KR20120093203A - Air conditioner - Google Patents

Abstract

An indoor unit provided with an imaging device, the imaging device includes human body detecting means for detecting the presence of a person and obstacle detecting means for detecting the presence or absence of an obstacle. And the wind direction changing means installed in the indoor unit based on the detection signal of the human body detecting means and the detection signal of the obstacle detecting means. In addition, when the operation of the air conditioner is stopped, the imaging device is covered with a part of the indoor unit. In addition, it set so that the imaging device may face the same direction at the start of operation of an air conditioner.

Description

Air Conditioner {AIR CONDITIONER}

The present invention relates to an air conditioner in which an indoor unit is provided with a human body detecting device for detecting the presence of a person and an obstacle detecting device for detecting the presence or absence of an obstacle, and is based on the detection result of the human body detecting device and the detection result of the obstacle detecting device. It relates to an air conditioner to control the wind direction change wing.

A conventional air conditioner is provided with a human position detecting means and an obstacle detecting means in an indoor unit, and controls the wind direction changing means based on detection signals of both the human position detecting means and the obstacle position detecting means, thereby improving the air conditioning efficiency. There is.

In the air conditioner, when the heating operation is started, the human position detecting means first determines whether or not there is a person in the room, and when there is no human, the obstacle position detecting means determines whether there is an obstacle, and there is no obstacle. In this case, the wind direction changing means is controlled so that the air conditioning wind is spread throughout the room.

In addition, if there is no person but an obstacle that can be avoided is detected, the wind direction changing means is controlled in a direction where there is no obstacle, and if an obstacle that cannot be avoided is detected, the air conditioning wind does not directly touch the obstacle. In addition, the wind direction changing means is controlled so that the air conditioning wind is spread throughout the room.

Furthermore, if there is a person, it is determined whether there is a member region, and if there is no member region, the wind direction changing means is controlled to spread the air conditioning wind throughout the room, and if there is a member region, It is determined whether there is an obstacle, and if there is an obstacle, the wind direction control means is controlled in the direction of the obstacle so that the air conditioner wind does not strongly reach the obstacle, and if there is no obstacle, the wind direction control means is controlled in the direction without the obstacle (for example, , Patent Document 1).

Patent Document 1: Japanese Utility Model Publication No. Hei 3-72249

In the case of the air conditioner described in Patent Literature 1, an ultrasonic sensor is employed as a distance measuring device as a means for detecting a distance to a person or an obstacle, and the ultrasonic sensor is driven to perform obstacle detection scanning in the entire area of the room. The distance measuring device, such as an ultrasonic sensor, has a narrow measurement range and cannot detect and grasp a person or an obstacle in the entire area of the room without a complicated and time-consuming scan.

In addition, in a configuration in which a distance measuring device such as an ultrasonic sensor is exposed regardless of driving or stopping of the air conditioner, the distance measuring device is affected by dust, cigarette smoke, or the like, and causes a decrease in recognition performance.

In addition, when a movable device is adopted as a distance measuring device such as an ultrasonic sensor, and the detection operation of the distance measuring device is stopped after scanning the whole area in the room, the distance measuring device is directed to the corner of the room, which causes discomfort for the occupants. . In addition, every time scanning of the distance measuring device, such as an ultrasonic sensor, is complete | finished, if the direction which a distance measuring device faces differs, a resident will be inferior.

SUMMARY OF THE INVENTION The present invention has been made in view of such problems of the prior art, and detects the presence of a person (human detection means) by using a fixed imaging device or a driving imaging device provided in an indoor unit, and checks for the presence or absence of an obstacle. In the case of an imaging device which is configured to detect (obstacle detecting means) and does not expose the imaging device when the air conditioner is stopped, and to operate the air conditioner, the imaging device always moves in the same direction. It aims at providing the air conditioner which can suppress the fall of the recognition performance of an imaging device, and can give a resident a sense of security by setting so that it may face.

In order to achieve the above object, the present invention detects the presence of a person (human detection means) using a fixed imaging device or a driving imaging device provided in an indoor unit, and detects the presence or absence of an obstacle (obstacle detection means). An air conditioner configured to control the wind direction change blade provided in the indoor unit based on the detection result of the human body detecting means and the obstacle detecting means, and when the air conditioner stops operating, the imaging device is covered with a part of the indoor unit. In the case of the imaging device to be driven, the imaging device is set to face the same direction at the start of the operation of the air conditioner.

According to the present invention, the imaging device is not exposed when the air conditioner is stopped, thereby suppressing the deterioration in recognition performance of the imaging device, and at the start of operation of the air conditioner, the imaging device is the same. By setting the direction so that the residents can feel secure. In addition, the anxiety from the viewpoint of privacy that the imaging device is exposed even when the driving is stopped and the room may be photographed at all times can also be eliminated.

1 is a front view of an indoor unit of an air conditioner according to the present invention;
2 is a longitudinal sectional view of the indoor unit of FIG. 1;
FIG. 3A is a longitudinal sectional view of the indoor unit of FIG. 1 with the movable front panel opening the front opening and the upper and lower wings opening the tuyeres; FIG.
3B is a longitudinal sectional view of the indoor unit of FIG. 1 in a state in which a lower wing constituting the upper and lower wings is set downward;
4 is a cross-sectional view of the imaging device provided in the indoor unit of FIG. 1;
5 is a flowchart showing a flow of a process of estimating a person's position in the present embodiment;
6 is a schematic diagram for explaining a background difference process in estimating a person's position in the present embodiment;
7 is a schematic diagram for explaining a process of creating a background image in the background difference processing;
8 is a schematic diagram for explaining a process of creating a background image in the background difference processing;
9 is a schematic diagram for explaining a process of creating a background image in the background difference processing;
10 is a schematic diagram for explaining an area division process in the person position estimation in the present embodiment;
11 is a schematic diagram for explaining two coordinate systems used in the present embodiment;
12 is a schematic diagram showing a distance from an imaging sensor unit to a center position of a person;
13 is a schematic diagram showing a person position determining region detected by an imaging sensor unit constituting a human body detecting means;
14 is a schematic diagram when a person is present in a person position determination area detected by an imaging sensor unit constituting a human body detecting means;
FIG. 15 is a flowchart for setting region characteristics in each region shown in FIG. 13;
FIG. 16 is a flowchart for finally determining whether a person is present in each region shown in FIG. 13;
17 is a schematic plan view of a house in which the indoor unit of FIG. 1 is installed;
18 is a graph showing a long-term cumulative result of each imaging sensor unit in the housing of FIG. 17;
19 is a schematic plan view of another housing in which the indoor unit of FIG. 1 is installed;
20 is a graph showing a long-term cumulative result of each imaging sensor unit in the housing of FIG. 19;
21 is a flowchart showing a flow of a human position estimation process using a process of extracting a human-like area from a frame image;
Fig. 22 is a flowchart showing a flow of a process for estimating person position using a process of extracting a face-like area from a frame image;
23 is a schematic diagram showing an obstacle position determining region detected by the obstacle detecting means;
24 is a schematic diagram for explaining obstacle detection by a stereo method;
25 is a flowchart showing a processing flow of distance measurement to an obstacle;
26 is a schematic diagram showing a distance from an imaging sensor unit to a position P;
27 is a schematic diagram showing a measurement result of an obstacle detecting means in an elevation view of a living space;
28 is a schematic view showing the definition of the wind direction in each position of the left wing and the right wing constituting the left and right wings;
29 is a schematic plan view of a room for explaining a wall detection algorithm for measuring a distance from an indoor unit to a surrounding wall surface to obtain a distance number;
30 is a front view of an indoor unit of another air conditioner according to the present invention;
31 is a schematic diagram showing the relationship between an imaging sensor unit and a light projecting portion;
32 is a flowchart showing a flow of processing of distance measurement to an obstacle using a light projecting unit and an imaging sensor unit;
33 is a front view of an indoor unit of another air conditioner according to the present invention;
34 is a flowchart showing the flow of processing of the human body distance detecting means using the human body detecting means;
35 is a schematic diagram for explaining a process of estimating a distance from an imaging sensor unit to a person using v1, which is the v-coordinate at the top of the image, FIG.
36 is a flowchart showing the flow of processing of the obstacle detecting means using the human body detecting means;
37 is a schematic diagram for explaining a process of estimating a height v2 of a person on an image using distance information from the imaging sensor unit estimated by the human body distance detecting unit to the person;
38 is a schematic diagram for explaining a process of estimating whether an obstacle exists between an imaging sensor unit and a person;
39 is a schematic diagram for explaining a process of estimating whether an obstacle exists between the imaging sensor unit and the person.

