JP4188251B2 - Automatic door sensor - Google Patents

Description

  The present invention relates to an automatic door sensor, and more particularly, to updating a reference value used when detecting the presence or absence of a person or an object.

  As an automatic door sensor, for example, there is one disclosed in Patent Document 1. In this technique, an automatic door is provided at a door opening located at the boundary between the indoor and the outdoor of a building so as to open and close the door opening. An infrared sensor is provided on each of the indoor and outdoor surfaces above the door opening. Some of the detection areas formed by these sensors cross each other above the vertical direction of the track on which the automatic door travels. When both sensors are in the detection state at the same time, it is determined that a person or an object exists on the track, and the open state of the automatic door is maintained.

Japanese Patent Laid-Open No. 2002-322872

  According to the above technique, when a person or an object is present on the track of the door, the safety of the automatic door can be improved without closing the door. By the way, in an infrared sensor, in order to prevent malfunctioning by the change of the infrared reflectance in a detection area, what is called background learning may be performed. Background learning is, for example, a result of an infrared sensor emitting light reflected from an infrared projector and reflected by some object by an infrared receiver, and comparing the received light signal with a predetermined reference signal. In the case of generating a detection signal, the state where the amount of reflected infrared light received by the infrared light receiver exceeds the predetermined range continues for a predetermined time, and the predetermined signal is generated. When the learning condition that the amount of change in time is within a predetermined allowable range is satisfied, the reference signal is updated to a value based on the light reception signal of the light receiver at that time. When this background learning function is applied to the above-described automatic door sensor disclosed in Patent Document 1, when a passer such as an elderly person is passing through a non-intersection portion of the detection areas of the two sensors. The above-described learning condition is satisfied by a sensor having a detection area in transit. Moreover, when a passerby falls on the track | orbit of an automatic door, said learning conditions may be satisfied in both sensors. In this case, background learning is performed by a sensor that satisfies the learning conditions, and thereafter, it may not be possible to detect a passerby whose traveling speed is extremely slow or a passerby who has fallen on the track. As the infrared sensor, a distance measuring sensor that projects infrared light from a projector and measures the distance from the incident position of the reflected infrared light to the position detection device (PSD) to the reflector from the infrared sensor may be used. In this distance measuring sensor, for example, a reference signal corresponding to a distance from the floor surface by a predetermined distance and the distance to the distance measuring sensor is used, and if the measurement distance is shorter than this reference signal, a person or object Is detected and a detection signal is generated. In order to determine the reference signal, the distance from the distance measuring sensor to the floor surface is measured. This distance may also change due to changes in the amount of reflected infrared rays, and background learning is performed in the same manner as described above. Therefore, the same problem as described above also occurs in the distance measuring sensor.

  It is an object of the present invention to provide an automatic door sensor that allows background learning to be performed at an optimal time in consideration of detection results of other sensors when background learning is performed in one infrared sensor. To do.

An automatic door sensor according to an aspect of the present invention includes first and second sensors, each of which forms a detection area adjacent to a track of the automatic door, and a person or a person in the corresponding detection area. A detection signal is generated when an object is detected. The detection area extends from a reference point above the automatic door to a predetermined area on the floor. At least the first sensor is an infrared sensor. The second sensor can also be an infrared sensor. As an infrared sensor, a light projecting unit that projects infrared light, a light receiving unit that generates a received light signal that represents the amount of reflected light of the projected infrared light, and a reference signal that represents the received light signal and a predetermined reflected light received amount It is possible to use a reflection type including a determination unit that generates a detection signal according to the result. Alternatively, as an infrared sensor, a light projecting means for projecting infrared light and a ranging signal representing the distance from the infrared sensor to the reflector reflecting the infrared light are generated based on the incident position of the reflected infrared light. A distance measuring type comprising a position detecting means for comparing the distance measuring signal with a reference signal representing a predetermined distance from the infrared sensor and generating a detection signal according to the comparison result. It can also be used. At least the first sensor includes: a determination unit that determines whether or not a learning condition is satisfied; and a learning unit that learns the background of the corresponding detection area when the determination unit determines that the learning condition is satisfied. Have. As a learning condition, a state in which the output value from the light receiving means or the position detecting means is not less than a predetermined upper limit value or is not more than a predetermined lower limit value continues for a predetermined time and the output value changes during the predetermined time There can be used whether or not the change in the permissible range determined in advance. As the learning unit, for example, a unit that updates the reference signal based on the output signal of the light receiving unit or the position detecting unit after the learning condition is satisfied is used. As the updated reference signal, for example, an averaged output signal of the light receiving means or the position detecting means can be used. When the second sensor is in the detection state, the temporary stop means temporarily stops the determination means or the learning means. For example, when the second sensor is in the detection state, the suspension unit temporarily stops the determination unit from determining whether or not the learning condition is satisfied when the second sensor is in the detection state. Alternatively, when the second sensor is in the detection state, the temporary stop unit temporarily stops the operation of the learning unit even if the determination unit determines that the learning condition is satisfied. As described above, the learning means performs learning when the second sensor is in the non-detection state and the learning condition is satisfied.

