that the detection output level becomes a constant value (step S34). ). If no long-term fluctuation is detected, the process returns to step S24. After that, steps S24 to S34 are repeatedly executed at a predetermined cycle. Thus, it becomes possible to detect the human body by removing the noise signal based on the opening / closing movement of the automatic door. In the noise removal signal processing based on such door movement, as shown in FIG. 7D, when the automatic return signal of the automatic door is output to the sensing unit 10 while the person is stopped and the automatic door starts to close, for example, Since the detection output level has already dropped significantly at time t33 in FIG. 7D, it is difficult for the control device using the conventional DC amplifier to detect the minute change as at time t34. In the present application, even in such a case,
It is possible to stably detect the movement of a person while correcting the noise signal based on the movement of the automatic door, and it is also possible to correct long-term noise fluctuation such as temperature drift.

FIG. 11 corresponding to FIG. 8 shows still another embodiment of the present invention, in which the devices with the same reference numerals perform the same functions and the sensing unit 10 is used.
, 10n are installed and the outputs of these sensing plates are respectively detected by the oscillation detectors 120a, 120b, 120b ,.
..., input to 120n, and the output is DC amplification type A
The digital outputs VT1, VT2, ... VTn, etc. are input to the D conversion means 140a, 140n, taken into the MPU 160 via the interrupt input INT or the digital input DI, and converted into time signals by the software timing processing unit 172. It was done. Also, DA conversion means 162a, 162
b, ... 162n through the oscillation detectors 120, 120b,
, 120n oscillation frequency and output DC level are automatically corrected, the reference voltage generating means incorporated in the AD converting means 140a to 140n is reset via the digital output DO1, and the motor control is performed via the digital output DO2. The control signals MC1, MC2, ..., MCn to the circuit are ON-OFF controlled.

In the control device having such a configuration, the sensing unit 10
The switching means for a, 10b, ..., 10n are not required, and the reference voltage is MPU in each voltage-time signal generating means.
Since it is generated independently of the operation of 160, it is stabilized, and its output signals VT1, VT2, ..., VTn are also MPUs at high speed.
Since digital input can be made to 160, there is an advantage that the accuracy of AD conversion can be maintained as compared with the case where the detection output is switched by the analog switching unit and AD conversion is performed even if the number of installation locations of the sensing unit is increased.

[0038]

As described above, according to the automatic door control apparatus of the present invention, it is easy to improve the AD conversion accuracy of the detection output, which was conventionally converted to around 8 bits, to 16 bits to 24 bits. As a result, the background noise component of the detection output and the detection signal component of the human body can be easily distinguished, and highly safe and reliable automatic door control can be realized.

[Brief description of drawings]

FIG. 1 is a diagram showing a relationship between an automatic door and a sensing unit.

FIG. 2 is a diagram showing a mechanism unit for opening / closing control of an automatic door.

FIG. 3 is a block diagram showing an embodiment of an automatic door control device of the present invention.

FIG. 4 is a diagram for explaining a detection output of a sensing unit.

5 is a flowchart for explaining the operation of the control device of FIG.

FIG. 6 is a time chart for explaining the operation of the AD conversion means.

FIG. 7 is a time chart for explaining a noise signal removing operation based on movement of an automatic door.

FIG. 8 is an example of a block diagram showing another configuration example of the present invention.

FIG. 9 is a diagram for explaining a background data recording process.

FIG. 10 is a flowchart for explaining a control operation with automatic door movement correction.

FIG. 11 is a block diagram showing another embodiment of the automatic door control device of the present invention.

[Explanation of symbols]

 2,4 Doors 10, 10 ', 20, 10a, 10b, 10n Sensing part 24 High permeability magnetic material 30 Feeder 100, 100', 100a Control device 120 Oscillation detecting part 140 AD conversion means 142, 148 Operational amplifier 146, 146 ′ Switching means 150 Storage means 160 MPU 162, 164 Voltage adjusting means 170 Timer 172 Software timekeeping processing unit

Claims (11)

[Claims] 1. A sensing unit embedded in an entrance of an automatic door and acting as a capacitor to sense a human body, a high-frequency oscillation detection unit coupled to the sensing unit, and a digital output for amplifying the output of the high-frequency oscillation detection unit. In an automatic door control device comprising an AD conversion means for converting the output of the AD conversion means and a control means for controlling the opening and closing of the automatic door by arithmetically processing the output of the AD conversion means, a storage means readable and writable by the control means is provided. A digital output of the high-frequency oscillation detection unit is stored in the storage unit in a predetermined synchronization, and the digital output waveform of the oscillation detection unit stored in the storage unit is time-series analyzed by the control unit to obtain the digital output waveform. Generates a control signal for opening and closing the automatic door from the short-term fluctuation component of the high-frequency oscillation detection unit from the long-term fluctuation component of the digital output waveform. An automatic door control device characterized in that a correction signal is generated. 2. A sensing unit which is embedded in an entrance of an automatic door and acts as a capacitor to detect a human body, a high frequency oscillation detecting unit coupled to the sensing unit, and a digital output which amplifies an output of the high frequency oscillation detecting unit. An automatic door control device comprising: an AD conversion means for converting the output of the AD conversion means; and a control means for controlling the opening and closing of the automatic door by processing the output of the AD conversion means. Means for reading and writing from the means, and at the time of recording the background data of the automatic door, the automatic door is opened and closed with no sensing object in the moving space of the automatic door or in the vicinity thereof, The output data is measured at predetermined intervals and stored as background data in the storage means, and during the automatic operation of the automatic door, the output data of the high frequency oscillation detection unit is output. The automatic door control device is characterized in that the presence or absence of the sensing object is detected based on the difference data between the background data at the present position of the automatic door and the opening / closing control signal of the automatic door. 3. The difference data between the output data of the high frequency oscillation detector and the background data at the current position of the automatic door is time-series analyzed by the control means, and the automatic door is calculated from a short-term fluctuation component of the difference data waveform. And a noise correction signal for the high-frequency oscillation detector based on the long-term fluctuation component of the difference data waveform. 4. The control device for an automatic door according to claim 1, wherein the AD conversion means is a voltage-time signal conversion means. 5. The position measurement of the automatic door is realized by time measurement after the automatic door starts moving, and the position measurement data of the automatic door is respectively measured at a fully closed position and a fully opened position of the automatic door. 5. The automatic door control device according to claim 2, wherein the control device is reset to a predetermined value. 6. The automatic door is opened and closed a plurality of times when the background data of the automatic door is recorded, and the output background data of the high frequency oscillation detector and the automatic door position data are automatically learned. The automatic door control device according to any one of 2 to 5. 7. A detection output level adjusting means is provided between the high-frequency oscillation detection section and the control means, and the detection output is set so that the AD conversion data has a constant value while a sensing object is not detected by the sensing section. 7. The automatic door control device according to claim 1, wherein the level adjusting means is controlled. 8. The background data is stored only during a period in which the automatic door moves from a fully open position to a fully closed position, and
8. The method according to claim 2, wherein the sensing object detection process is used.
Automatic door control device described in 9. The control device for an automatic door according to claim 1, wherein the storage means is a battery-backed RAM memory or flash memory. 10. The automatic door control device according to claim 1, wherein the output of the high-frequency oscillation detector is subjected to DC amplification and AD conversion. 11. The n number of the sensing section, the oscillation detection section and the AD conversion section are provided, and the outputs of the n oscillation detection sections are input to the control section via the AD conversion section. Item 1. An automatic door control device according to items 1 to 10.