The first invention detects the presence of a person (human detection means) using an imaging device to be fixed to an indoor unit or a driving imaging device, and also detects the presence or absence of an obstacle (obstacle detection means), and detects a human body. The air conditioner which controls the wind direction change vane provided in the indoor unit based on the detection result of the detection result and the obstacle detection means, It is set as the structure which covers an imaging device with a part of indoor unit at the time of stopping operation of an air conditioner.

By this structure, the fall of the recognition performance of an imaging device can be suppressed, and a security can be provided to a resident. In addition, the image pickup device is exposed even when the vehicle is stopped, so that the anxiety from the viewpoint of privacy that the room may be photographed at all times can be eliminated.

In the second invention, in the case of the imaging device to be driven, the imaging device is set to face the same direction at the start of operation of the air conditioner.

In the third invention, in the case of the imaging device to be driven, the imaging device is set to face the front of the indoor unit at the start of operation of the air conditioner.

By this structure, a resident does not receive discomfort.

4th invention is a 2nd and 3rd invention by setting so that the optical axis of an imaging device may be substantially perpendicular to an installation surface looking at the indoor unit from the top at the start of operation of an air conditioner in the case of the imaging device to drive. It can produce the same effect.

In the fifth invention, in the case of an imaging device to be driven, the imaging device is configured to be able to change its direction in a predetermined angle range in the vertical direction and the horizontal direction, and at the start of operation of the air conditioner, the direction of the imaging device is changed. By setting to the upper limit position or the lower limit position in the vertical direction, the same effects as in the second and third inventions can be achieved.

In addition, when the viewing angle in the vertical direction of the imaging device is sufficient, the structure for operating the imaging device only needs to be in the horizontal direction. Similarly, when the viewing angle in the horizontal direction of the imaging device is sufficient, the structure for operating the imaging device is in the vertical direction. Needless to say that it is all. It goes without saying that the imaging device may be fixed when the viewing angle of the imaging device is sufficient in both the vertical direction and the horizontal direction.

In the sixth aspect of the invention, when the air conditioner is stopped, the imaging device is covered with a movable front panel or an up-down wind direction change wing, so that the imaging device is not affected by dust, cigarette smoke, or the like, thereby suppressing a decrease in recognition performance. can do. In addition, the image pickup device is exposed even when the vehicle is stopped, so that the anxiety from the viewpoint of privacy that the room may be photographed at all times can also be eliminated.

EMBODIMENT OF THE INVENTION Hereinafter, embodiment of this invention is described, referring drawings.

<Overall Configuration of Air Conditioner>

An air conditioner used in a general home is usually composed of an outdoor unit and an indoor unit connected to each other by a refrigerant pipe, and FIGS. 1 to 3B show the indoor unit of the air conditioner according to the present invention.

The indoor unit has a main body 2 and a movable front panel (hereinafter, simply referred to as a 'front panel') 4 that can open and close the front opening portion 2a of the main body 2, and at the time of stopping the air conditioner, The front panel 4 is in close contact with the main body 2 to close the front opening 2a. On the other hand, during operation of the air conditioner, the front panel 4 moves away from the main body 2 and the front opening 2a. To open. 1 and 2 show a state in which the front panel 4 closes the front opening 2a, and FIGS. 3A and 3B show a state in which the front panel 4 opens the front opening 2a. .

1 to 3B, inside the main body 2, the heat exchanger 6 and the indoor air taken in from the front opening 2a and the surface opening 2b are heat exchanged with the heat exchanger 6. The upper and lower wings 12 for opening and closing the indoor fan 8 for blowing air into the room, the blower 10 for blowing the heat-exchanged air into the room, and changing the air blowing direction up and down, and blowing the air. It is provided with the left-right wing | wing 14 which changes a direction from side to side, and is received from the front opening part 2a and the surface opening part 2b between the front opening part 2a and the surface opening part 2b, and the heat exchanger 6. As shown in FIG. The filter 16 for removing the dust contained in indoor air is provided.

In addition, the upper part of the front panel 4 is connected to the upper part of the main body 2 via the two arms 18 and 20 provided in the both ends, and drive control of the drive motor (not shown) connected to the arm 18 is carried out. In the air conditioner operation, the front panel 4 moves obliquely upward from the front from the position at the time of stopping the air conditioner (closed position of the front opening 2a).

In addition, the upper and lower wings 12 are composed of an upper wing 12a and a lower wing 12b, and are mounted to the lower part of the main body 2 so as to be swingable. The upper vane 12a and the lower vane 12b are connected to different drive sources (e.g., stepping motors), and are each independently controlled by a control device (the first substrate 48 described later, for example, a microcomputer), which is incorporated in the indoor unit. Angle is controlled. 3A and 3B, the changeable angle range of the lower blade 12b is set larger than the changeable angle range of the upper blade 12a.

In addition, the driving method of the upper blade 12a and the lower blade 12b is mentioned later. In addition, the upper and lower wings 12 may be configured by three or more upper and lower wings, and in this case, at least two sheets (in particular, the uppermost wing and the lowermost wing) can be independently angle controlled. It is desirable to have.

In addition, the left and right wings 14 are comprised by the total of 10 blades arrange | positioned 5 each from left to right from the center of an indoor unit, respectively, and are attached to the lower part of the main body 2 so that rotation is possible. In addition, the left and right five sheets are connected to another drive source (for example, a stepping motor) using one unit as a unit, and the left and right five blades are independently angle-controlled by a control device built in the indoor unit. In addition, the drive method of the left and right wings 14 is also mentioned later.

<Configuration of the human body detecting means>

As shown in FIG. 1, the imaging sensor unit 24 is assembled to the imaging device 25 at both left and right ends or one of its lower sides as viewed from the front of the main body. The imaging device 25 is illustrated in FIG. 4. Explain while referring to.

The imaging sensor unit 24 is comprised of the circuit board 51, the lens 52 attached to the circuit board, and the imaging sensor 53 mounted in the lens. In addition, the human body detection means judges the presence or absence of a person by the circuit board 51 based on the difference process mentioned later, for example. That is, the circuit board 51 acts as presence or absence determination means which judges the presence or absence of a person.

The imaging sensor 53 is a support (sensor holder) 54 of a sphere that is rotatably supported, and an imaging direction changing means (driving means) that changes the direction of the imaging sensor 53 so as to scan-drive all the necessary fields of view. ).

Moreover, the support body 54 has the horizontal (horizontal) rotation axis 55 and the vertical (vertical) rotation axis 56 extended in the direction orthogonal to the horizontal rotation axis 55, and for horizontal rotation The rotating shaft 55 is connected to and driven by the horizontal rotating motor 57, and the vertical rotating rotating shaft 56 is connected to and driven by the vertical rotating motor 58. That is, the imaging direction changing means is comprised by the horizontal rotation motor 57, the vertical rotation motor 58, etc., The direction angle of the imaging sensor 53 can be changed in two dimensions, and the imaging sensor 53 You can recognize the direction in which) is facing.

<Personal Position Estimation by Image Sensor Unit>

Hereinafter, a method for estimating a person's position by the imaging sensor unit will be described. However, for simplicity (assuming that the visual field of the imaging sensor is sufficiently secured in both the horizontal and vertical directions), the driving of the imaging sensor is not necessary, that is, fixed. The case will be described.

In order to perform the human position estimation by the imaging sensor unit 24, the difference method which is a well-known technique is used. This is a difference process between a background image which is an image without a person and an image picked up by the imaging sensor unit 24, and it is assumed that a person exists in an area where a difference occurs.

5 is a flowchart showing the flow of the process of estimating a person's position in the present embodiment. In step S101, a pixel having a difference is detected in the frame image by using the background difference process. Background difference processing is not present in a background image by comparing a background image photographed under specific conditions with an image captured in a situation where the background image and the imaging conditions such as the field of view, the viewpoint, and the focal length of the imaging sensor unit are the same. This is a technique for detecting an object present in a captured image. In order to detect a person, an image without a person as a background image is created.