  With the automatic door sensor configured as described above, when the passerby moves at a very low speed in the detection area of the first sensor, the learning condition may be satisfied in the first sensor. . However, even a slow passer is likely to move into the detection area of the second sensor before the learning condition is satisfied, and the second sensor detects this passer. As a result, the temporary stop means temporarily stops the operation of the determination means or the learning means, so that learning is not performed and it is possible to prevent an erroneous reference signal from being set. Alternatively, even if the passerby has fallen on the track of the automatic door, the second sensor generates a detection signal, so that the temporary stop means temporarily stops the operation of the determination means or the learning means. Therefore, learning is not performed. In this way, learning in the first sensor is performed only when the learning condition is satisfied and the second sensor detects neither a person nor an object, so that the reference signal is updated to an incorrect value. Can be prevented.

  When the second sensor enters a non-detection state, the temporary stop means can start the operation of the determination means or the learning means. With this configuration, when the second sensor enters a state in which neither a person nor an object is detected, the determination unit or the learning unit is immediately operated, so that learning can be resumed promptly.

  The first and second sensors can form the detection area corresponding to an area on the side where the first and second sensors are provided, of the two areas partitioned by the automatic door. That is, when the first sensor is provided on the first area side partitioned by the automatic door, the detection area by the first sensor is also provided on the first area side, and the second sensor is provided by the automatic door. If it is provided on the partitioned second region side, the detection area is also provided on the second region side by the second sensor.

  For example, the first sensor is provided on the first region side, the second sensor is provided on the second region side, and the lowermost detection areas of the first and second sensors are on the floor surface of the first region. When a stationary object, for example, a mat or a flowerpot, is arranged in these detection areas in a state set in a certain manner, even if the learning condition is satisfied in the first sensor, the second sensor detects the stationary object. So I can't learn. However, as described above, the detection area of the first sensor is formed on the side where the first sensor is installed, and the detection area of the second sensor is formed on the side where the second sensor is installed. Even if a stationary object is placed in the detection area of the first sensor, the second sensor does not detect it and learning can be performed.

  Further, the first and second sensors are provided above the automatic door, and the detection area by the first sensor is the two areas partitioned by the automatic door on the side where the first sensor is provided. A detection area formed by the second sensor and formed by the second sensor can be formed in an area including on the track, or an area extending vertically above the track and extending to a region on the side where the first sensor is provided. If comprised in this way, when a person or an object exists on the track | orbit which an automatic door drive | works, since a 2nd sensor detects this, it can prevent a 1st sensor performing learning. . Therefore, it is possible to prevent learning that a person or an object is erroneously present on the trajectory but is not detected.

  As described above, according to the present invention, the timing for performing background learning in one infrared sensor is determined in consideration of the detection results of the other sensors, so that it can be performed at the optimal timing.

  An automatic door sensor according to an embodiment of the present invention is provided in an automatic door device as shown in FIG. This automatic door device opens and closes, for example, a rectangular door opening 4 that is formed to communicate with the inside and outside of a building 2 by a door 6. The door 6 is a sliding door type that opens and closes the door opening 4 by running on a track along the length direction of the door opening 4, and is a double-drawing or single-drawing type. On the upper portion of the door opening 4, an intangible 8 is provided along the length direction of the door opening 4. A first sensor, for example, an infrared sensor 10 is provided on the inner surface of the invisible 8 building, and a second sensor, for example, an infrared sensor 12 is provided on the outer surface of the invisible 8 building.

  These infrared sensors 10 and 12 have the same configuration, for example, as shown in FIG. These infrared sensors 10 and 12 have a light projector 14 made of light projecting means, for example, an infrared light emitting diode. The infrared rays from these projectors 14 are projected toward the floor surface inside and outside the building 2 as shown in FIG. The projected infrared rays are reflected by the floor surface or the like and received by a light receiver 16 including a light receiving means, for example, a light receiving transistor. The light receiver 16 generates an output signal, for example, a light reception signal corresponding to the amount of received light.