6 is a schematic diagram for explaining a background difference process. Fig. 6A shows the background image. Here, the field of view is set to be approximately equal to the air conditioning space of the air conditioner. In this figure, 101 denotes a window existing in the air conditioning space and 102 denotes a door. Fig. 6B shows a frame image picked up by the imaging sensor unit. Here, the field of view, the viewpoint, the focal length, and the like of the imaging sensor unit are the same as those of the background image of Fig. 6A. 103 represents a person present in the air conditioning space. In the background difference processing, the person is detected by creating the difference image of Figs. 6 (a) and 6 (b). Fig. 6C shows a difference image, where white pixels represent pixels with no difference, and black pixels show pixels with difference. It can be seen that the area of the person 103 which does not exist in the background image but exists in the picked-up frame image is detected as the area 104 having a difference. In other words, it is possible to detect the person area by extracting the area where the difference is generated from the difference image.

In addition, the above-mentioned background image can be created by using the interframe difference processing. 7-9 is a schematic diagram for demonstrating this process. 7 (a) to 7 (c) are schematic diagrams showing images of three consecutive frames captured by the imaging sensor unit in a scene in which the person 103 moves the front of the window 101 from right to left. to be. Fig. 7 (b) shows an image of the next frame of Fig. 7 (a) and Fig. 7 (c) shows an image of the next frame of Fig. 7 (b). 8A to 8C show interframe difference images in which interframe difference processing is performed using the image of FIG. 7. White pixels represent pixels with no difference, and black pixels 105 represent pixels with differences. Here, if only the object moving in the field of view is a person, it is considered that there is no person in the region where no difference occurs in the interframe difference image. Here, the background image is replaced with the image of the current frame in the region where the difference between frames does not occur. By performing this process, a background image can be automatically created. 9 (a) to 9 (c) are diagrams schematically showing updating of a background image in each frame of FIGS. 7 (a) to 7 (c), respectively. The area 106 indicated by the oblique line indicates the area where the background image has been updated, the black area 107 indicates the area where the background image has not been created yet, and the white area 108 indicates the area where the background image has not been updated. That is, the total area | region of the black area | region 107 and the white area | region 108 of FIG. 9 becomes the same as the black area | region of FIG. As shown in the figure, when the person is moving, it can be seen that the black area 107 gradually decreases and a background image is automatically created.

Next, in step S102, the obtained difference area is divided into areas, and when there are a plurality of persons, they are divided into a plurality of difference areas. This may be achieved by using a known image clustering method. For example, the difference image may be divided in accordance with a rule that "a pixel having a difference and a pixel having a difference present in the vicinity thereof are the same area". 10 is a schematic view of performing this area division process. Fig. 10A shows the difference image calculated by the difference processing, and black pixels 111 and 112 are pixels in which difference is generated. Fig. 10 (b) shows that when Fig. 10 (a) is obtained as a difference image, region division is performed in accordance with the above-mentioned rule that "the pixel with the difference and the pixel with the difference existing in the vicinity are the same area". The result of the experiment is shown. Here, it is determined that the horizontal stripe region 113 and the vertical stripe region 114 are different regions. At this time, you may perform the noise removal process, such as the morphology process widely used for image processing.

Next, in step S103, the position of the detected person is detected by calculating the center position of each obtained area. Perspective projection transformation may be used to detect the position of the person from the center position of the image.

Two coordinate systems are described to illustrate the perspective projection transformation. 11 is a schematic diagram for explaining two coordinate systems. First, consider the image coordinate system. This is a two-dimensional coordinate system in the picked-up image, where the pixel on the upper left of the image is the origin, u in the right direction, and v in the downward direction. Next, a camera coordinate system, which is a three-dimensional coordinate system based on the camera, is considered. This sets the focal position of the imaging sensor unit to the origin, Zc as the optical axis direction of the imaging sensor unit 24, Yc as the camera upward direction, and Xc as the camera left direction. At this time, the following relationship is established by perspective projection conversion.

Here, f is the focal length [mm], (u0, v0) is the image center [Pixel] in the image coordinates, and (dpx, dpy) represents the size [mm / Pixel] of one pixel of the image pickup device. Note that Xc, Yc, and Zc are unknowns. When Equation 1 knows the coordinates (u, v) on the image, the actual three-dimensional position corresponding to the coordinates passes through the origin of the camera coordinate system. It can be seen that on a straight line.

As shown in Figs. 12A and 12B, the center position of the person on the image is (ug, vg) and the three-dimensional position in this camera coordinate system is (Xgc, Ygc, Zgc). Here, FIG. 12 (a) has shown the schematic diagram which looked at the air control space from the side, and FIG. 12 (b) has shown the schematic diagram seen from the top. In addition, it is assumed that the height where the imaging sensor unit is installed is the same in the horizontal direction in the H and Xc directions, and the optical axis Zc is provided at an angle of θ from the vertical direction. In addition, the direction in which the imaging sensor unit 24 is directed is the angle in the vertical direction (angle measured from the vertical line) and the angle in the horizontal direction (viewed from the room to the right from the reference line on the front side). Measured angle) β. In addition, if the center height of a person is h, the distance L from the imaging sensor unit which is a three-dimensional position in an air conditioning space, and a center position can be computed by following Formula.

Here, in consideration of the fact that the imaging sensor unit is normally installed at a height of H = about 2m, and the height h of the center of the person is about 80cm, equations (3) and (5) are provided with the imaging sensor unit 24. When the height H and the center height h of the person are defined, it indicates that the center positions L and W of the person in the air conditioning space are uniquely obtained from the center positions ug and vg on the screen. . 13 (a) and 13 (b) show which areas exist in the air conditioning space when the central position on the image exists in each area of A to G. FIG. 14 (a) and 14 (b) show schematic diagrams when a person is present. In FIG. 14A, since the center position of the person exists in the areas A and F, it is determined that the person exists in the areas A and F of FIG. 13B. On the other hand, in FIG. 14B, since the center position of the person exists in the area D, it is determined that the person exists in the area D of FIG. 13B.

FIG. 15 is a flowchart for setting area characteristics to be described later in each of the areas A to G using the imaging sensor unit, and FIG. 16 determines whether any area of the areas A to G is present using the imaging sensor unit. This is a flowchart to describe a person's position determination method with reference to this flowchart.

In step S1, the presence or absence of a person in each area is first determined by the above-described method at a predetermined period T1 (for example, 200 milliseconds if the frame rate of the imaging sensor unit 24 is 5 fps). Based on this determination result, each area | region A-G is divided into the 1st area | region (often frequent place) where a person exists frequently, and the 2nd area | region where the time where a person exists is short (an area which a person just passes, staying time). It is discriminated into a passage area | region, such as a short area | region, and the 3rd area | region (a wall, a window, non-living area which few people, such as a window) have very short human time. Hereinafter, the first area, the second area, and the third area are referred to as the life division I, the life division II, and the life division III, respectively, and the life division I, the life division II, and the life division III are the areas of the area characteristics I, It may also be referred to as the region of region characteristic II and the region of region characteristic III. In addition, life classification I (area characteristics I) and life classification II (area characteristics II) are defined as living areas (areas where people live), while life classification III (area characteristics III) is defined as non-living areas. (Areas not to live), the areas of living may be largely classified according to the frequency of presence or absence of a person.

This determination is performed after step S3 in the flowchart of FIG. 15, and this determination method will be described with reference to FIGS. 17 and 18.

FIG. 17 shows the case where the indoor unit of the air conditioner according to the present invention is installed in an LD of 1LDK consisting of one room, an LD (living room and a dining room), and a kitchen, and an area represented by an ellipse in FIG. 17 is a subject. We show frequent place that we reported.

As described above, the presence or absence of a person in each region A to G is determined for each cycle T1, but as a result (determination) of the cycle T1, 1 (with response) or 0 (no response) is output, and this is repeated several times. After the repetition, all sensor outputs are cleared in step S2.

In step S3, it is determined whether the cumulative operating time of the predetermined air regulator has elapsed. If it is determined in step S3 that the predetermined time has not elapsed, the process returns to step S1. If it is determined that the predetermined time has elapsed, the result of the reaction accumulated in the corresponding predetermined time in each of the areas A to G is the minimum physical quantity that causes two reactions. By comparing, each area | region A-G is discriminated as one of the life divisions I-III respectively.

Referring to FIG. 18 showing the long term cumulative result, the minimum physical quantity causing the first reaction and the minimum physical quantity causing the second reaction smaller than the minimum physical quantity causing the first reaction are set, and in step S4 It is determined whether the long-term cumulative result of each area | region A-G is more than the minimum physical quantity which causes a 1st reaction, and it is determined in step S5 that the area | region determined as many is life division I. In addition, if it is determined in step S4 that the long-term cumulative result of each of the regions A to G is less than the minimum physical quantity causing the first reaction, the minimum physical amount of the long-term cumulative result of each of the regions A to G causes the second reaction in Step S6. It is determined whether there are more, and the area judged to be large is determined as life category II in step S7, while the area judged as small is determined as life category III in step S8.

In the example of FIG. 18, areas C, D, and G are discriminated as life division I, areas B and F are discriminated as life division II, and areas A and E are discriminated as life division III.

19 shows the case where the indoor unit of the air conditioner according to the present invention is installed in another LD of 1LDK, and FIG. 20 shows the result of determining the respective areas A to G based on the long term cumulative result in this case. have. In the example of FIG. 19, areas B, C, and E are identified as life division I, areas A and F are identified as life division II, and areas D and G are identified as life division III.

In addition, although the above-mentioned determination of the area characteristics (life division) is repeated every predetermined time, the determination result hardly changes unless a sofa, a dining table, or the like arranged in the room to be discriminated is moved.

Next, with reference to the flowchart of FIG. 16, the final determination of the presence or absence of a person in each area | region A-G is demonstrated.

Steps S21 to S22 are the same as steps S1 to S2 in the flowchart of FIG. 15 described above, and thus description thereof is omitted. In step S23, it is determined whether or not the result of the reaction of the predetermined number M (e.g., 45 times) of the period T1 is obtained, and if it is determined that the period T1 does not reach the predetermined number M, the process returns to step S21 while the period T1 is predetermined. When it is determined that the number M has been reached, in step S24, the cumulative reaction period number of times is calculated as the cumulative reaction period number as the sum of the reaction results in the period T1 × M. When the calculation of the cumulative reaction period number of times is repeated several times, and it is determined in step S25 whether the calculation result of the cumulative reaction period number of times for a predetermined number of times (for example, N = 4) has been obtained, and it is determined that the predetermined number of times has not been reached, Returning to step S21, if it is determined that the predetermined number of times has been reached, the presence or absence of a person in each of the areas A to G is estimated based on the area characteristics already determined in step S26 and the cumulative reaction period of the predetermined number of times.

In addition, by subtracting 1 from the calculation number N of the cumulative reaction period number N in step S27, and returning to step S21, the calculation of the cumulative reaction period number of times for a predetermined number of times is repeatedly performed.

Table 1 shows the history of the reaction result of the latest one time (time T1 x M), and, for example, (Sigma) A0 in Table 1 means the number of times of the cumulative reaction period in the area A.

Here, the number of cumulative reaction periods for the first time immediately before ΣA0 is ΣA1, and the number of cumulative reaction periods for the previous one is ΣA2... In the case of N = 4, it is determined that there is a person, if there is at least one cumulative reaction period number of times in the life division I in the past four times of history (ΣA4, ΣA3, ΣA2, ΣA1). In addition, about life division II, if there is more than one cumulative reaction period number of times in the past four times, it is judged that there is person, and about life division III, it is more than two times in the past four times history If there are three or more cumulative reaction periods, it is determined that there is a person.

Next, after time T1xM from the above-mentioned judgment of presence of a person, estimation of presence of a person is similarly performed from the past four times of history, a classification of life, and the number of cumulative reaction periods.

That is, in the indoor unit of the air conditioner according to the present invention, the past of the cumulative reaction period count of each region obtained by accumulating the region determination result for each predetermined period for a long term and the region determination result for each predetermined cycle for N times. By estimating the location of a person from histories, the position estimation result of a person with a high probability is obtained.

Table 2 determines the presence or absence of a person in this way, and shows the time required for existence estimation and the time required for absence estimation when T1 = 0.2 seconds and M = 45 cycles.

Thus, after dividing the air regulation area | region by the indoor unit of the air conditioner which concerns on this invention into the some area | region A-G by the imaging sensor unit, the area | region characteristic of each area | region A-G (life division I-III) In addition, the time required for the existence estimation and the time required for the member estimation are changed according to the region characteristics of the respective regions A to G.

In other words, after changing the air conditioning setting, it takes about 1 minute to reach the wind, so changing the air conditioning setting in a short time (for example, a few seconds) not only impairs comfort, As for the place, it is preferable not to perform air regulation very much from an energy saving viewpoint. Therefore, the presence or absence of a person in each area | region A-G is first detected, and the air conditioning setting of the area | region in which a person exists is especially optimized.

Specifically, on the basis of the time required for estimating the presence or absence of the area identified as life division II, in the area determined as life division I, the presence of a person is estimated at a shorter time interval than the area determined as life division II. On the other hand, when a person disappears from the area, by estimating the person's absence at a time interval longer than the area determined as the life classification II, the time required for the existence estimation is shortened, and the time required for the absence estimation is long. Will be set. On the contrary, in the area judged as life division III, the presence of a person is estimated at a longer time interval than the area determined as life division II, whereas when a person disappears from this area, the time is shorter than the area determined as life division II. By estimating the absence of a person at intervals, the time required for presence estimation is lengthened and the time required for absence estimation is set short. Furthermore, as described above, according to the long-term cumulative result, the division of life of each area is changed, and accordingly, the time required for the existence estimation and the time required for the absence estimation are also variablely set.

In the above description, although the difference method was used as the estimation of the position of the person by the imaging sensor unit, of course, other methods may be used. For example, it is possible to extract an area such as a person from the frame image by using image data of the whole body of the person. As such a method, for example, a method using a histograms of oriented gradients (HOG) characteristic amount or the like is widely known (N. Dalal and B. Triggs, "Histograms of Oriented Gradients for Human Detection", In Proc. IEEE Conf. On Computer). Vision and Pattern Recognition, Vol. 1, pp. 886-893, 2005.). The HOG feature amount is a feature amount that pays attention to the edge strength in each edge direction in the local area. Even if the feature amount is learned and identified by SVM (Support Vector Machine) or the like, the person area can be detected from the frame image. Does not matter.

21 is a flowchart showing a flow of a human position estimation process using a process of extracting a human-like area from a frame image. In this figure, the same code | symbol is attached | subjected about the step like FIG. 5, and the detailed description is abbreviate | omitted here.

In step S104, the human-like area is extracted as the human area in the frame image by using the above-described HOG feature amount.

In step S103, the position of the detected person is detected by calculating the center position of the obtained human area. In order to detect the position of the person from the center position of the image, as described above, Equations 3 and 5 may be used.

Further, instead of using the whole body image data of the person, it is also possible to extract an area such as a face from the frame image. As such a technique, for example, a technique using a Haar-Like feature amount or the like is widely known (P. Viola and M. Jones, "Robust real-time face detection", International Journal of Computer Vision, Vol. 57, no. 2). , pp. 137-154, 2004.). The Haar-Like feature amount is a feature amount that pays attention to the luminance difference between the local areas. The haar-like feature may be learned and identified by SVM (Support Vector Machine) or the like to detect the person area from the frame image. .

Fig. 22 is a flowchart showing the flow of a human position estimation process using a process of extracting a face-like area from a frame image. In this figure, the same reference numerals are given to the steps as in Fig. 5, and detailed description thereof is omitted here.

In step S105, the face-like area is extracted as the face area in the frame image by using the aforementioned Haar-Like feature amount.

In step S103, the position of the detected person is detected by calculating the center position of the obtained face area. In order to detect the position of the person from the center position of the image, perspective projection transformation may be used as described above. At this time, when detecting the position of the person from the center position using the whole body area of the person, h = about 80 cm as the center height of the person, but when using the face area, h = about the height to the center of the face The position of the person is detected by using the equation (3) and (5).

<Configuration of obstacle detection means>

The obstacle detecting means for performing obstacle detection using the above-described imaging sensor unit 24 will be described. In addition, the term "obstacle" used in the present specification refers to the entire object that is blown from the air vent 10 of the indoor unit to hinder the flow of air for providing a comfortable space to the occupants, for example, furniture such as a table or sofa It is a generic term for objects other than residents such as TV, TV, and audio.