  By such light projection and light reception, detection areas 18 and 20 that are enlarged from the infrared sensors 10 and 12 toward the floor as shown in FIG. 1 are formed. As shown in FIG. 3, these detection areas 18 are arranged in the first area, for example, the inside of the door 6, of the first and second areas partitioned by the door 6, and the direction of movement of the door 6 is approached. A plurality are provided along. Among the first and second regions, a plurality of detection areas 20 are provided in the second region, for example, outside the door 6, approaching the door 6 and along the moving direction thereof. In order to form a plurality of detection areas 18 and 20 in this way, the projector 14 has a plurality of infrared light emitting diodes, and the light receiver 16 has the same number of light receiving transistors as the infrared light emitting diodes.

  The light projector 14 and the light receiver 16 are controlled by a control unit, for example, a CPU 22. The CPU 22 operates according to a program stored in a storage unit, for example, the ROM 24, and controls the projector 14 and the light receiver 16 while using the storage unit, for example, the RAM 26 as a working area. Further, the CPU 22 compares the light reception signal from the light receiver 16 with a storage means, for example, a reference signal stored in the EEPROM 28, for example, a threshold value, and the value of the light reception signal becomes smaller than the threshold value. It is determined that a person or object has been detected, and a detection signal is generated.

  Moreover, CPU22 can switch the contact (not shown) provided in the detection signal input / output part 30 so that a detection signal can be output. The output detection signal is transmitted to an automatic door controller (not shown), and the door 6 is opened. Further, it is output to another sensor (if the sensor that outputs this detection signal is the sensor 10, it is output to the sensor 12, and if the sensor that outputs the detection signal is the sensor 12, it is output to the sensor 10). Further, the CPU 22 can switch the contact of the detection signal input / output unit 30 so as to input the detection signal. In this case, a detection signal is input to the CPU 22 from another sensor (from the sensor 12 if the sensor to input the detection signal is the sensor 10 or from the sensor 10 if the sensor to input the detection signal is the sensor 12). Is done.

  Further, the CPU 22 is in a non-detection state when the background learning condition is satisfied and the other sensor (the sensor 12 if the sensor to which the CPU 22 belongs is the sensor 10, the sensor 10 if the sensor to which the CPU 22 belongs is the sensor 12), For example, when the detection signal is off, background learning is performed. The background learning condition is for determining that there is substantially no moving person or object. For example, the value of the light reception signal is equal to or higher than a predetermined background learning upper limit value or a predetermined value. The condition for the background learning is determined when the state below the lower limit for background learning is continued for a predetermined time, and the change in the value of the received light signal during the predetermined time is within a predetermined allowable range. It is filled. In the background learning, when the learning condition is satisfied, for example, an average value of the received light signal for a predetermined time thereafter is obtained, and the threshold value is updated based on the average value.

  FIG. 4 shows a timing chart when background learning is performed in the sensor 10 when there is a passerby from the outside to the inside of the building 2 at a very slow speed, for example.

  When a passerby enters the detection area 20, the detection signal of the sensor 12 is turned on and output at time t1. As a result, the door 6 is opened. Eventually, a passer-by passes through the opened door 6 and the rear part of the body remains in the detection area 20, while the front part enters the detection area 18, and at time t2, the detection signal is also turned on and output in the sensor 10. .

  Since only the rear part of the body of the passer-by travels slowly in the detection area 18, the light reception signal at the sensor 12 becomes large and exceeds the upper limit for background learning. If this continues for the predetermined time or more and the amount of change in the received light signal is within the allowable range during this time, the learning condition is satisfied at time t3. At this time, the CPU 22 of the sensor 12 inputs the detection signal of the other sensor, that is, the sensor 10, and determines whether it is on. In this case, since it is on, the learning operation is not performed. The start of the operation is temporarily stopped, and it waits for the learning condition to be satisfied again.

  On the other hand, the learning condition is satisfied at time t4 as the passer-by of the passer-by of the sensor 10 progresses. At this time, since the detection signal of the sensor 12 is on, the start of the learning operation is temporarily stopped. And wait for the learning condition to be satisfied again.

  In the sensor 12, the learning condition is satisfied again at time t5, but since the detection signal of the sensor 10 is on at this time, the learning operation is not started and the next waiting for the learning condition to be satisfied is waited for. However, eventually, the passerby completely leaves the detection area 20 at time t6, so that the detection signal is turned off and the sensor 12 does not perform background learning after all.