In the present embodiment, the obstacle detecting means divides the bottom surface of the living space into a subdivision as shown in FIG. 23 based on the angle α in the vertical direction and the angle β in the horizontal direction as shown in FIG. 12. Each is defined as an obstacle position discrimination area or a "position" to determine which position exists an obstacle. In addition, the overall position shown in FIG. 23 substantially coincides with the entire area of the human position determination region shown in FIG. 13 (b), and the region boundary of FIG. 13 (b) is approximately coincided with the position boundary of FIG. By corresponding the positions as follows, it is possible to easily perform the air regulation control described later, and the memory to be stored is reduced as much as possible.

Area A: Position A1 + A2 + A3

Area B: Position B1 + B2

Area C: Position C1 + C2

Area D: Position D1 + D2

Area E: Position E1 + E2

Area F: Position F1 + F2

Area G: Position G1 + G2

In addition, the area division of FIG. 23 sets the number of areas of a position more than the number of areas of a person position determination area | region, and at least two positions belong to each person position determination area | region, and these at least two obstacle position determination area | regions are removed from an indoor unit. Although it is arrange | positioned at the left and right side, air regulation control can also be performed by region-dividing so that at least 1 position may belong to a position determination area | region respectively.

In addition, in the area segmentation of FIG. 23, each of the plurality of human location determination areas is divided according to the distance to the indoor unit, and the number of areas belonging to the person location determination area in the near area belongs to the person location determination area in the far area. Although the number is set to be larger than the number of areas, the number of positions belonging to the position determining area may be the same number regardless of the distance from the indoor unit.

<Detection operation and data processing of obstacle detection means>

As described above, the air conditioner according to the present invention detects the presence or absence of a person in the areas A to G by the human body detecting means, and the presence or absence of an obstacle in the positions A1 to G2 by the obstacle detecting means. Is detected, and drive control of the upper and lower blades 12 and the left and right blades 14, which are the wind direction changing means, based on the detection signal (detection result) of the human body detection means and the detection signal (detection result) of the obstacle detection means, To provide space.

As described above, the human body detecting means can detect the presence or absence of a person by detecting a moving object in the air control space, for example, by using a person moving, while the obstacle detecting means has an image sensor unit 24. By detecting the distance of the obstacle, the person and the obstacle cannot be discriminated.

If people are mistaken as an obstacle, they may not be able to control the area where the person is, or they may hit the air conditioning wind directly on the person, resulting in inefficient air control or unpleasant air control. There is a risk of control.

Here, the obstacle detecting means is subjected to the data processing described below so as to detect only the obstacle.

First, obstacle detection means using the imaging sensor unit will be described. In order to detect an obstacle using an imaging sensor unit, the stereo method is used. The stereo method is a method of estimating the distance to a subject using the some imaging sensor units 24 and 26, and using the parallax. It is a schematic diagram for demonstrating obstacle detection by the stereo method. In the figure, the distance to the point P which is an obstacle is measured using the imaging sensor units 24 and 26. FIG. Further, f is the focal length, B is the distance between the focal points of the two imaging sensor units 24 and 26, u1 is the u-coordinate of the obstacle on the image of the imaging sensor unit 24, and u1 is also the imaging sensor unit 26. U2 and X represent the distance from the imaging sensor unit to the point P in the u-coordinate of the corresponding point on the image. In addition, the image center positions of the two imaging sensor units 24 and 26 shall be the same. At this time, the distance X from the imaging sensor unit to the point P is obtained from the following equation.

From this equation, it can be seen that the distance X from the imaging sensor unit to the point P of the obstacle depends on the parallax | u1-u2 | between the imaging sensor units 24 and 26.

In addition, searching for a corresponding point may be performed using a block matching method using a template matching method. As described above, distance measurement (position detection of obstacles) in the air conditioning space is performed by using the imaging sensor unit.

From Equations 3, 5, and 6, it can be seen that the position of the obstacle is estimated by the pixel position and parallax. I and j in Table 3 indicate pixel positions to be measured, and the angle in the vertical direction and the angle in the horizontal direction indicate the elevation angle α described above and the angle β measured from the front reference line to the right direction from the indoor unit, respectively. . That is, each pixel is set in the range of 5 degrees-80 degrees in the vertical direction and -80 degrees-80 degrees in the horizontal direction from the indoor unit, and the imaging sensor unit counts the parallax of each pixel.

That is, the air conditioner performs distance measurement (position detection of an obstacle) by measuring parallax at each pixel from the pixels [14, 15] to the pixels [142, 105].

In addition, you may limit the detection range of the obstacle detection means at the start of operation of an air conditioner to 10 degrees or more elevation angles. This is because the measurement data can be effectively used by measuring distances only in the areas where people are likely to be present at the start of operation of the air conditioner and are not likely to detect people, that is, the areas where walls are present. Since it is not an obstacle, as described later, data of a human area is not used).

Next, the distance measurement to an obstacle is demonstrated, referring the flowchart of FIG.

First, in step S41, when it is determined that there is no person in the area corresponding to the current pixel (any one of areas A to G shown in FIG. 13), the process proceeds to step S42, and when it is determined that there is a person, step S43. To fulfill. That is, since the person is not an obstacle, in the pixel corresponding to the area where the person is determined to be present, the area is determined to be no person by using the previous distance data (not updating the distance data) without performing the distance measurement. The distance measurement is performed only for the pixel to be used, and the distance data newly measured is used (to update the distance data).

That is, when determining the presence or absence of an obstacle in each obstacle position determination region, the obstacle position determination region in each obstacle position determination region is determined according to the result of the presence or absence of a person in the person position determination region corresponding to each obstacle position determination region. By determining whether or not to update the determination result of the obstacle detecting means, the presence or absence of the obstacle is effectively determined. More specifically, in the obstacle position determination region belonging to the human position determination region judged that there is no person by the human body detection means, the previous determination result by the obstacle detection means is updated with a new determination result, while the human body detection means In the obstacle position determination area belonging to the person position determination area judged to be a person, the previous determination result by the obstacle detection means is not updated with the new determination result.

In step S42, the parallax of each pixel is calculated by using the above-described block matching method, and the process proceeds to step S44.

In step S44, eight times of data are acquired by the same pixel, it is determined whether the distance measurement based on the acquired data is completed, and when it is determined that distance measurement is not completed, it returns to step S41. On the contrary, if it is determined in step S44 that the distance measurement is completed, the process proceeds to step S45.

In step S45, the accuracy of the distance estimation is improved by evaluating the reliability. In other words, when it is determined that the reliability is determined, the distance number determination processing is performed in step S46, and when it is determined that there is no reliability, the processing is performed using the distance number of the vicinity as the distance data of the pixel in step S47.

In addition, since such a process is performed by the imaging sensors 24 and 26, the imaging sensor units 24 and 26 act as obstacle position detection means.

Next, although the distance number determination process in step S46 is demonstrated, the term "distance number" is demonstrated first.

"Distance number" means the approximate distance from the imaging sensor unit to the position P where the air adjusting space is located. As shown in Fig. 26, the imaging sensor unit is installed 2 m above the floor and is located from the imaging sensor unit. If the distance to the position P is called "distance number equivalent distance" X [m], the position P is represented by the following formula.

As shown in equation (6), the distance number equivalent distance X depends on the parallax between the imaging sensor units 24 and 26. In addition, the distance number is an integer value from 2 to 12, and the distance corresponding to each distance number is set as shown in Table 4.

In addition, Table 4 shows the position of the position P corresponding to the elevation angle α determined by the v-coordinate value of each pixel by the distance number and the number 2, and the black part has a negative value where h is a negative value ( h <0), the position under the floor is shown. In addition, the setting of Table 4 is applied to the air conditioner of capacity rank 2.2kw, This air conditioner is installed only in 6 jaw rooms (diagonal distance = 4.50m), and limits distance number = 9 ( It is set as the maximum value D). That is, in the 6 jaw square, the position corresponding to the distance number ≥ 10 becomes the position (the position outside the room) beyond the wall of the room at the diagonal distance> 4.50m, and is shown in black with the distance number having no meaning at all.

In addition, Table 5 applies to the air conditioner which is capacity rank 6.3kw, This air conditioner sets only by setting distance number = 12 as limit value (maximum value D) by being installed only in 20 squares (diagonal distance = 8.49m) Doing.

Table 6 shows the limit value of the distance number set according to the capability rank of the air conditioner and the elevation angle of each pixel.

Next, the reliability evaluation process in step S45 and the distance number determination process in step S46 are demonstrated.

As described above, in the distance number, a limit value is set according to the capability rank of the air conditioner and the elevation angle of each pixel. Even when the distance number estimation result is N> maximum value D, all the results are distances in the plurality of measurement results. If number = N, distance number = D is set.

The distance numbers of eight times are determined in each pixel, and the distance numbers are determined by taking the average of the remaining four distance numbers except for the two distance numbers in order from the largest and the two distance numbers in the order from the smallest. In the case of using the stereo method according to the block matching method, when detecting an obstacle with no luminance change, the parallax calculation is not stable, and a largely different parallax result (distance number) is detected every measurement. Therefore, in step S45, when the values of the remaining four distance numbers are compared, and the gap is equal to or more than the minimum physical quantity that causes a reaction, the value of the distance number is assumed to be unreliable in step S47, and the distance estimation in the pixel is performed. It gives up and uses the distance number estimated in the neighborhood pixel. In addition, the average value is made into the integer value quantized by discarding below a decimal point, and the position corresponded to the distance number determined in this way is as Table 4 or Table 5.

In addition, in the present embodiment, eight distance numbers are determined for each pixel, and the distance numbers are determined by taking the average of the remaining four distance numbers except for two distance numbers, respectively, to determine the distance numbers. The number is not limited to eight, and the distance numbers taking the average value are not limited to four.

That is, when determining the presence or absence of an obstacle in each obstacle location determination area, obstacle detection means in each obstacle location determination area according to the result of the presence or absence determination result of the person in the person location determination area corresponding to each obstacle location determination area. By determining whether or not to update the determination result, the presence or absence of the obstacle is effectively determined. More specifically, in the obstacle position determination region belonging to the human position determination region judged that there is no person by the human body detection means, the previous determination result by the obstacle detection means is updated with a new determination result, while the human body detection means In the obstacle position determination region belonging to the person position determination region determined to be a person, the previous determination result by the obstacle detection means is not updated with the new determination result.

In addition, although the previous distance data was used in step S43 in the flowchart of FIG. 25, since the previous data does not exist immediately after the installation of the air conditioner, in each obstacle position determination region by the obstacle detecting means. In the case of the first determination of, the default value is used, and the above-mentioned limit value (maximum value D) is used as the default value.

FIG. 27 is an elevation view (a longitudinal cross-sectional view through an imaging sensor unit) of a living space, in which the bottom 2m of the imaging sensor unit has a bottom surface, and a measurement result when there is an obstacle such as a table from 0.7 to 1.1m from the bottom surface. In the drawing, the meshed hatched portion, the right upward oblique line portion, and the right downward oblique line portion are judged to have obstacles at near, middle, and far distances (to be described later).

<Learning control of obstacle detection>

As described above, the stereo method increases the likelihood that the obstacle detection will fail depending on the subject, such as when detecting an obstacle with no change in luminance.

As an example, when a table such as a dining table having a flat top surface without any change in luminance is assumed, when there is nothing on the table, the parallax calculation fails by the stereo method, so that positioning of the table is difficult. However, when household goods (tableware, remote control, books, newspapers, tissue boxes, etc.) exist on the table, luminance differences (textures) are generated on the surface, so that positioning of the table by the stereo method is easy.

Therefore, in this learning control, the obstacle detection is performed not only by the obstacle but also by the interaction with the surrounding accessories in the vicinity of the obstacle. However, the furniture, etc. (actually, household goods placed on the furniture rather than the furniture) actually placed in the room is likely to change in place every day, and the change of the angle of the obstacle or the interaction of the surrounding accessories near the obstacle is changed. By repeatedly performing the obstacle detection, the detection miss can be reduced to the maximum. In this learning control, as shown in the flowchart shown in Fig. 28, the obstacle position is learned based on each scan result, the location of the obstacle is determined from the learning control result, and airflow control described later is performed.

FIG. 28 shows a flowchart showing the presence or absence of an obstacle determination, and the obstacle existence determination is sequentially performed for all positions (obstacle position determination regions) shown in FIG. 23. Here, position A1 will be described as an example.

When the obstacle detection operation is started by the imaging sensor units 24 and 26, first in step S71, the detection sensor (stereo method) is performed by the imaging sensor units 24 and 26 in the first pixel of the position A1, and step S72. In the above, the presence or absence of the above-described obstacle is determined. If it is determined in step S72 that there is an obstacle, in step S73, "1" is added to the first memory, and if it is determined that there is no obstacle, in step S74, "0" is added to the first memory.

In step S75, it is determined whether the detection in all the pixels of the position A1 is finished. When the detection in all the pixels is not finished, in step S76, the detection operation is performed in the next pixel by the stereo method, and in step S76, Return to S72.

On the other hand, when the detection in all the pixels is complete | finished, in step S77, the numerical value (sum total of the pixel determined to have an obstacle) recorded in the 1st memory is divided by the number of pixels of position A1 (dividing is performed), and next. In step S78, the quotient is compared with a predetermined threshold value. If the quotient is larger than the threshold value, it is temporarily determined that there is an obstacle in the position A1 in step S79, and "5" is added to the second memory in step S80. On the other hand, if the quotient is less than the threshold value, it is temporarily determined that there is no obstacle in the position A1 in step S81, and "-1" is added to the second memory in step S82 ("1" is subtracted).

In addition, the obstacle detection by the imaging sensor units 24 and 26 is so difficult that the distance from the imaging sensor units 24 and 26 to an obstacle becomes difficult, and the threshold value used here is based on the distance from an indoor unit, For example, it is set as follows.

Nearest: 0.4

Medium range: 0.3

Long range: 0.2

In addition, since this obstacle detection operation is performed every time the air conditioner is operated, "5" or "-1" is repeatedly added to the 2nd memory. Therefore, the numerical value recorded in the second memory has set the maximum value to "10" and the minimum value to "0".

Next, in step S83, it is determined whether or not the numerical value (total after addition) recorded in the second memory is equal to or greater than the determination reference value (e.g., 5). On the other hand, if it is less than the determination reference value, it is finally determined that there is no obstacle in the position A1 in step S85.

In addition, the first memory can be used as a memory for the obstacle detection operation at the next position by clearing the memory when the obstacle detection operation at one position is completed. However, the second memory has one memory for each operation of the air conditioner. By accumulating the addition values at the positions (where the maximum value> the total value> the minimum value), the same number of memories as the number of positions are prepared.

In the above-described obstacle detection learning control, "5" is set as the determination reference value, and when it is finally determined that there is an obstacle in the first obstacle detection in any position, "5" is recorded in the second memory. In this state, if it is finally determined that there is no obstacle in the next obstacle detection, the value obtained by adding "-1" to "5" becomes less than the determination reference value, so that no obstacle exists in the position.

However, if it is finally determined that there is an obstacle even in the next obstacle detection, the value "10" which adds "5" to "5" is recorded in the 2nd memory, and since the sum total is more than the determination reference value, there exists an obstacle in the position. Even if it is determined that there are no obstacles in five obstacle detections after the next time, since the value of adding "-1" to "10" is "5", the obstacle still exists in the position.

That is, the learning control of this obstacle detection determines that there is no obstacle when it is determined that there is an obstacle when the final presence or absence of the obstacle is determined on the basis of several addition total values (or addition subtraction total values). It is characterized by the fact that it is set to a sufficiently large number than the value to be subtracted.

In addition, by setting the maximum value and the minimum value to the numerical values recorded in the second memory, even if the position of the obstacle changes greatly due to moving or relocation, the change can be followed as soon as possible. If it is determined that there is an obstacle every time when the maximum value is not provided, the product is gradually increased, and even if the position of the obstacle is changed by moving or the like, and there is no obstacle in the area determined each time that there is an obstacle, the time is shorter than the determination standard value. I get caught. If the minimum value is not provided, the opposite phenomenon occurs.

FIG. 29 shows a modification of the learning control for obstacle detection shown in the flowchart of FIG. 28, and since only the processing in steps S100, S102, S103 differs from the flowchart in FIG. 28, these steps will be described.

In this learning control, if it is temporarily determined that position A1 has an obstacle in step S99, "1" is added to 2nd memory in step S100. On the other hand, if it is temporarily determined that there is no obstacle in the position A1 in step S101, "0" is added to the second memory in step S102.

Next, in step S103, if the total value recorded in the second memory based on the past 10 obstacle detections including the current obstacle detection is greater than or equal to the determination reference value (for example, 2), the position A1 is entered in step S104. If it is finally determined that there is an obstacle, and it is less than the determination reference value, it is finally determined that the position A1 is not an obstacle in step S105.