  On the other hand, in the sensor 10, the learning condition is satisfied again at time t7 (the learning condition is satisfied in a state where the rear part of the passer's body exists in the detection area 18). At this time, since the detection signal of the sensor 12 is off, a learning operation is performed. As a result, the detection signal of the sensor 10 is turned off at time t8. At this time, a small part of the body of the passerby may remain in the detection area 18, but this hardly affects the update of the threshold value. Note that when the passerby passes from the inside to the outside of the building, learning is performed in the sensor 12 in the same manner as described above.

  In order to perform such control, the CPU 22 in the sensors 10 and 12 performs an operation as shown in the flowchart of FIG.

  That is, initialization is first performed (step S2). For example, a threshold value or the like is read from the EEPROM 28 and stored in the RAM 26. Next, infrared rays are projected and received by the sensors 10 and 12 (step S4). It is determined whether the value of the received light signal exceeds a threshold value (step S6). If the answer to this determination is yes, the contact of the detection signal input / output unit 30 is switched to output so as to output the detection signal (step S8). Next, this contact is turned on and a detection signal is output (step S10). If the answer to the determination in step S6 is no, the contact of the detection signal input / output unit 30 is switched to output (step S12), the contact is turned off, the detection signal is stopped (step S14), and the process is executed again from step S4. To do.

  Following step S10, it is determined whether the learning conditions described above are satisfied (step S16). If the answer to this determination is no, the process is executed again from step S4. In the execution from step S4 executed again, if a passerby or the like still exists in the detection area 18 or 20, steps S6, S8, and S10 are executed, and the state where the detection signal is output continues. This is the period from time t1 to t3 and the period from time t2 to t4.

  If the answer to the determination in step S16 is yes, since the learning condition is satisfied, the contact of the detection signal input / output unit 30 is switched to input (step S18), and the detection signal of the other sensor is input (step S20). . That is, when the CPU 22 that has executed Step S20 belongs to the sensor 10, the detection signal of the sensor 12 is input. When the CPU 22 that has executed Step S20 belongs to the sensor 12, the detection signal of the sensor 10 is input.

  Based on the input detection signal of the other sensor, it is determined whether the other sensor is in a detection state (step S22). If the answer to this determination is yes, since there are passersby in the detection area of the other sensor, the contact of the detection signal input / output unit 30 is switched to output without performing learning (step S24). The contact is turned on (step S26), and the process is executed again from step S4. Also in this case, the state where the detection signal is output continues as shown at times t3, t4, and t5. It should be noted that steps S4 and S6 are executed after the state at time t5, and the case where steps S12 and S14 are executed as a result of the determination at step S6 corresponds to time t6.

  If the answer to the determination in step S22 is no, learning is performed (step S28). This start is time t7. As a result, at time t8, learning ends and no detection signal is output. Step S16 corresponds to learning condition determination means, and step S22 corresponds to temporary stop means.

  If the answer to the determination at step S22 is yes, steps S24 and S26 are executed, and again from step S4.

  A flowchart of the second embodiment is shown in FIG. In this embodiment, the configuration is the same as that of the first embodiment, but the control performed by the CPU 22 is slightly different. In the first embodiment, after the learning condition is satisfied, it is determined whether the other sensor is in the detection state. In the second embodiment, it is determined whether the other sensor is in the detection state. After it is determined that the state is not detected, it is determined whether the learning condition is satisfied.

  That is, as shown in FIG. 6, steps S2, S4, S6, S8, S10, S12, and S14 are performed in the same manner as in the first embodiment. However, after turning on the contact in step S10, it is determined whether a predetermined time T1 has elapsed (step S30), and the loop of steps S10 and S30 is repeated until the answer to this determination is yes. That is, the state in which the detection signal is output is continued for T1 time. Thereafter, the contact is switched (step S32), the detection signal of the other sensor is input (step S34), the contact is switched to the output side (step S36), the contact is turned on, and the detection signal is output (step S38). . Therefore, the state where the detection signal is substantially output is continued. Thereafter, it is determined whether or not the other sensor is in a detection state (step S40). If the answer to the determination is yes, the process is executed again from step S4. As a result, if a passerby is not present in the detection area, steps S12 and S14 are executed, and output of the detection signal is stopped. If a passerby is present in the detection area, steps S8, S10, and S30 are executed. Then, the detection signal is output again for the time T1. Thereafter, Steps S32, S34, S36, and S40 are executed to determine whether the other sensor is in a detection state. Here, if it is determined that the other sensor is in the detection state, it is determined whether the learning condition is satisfied (step S42). If the answer to this determination is no, the process is executed again from step S4. If the answer to this determination is yes, learning is performed (step S44), and the learning is executed again from step S4 after completion of the learning.