In other words, the learning control of the obstacle detection described above is finally determined that there is an obstacle if it can be detected twice, even if the obstacle cannot be detected eight times in the past ten obstacle detections in any position. Therefore, this learning control is characterized by setting the obstacle detection number (here, 2) which finally determines that there is an obstacle to a number sufficiently smaller than the number of obstacle detections in the past to refer to. This makes it easy to get results.

In addition, a button for resetting the data recorded in the second memory may be provided in the indoor unit main body or the remote controller, and the data may be reset by pressing this button.

Basically, the position of the obstacle and the wall surface which have a big influence on the airflow control rarely change, but when a change of the installation position of the indoor unit accompanied by moving or the change of the furniture position by repositioning in a room occurs now It is not preferable to perform airflow control based on the data obtained until now. This is because, by the learning control, it eventually becomes a control suitable for the room, but it takes time to achieve the optimum control (especially when the obstacle is eliminated in the area). Therefore, when the reset button is provided and the relative positional relationship between the indoor unit and the obstacle or the wall surface is changed, by resetting the data so far, it is possible to prevent improper air adjustment based on past wrong data and to control the learning. By restarting from the beginning, it is possible to make the control suitable for the situation more quickly.

Obstacle Avoidance Control

Based on the presence or absence of the obstacle, the upper and lower blades 12 and the left and right blades 14 as the wind direction changing means are controlled as follows during heating.

In the following description, the terms "block", "field", "near distance", "medium distance", and "far distance" are used, and these terms will be described first.

Regions A to G shown in FIG. 13 belong to the next block, respectively.

Block N: Area A

Block R: Areas B and E

Block C: Area C, F

Block L: Area D, G

In addition, the areas A to G belong to the following fields, respectively.

Field 1: Zone A

Field 2: Zone B, D

Field 3: Zone C

Field 4: Zones E, G

Field 5: Zone F

In addition, the distance from an indoor unit is defined as follows.

Near: Zone A

Medium range: zones B, C, D

Remote: Zone E, F, G

Table 7 shows the target setting angles in the positions of the five left wings and the five right wings constituting the left and right wings 14, and the symbols attached to the numbers (angles) are shown in FIG. Similarly, the case where the left wing or the right wing faces inward is defined as the positive (+, no symbol in Table 7) direction and the outward direction as the negative (-) direction.

In addition, the "heating B area" in Table 7 is a heating area which performs obstacle avoidance control, and "normally automatic wind direction control" is wind direction control which does not perform obstacle avoidance control. Here, the determination of whether or not to perform obstacle avoidance control is based on the temperature of the indoor heat exchanger 6, and when the temperature is low, the wind direction control that does not blow wind to the occupants; In the case of control and moderate temperature, the wind direction control to the heating B area | region is performed. In addition, the term "low temperature", "too high", "wind direction control which does not blow a resident", and "wind direction control of the maximum air volume position" here mean as follows.

? Low temperature: The temperature of the indoor heat exchanger 6 sets the skin temperature (33-34 ° C.) as the optimum temperature, and the temperature that can be below this temperature (eg 32 ° C.).

? Too high temperature: for example 56 ° C. or higher

? Wind direction control that does not hit the wind inhabitants: Wind direction control along the ceiling by controlling the angle of the upper and lower wings 12 without sending wind to the living space

? Wind direction control of the maximum air volume position: The air conditioner is infinitely close to zero in that the maximum air volume position is infinitely close to zero in that resistance (loss) always occurs when the airflow is bent by the upper and lower blades 12 and the left and right blades 14. Lose wind direction control (in the left and right wing 14, the position facing directly front, in the case of the upper and lower wing 12, 35 degrees from the horizontal down)

Table 8 has shown the target setting angle in each field of the up-and-down blade | wing 12 in the case of obstacle avoidance control. In addition, the angle (gamma) 1 of the upper blade | wing and the angle (gamma) 2 of the lower blade | wing in Table 8 are the angle (angular angle) measured from the vertical line to the upward direction.

Next, although the obstacle avoidance control according to the position of an obstacle is demonstrated concretely, the terms "swing operation", "position rectification operation", and "block rectification operation" used in obstacle avoidance control are demonstrated first.

The swing motion is an oscillation motion of the left and right blades 14, basically a motion of swinging at a predetermined left and right angular width with respect to one position of the target, and having no fixed time at both ends of the swing.

In addition, position rectification operation | movement correct | amends Table 9 with respect to the target setting angle (angle of Table 7) of a certain position, and sets it as a left end and a right end, respectively. As the operation, the wind direction fixation time (time to fix the left and right wings 14) at the left end and the right end, respectively, for example, when the wind direction fixation time has elapsed at the left end, it moves to the right end and the right end until the wind direction fixation time has elapsed at the right end The wind direction is maintained, and the wind direction is fixed to the left end after the fixed time elapses, and repeats this. The wind direction holding time is set to, for example, 60 seconds.

That is, if there is an obstacle at a position, if the target setting angle of the position is used as it is, the warm air always corresponds to the obstacle, but by correcting Table 9, the warm air is reached from the side of the obstacle to the position where a person is located. I can do it.

In addition, block rectification operation | movement determines the setting angle of the left and right blade | wing 14 corresponding to the left end and the right end of each block based on Table 10, for example. As an operation, each of the blocks has a wind direction fixation time at the left end and the right end of each block, for example, when the wind direction fixation time has elapsed at the left end, it moves to the right end, and maintains the wind direction of the right end until the wind direction fixation time has elapsed at the right end, After the fixed time has elapsed, it moves to the left end and repeats this. The wind direction settling time is set to, for example, 60 seconds as in the position rectification operation. In addition, since the left end and the right end of each block coincide with the left end and the right end of the person position determination area belonging to the block, the block rectification operation may be referred to as the rectification operation of the person position determination area.

In addition, position rectification operation | movement and block rectification operation are used according to the magnitude | size of an obstacle. In the case where the obstacle in front is small, the position rectification operation is performed around the position where the obstacle is located, while the blowing is avoided by the obstacle, while the obstacle in front is large, for example, when there is an obstacle in front of the human area, the block rectification operation is performed. By blowing, the air is blown over a wide range.

In this embodiment, the swing motion, the position commutation motion and the block commutation motion are collectively referred to as the swing motion of the left and right blades 14.

Hereinafter, although the control example of the up-down wing 12 or the left-right wing 14 is demonstrated concretely, when a human body detection means determines that a person is only in a single area | region, the human position determined by a human body detection means that a person exists. When it is determined by the obstacle detecting means that there is an obstacle in the obstacle position determining area located in front of the discriminating area, the upper and lower blades 12 are controlled to perform airflow control to avoid the obstacle from above. Further, when it is determined by the obstacle detecting means that there is an obstacle in the obstacle location determining area belonging to the person location determining area judged as human by the human body detecting means, at least one obstacle belonging to the person location determining area determined to be human 1st air flow control which oscillates the left and right wing 14 in the position determination area | region, and does not provide the fixed time of the left and right wing 14 at the both ends of a swing range, and is adjacent to the person position determination area | region or the person judged to be a person. Oscillate the left and right blades 14 in at least one obstacle positioning region belonging to a person positioning region, and select one of the second airflow controls provided with a fixed time of the left and right blades 14 at both ends of the swing range. have.

In addition, in the following description, although the control of the upper and lower blades 12 and the control of the left and right blades 14 are distinguished, according to the position of a person and an obstacle, the control of the upper and lower blades 12 and the control of the left and right blades 14 are different. Is suitably combined.

A. Up and down wing control

(1) When there is a person in any of the areas B to G, and there is an obstacle in positions A1 to A3 in front of the area where the person is located.

The set angle of the upper and lower blades 12 is corrected as in Table 11 with respect to the normal field wind direction control (Table 8), and airflow control is performed in which the upper and lower blades 12 are set upward.

(2) When there is a person in any of the areas B to G and there is no obstacle in the area A in front of the area where the person is present (other than (1) above)

Usually automatic wind direction control is performed.

B. Left and Right Wing Control

B1. If a person is in area A (near field)

(1) When there is only one position without obstacles in area A

The first airflow control is performed by swinging left and right about the target setting angle of the position where there is no obstacle. For example, if there are obstacles in positions A1 and A3 and there are no obstacles in position A2, the swing operation is performed to the left and right around the target setting angle of position A2, and basically the air is adjusted to position A2 without obstacles. Since A3 cannot be said to have no people, by adding a swing motion, airflow is distributed to positions A1 and A3 to some extent.