  A third embodiment is shown in FIGS. 7 (a) and 7 (b). In the first embodiment, the detection area 18 of the sensor 10 is inside the door 6 and does not cover the door 6. On the other hand, in the third embodiment, the detection area 18a of the sensor 10 starts from the sensor 10, passes through a position vertically above the track of the door 6, and its lowermost end is the lowermost end of the detection area 20. It is formed so as to almost overlap. This is possible by adjusting the installation angle of the projector 14 and the light receiver 16. By forming the detection area 18a in this way, a passerby stopped on the door 6 can be reliably detected by the sensor 10, and the sensor 12 learns when there is a person who has stopped on the door 6 for a long time. Even if the conditions are satisfied, the sensor 12 does not learn and safety is improved. Of course, in the sensor 10, the learning condition is never satisfied, so that the learning is not performed in the sensor 10. Other configurations and operations are the same as those in the first or second embodiment.

  In addition, as shown in FIG. 8A, the detection area 18 a is formed in a state where the lowermost end of the detection area 18 a is located on the track of the door 6 but does not overlap the lowermost end of the detection area 20. Alternatively, as shown in FIG. 4B, the detection area 18a is formed in a state where the lowermost end of the detection area 18a is located on the track of the door 6 and a part thereof overlaps the lowermost end of the detection area 20. can do. If formed in this way, it is possible to reliably detect a state in which a passerby has fallen on the track of the door 6. Therefore, even if the passerby has fallen for a long time, the sensors 10 and 12 do not learn by mistake.

  In each of the above embodiments, the sensor is a reflective sensor using a projector and a light receiver. However, the sensor is not limited to this, and a distance measuring sensor using a projector and a position detector is used. Things can also be used. In this case, the sensors 10 and 12 are provided on the invisible 8, but the present invention is not limited to this. For example, the sensors 10 and 12 may be attached to columns or the like on both sides of the door opening.

It is a side view of the automatic door which implemented the sensor for automatic doors of a 1st embodiment of the present invention. It is a block diagram of the sensor for automatic doors of FIG. It is a top view which shows the detection area formed with the sensor for automatic doors in FIG. It is a timing chart of the sensor for automatic doors of FIG. It is a flowchart of the sensor for automatic doors of FIG. It is a flowchart of the sensor for automatic doors of the 2nd Embodiment of this invention. It is the side view of the automatic door which implemented the sensor for automatic doors of the 3rd Embodiment of this invention, and the top view which shows the detection area formed by this sensor. It is a top view which shows the detection area formed by the modification of the sensor for automatic doors of 3rd Embodiment.

Explanation of symbols

4 Opening 6 Automatic door 10 Infrared sensor (first sensor)
12 Infrared sensor (second sensor)
18 First detection area 20 Second detection area 22 CPU (determination means, temporary stop means)

Claims (4)

Each has a detection area adjacent to the track of the automatic door, and has first and second sensors for generating a detection signal when a person or object is detected in the corresponding detection area, and at least the first sensor is A sensor which is an infrared sensor having an infrared light projecting means and an output means for generating an output by receiving the infrared light ;
At least the first sensor is
Means for determining whether or not the learning condition is satisfied;
Learning means for learning the background of the corresponding detection area when the determination means determines that the learning condition is satisfied;
When the second sensor is in a detection state, a temporary stop means for temporarily stopping the determination means or the learning means,
Comprising the determination means, the as learning condition, the output signal of the output means is a predetermined time continues a predetermined lower limit value or less in a state where more than the upper limit value or a predetermined and in the predetermined time An automatic door sensor for determining whether or not a change in the value of an output signal is within a predetermined allowable range .   2. The automatic door sensor according to claim 1, wherein when the second sensor is in a non-detection state, the temporary stop means starts the operation of the determination means or the learning means.   2. The automatic door sensor according to claim 1, wherein the first and second sensors correspond to an area on the side where the first and second sensors are provided, of the two areas partitioned by the automatic door. An automatic door sensor forming the detection area.   2. The automatic door sensor according to claim 1, wherein the first and second sensors are provided above the automatic door, and a detection area by the first sensor is, among two regions partitioned by the automatic door, The detection area formed by the second sensor is formed in the area on the side where the first sensor is provided, and the area on the track includes the area on the track, or the area on the side where the first sensor is provided passing vertically above the track. A sensor for automatic doors formed in a wide area.

Images