More specifically, based on Tables 7 and 9, since the target setting angle and correction angle (swing angle at the time of swing operation) of position A2 are determined, both the left wing and the right wing are centered around 10 degrees, respectively. Continue to swing (swing) without stopping within an angle range of ± 10 degrees. However, the timing of shaking the left wing and the right wing to the left and right is set to be the same, and the rocking motions of the left wing and the right wing are interlocked.

(2) When there are two positions without obstacles in the area A and adjacent to each other (A1 and A2 or A2 and A3)

The first air flow control is performed by swinging the target setting angles of the two positions without obstacles at both ends, thereby basically adjusting the positions without obstacles.

(3) Two positions without obstacles in the area A and separated (A1 and A3)

The second airflow control is performed by block rectifying operation with the target setting angles of the two positions without obstacles at both ends.

(4) If there are obstacles in all positions in area A

Since it is unclear where to aim, the block N is rectified and block 2nd airflow control is performed. This is because the block rectifying operation becomes a directional wind direction rather than aiming at the entire area, so that it is easy to reach far and to avoid obstacles. That is, even when an obstacle is interspersed in the area A, there is usually a gap between the obstacle and the obstacle, and the air can be blown through the gap between the obstacles.

(5) all positions in area A have no obstacles

Normal automatic wind direction control of the area A is performed.

B2. If there is a person in any of areas B, C, and D (medium distance)

(1) There is an obstacle in only one of the two positions belonging to the area where the person is located.

The first airflow control is performed by swinging left and right about the target setting angle of the position where there is no obstacle. For example, when there is a person in the area D and there is an obstacle only in the position D2, the swing operation is performed left and right about the target set angle of the position D1.

(2) Obstacles in both positions belonging to the human area

The block including the human area is subjected to block rectification to perform the second air flow control. For example, when there is a person in the area D and there are obstacles in both the positions D1 and D2, the block L is started to commutate block.

(3) When there is no obstacle in the area where people are

Normal automatic wind direction control is performed in an area where a person is present.

B3. If there is a person in any of areas E, F, and G (far)

(1) If there is an obstacle in only one of the two positions belonging to the medium-distance zone in front of the person's area (for example, there is a person in area E, there is an obstacle in position B2, and there is no obstacle in position B1)

(1.1) If there are no obstacles on both sides of the position with obstacles (e.g. no obstacles in positions B1, C1)

(1. 1.1) If there are no obstacles behind the obstructed position (e.g. no obstacles in position E2)

The position rectification operation | movement is made centering on the position with an obstacle, and 2nd air flow control is performed. For example, if there is a person in the area E, there are obstacles at the position B2, and there are no obstacles at both sides and the rear, the airflow can be sent to the area E by avoiding the obstacle at the position B2 from the side.

(1. 1.2) If there is an obstacle behind the obstructed position (e.g. there is an obstacle at position E2)

The first airflow control is performed by swinging around the target setting angle of the position where the obstacle is free in the intermediate distance region. For example, if there is a person in the area E, there is an obstacle in the position B2, and there are no obstacles on both sides, but there is an obstacle behind it, it is advantageous to send airflow from the position B1 without the obstacle.

(1. 2) If there is an obstacle on either side of the neighboring position and there is no obstacle on one side

The first airflow control is performed by swinging around the target setting angle at the position where there is no obstacle. For example, if there is a person in the area F, an obstacle in the position C2, an obstacle in the position D1 of both neighbors of the position C2, and there is no obstacle in the C1, the air flow avoids the obstacle of the position C2 from the position C1 without the obstacle. Send to area F.

(2) Obstacles in both positions in the middle distance area in front of the human area.

The block including the human area is subjected to block rectification to perform the second air flow control. For example, when there is a person in the area F and there are obstacles in both the positions C1 and C2, the block C is operated for block rectification. In this case, since there is an obstacle in front of the person and there is no way to avoid the obstacle, the block rectifying operation is performed regardless of whether there is an obstacle in the block adjacent to the block C.

(3) If there are no obstacles in both positions in the middle distance area in front of the human area (for example, there is a person in area F and there are no obstacles in positions C1 and C2).

(3.1) If there is an obstacle in only one of the two positions that belong to the human area

The first airflow control is performed by swinging around the target setting angle at one position without obstacles. For example, if there is a person in the area F, there are no obstacles in the positions C1, C2, F1, and there is an obstacle in the position F2, the front of the area F where the person is open is open, so the long distance without the obstacle in consideration of the long distance obstacle. Adjust the air around position F1.

(3. 2) Obstacles at both positions belonging to the human area

The block including the human area is subjected to block rectification to perform the second air flow control. For example, if there is a person in the area G, there are no obstacles in the positions D1 and D2, and there are obstacles in both the positions G1 and G2, the front of the area G in which the person is open is open, but there is an obstacle in the whole area, It is unclear where to aim, so block L is commutated.

(3.3) If there are no obstacles in both positions that belong to the human area

Normal automatic wind direction control is performed in an area where a person is present.

In addition, although this obstacle avoidance control was made to control the upper and lower wings 12 and the left and right wings 14 based on the presence or absence of a person by the human body detection means and the presence or absence of the obstacle by the obstacle detection means, the obstacle detection is performed. The upper and lower blades 12 and the left and right blades 14 may be controlled based only on the determination of the presence or absence of obstacles by means.

Obstacle avoidance control based only on determination of obstacles

This obstacle avoidance control is basically for avoiding the area determined to be an obstacle by the obstacle detecting means and blowing the air toward the area determined to be no obstacle, and a concrete example thereof will be described below.

A. Up and down wing control

(1) When there is an obstacle in area A (near field)

When heating the upper and lower wings 12 to the lowest direction in order to suppress the rising warmth during heating, if there is an obstacle in the area A, warming up behind the obstacle (indoor side), or warming obstacles It may be considered that it does not reach the bottom surface by touching.

Therefore, when an obstacle is detected directly under or near the indoor unit, the set angle of the upper and lower blades 12 is corrected as in Table 11 with respect to the normal field control (Table 8), and the upper and lower blades 12 are set upward. Airflow control is performed and air is adjusted from the obstacle. If the entire air flow is lifted up too high to avoid obstacles, the warming will directly touch the occupant's face, causing discomfort, so that rising to the upper wing 12a while lifting the warm up with the lower wing 12b to avoid obstacles. To prevent it.

B. Left and Right Wing Control

(1) When there is an obstacle in any one of areas B, C, and D (medium distance)

Air conditioning is focused on the obstacle-free direction. For example, in the case where an obstacle is detected in the area C (room center), the blocks including the areas B and D on both sides without obstacles are alternately subjected to block rectification, so that an area free of obstacles (= likely to be present) can be obtained. Can focus air control.

In addition, when an obstacle is detected in the area | region B or D (a corner of a room), the block commutation operation | movement of the block containing area | region C and D or area | region B and C is made to operate. In this case, if the block C is rectified and operated in the area C and D or the areas B and C at a ratio of a plurality of times (for example, five times), the left and right blades 14 are swinged toward the area B or D. Not only can the air be controlled around an area where people are more likely to be present, it is also effective in terms of air conditioning throughout the room.

The position for determining the presence or absence of an obstacle (obstacle position determining region) may be subdivided as shown in FIG. 23 irrespective of the capability rank of the air conditioner. However, since the size of the room to be installed varies according to the capability rank, the divided region. You may change the number. For example, when the capacity rank is 4.0 kw or more, it may be divided as shown in FIG. 23, and when the capacity rank is 3.6 kw or less, the short distance may be divided into three and the intermediate distance area may be divided into six without providing a far distance.

In addition, as shown in FIG. 23, when a room is recognized in a radial shape and divided into near / middle / distant distances equidistantly from the indoor unit, the area becomes larger as it moves away from the indoor unit. Therefore, as the distance from the indoor unit increases, the number of discrimination zones increases, so that the size of each zone can be made almost uniform, and airflow control is easily performed.

<Person wall proximity control>

When a person and the wall exist in the same area, the person must be located in front of the wall and close to the wall, so that the warm air tends to stay near the wall during heating, and the room temperature near the wall tends to be higher than the room temperature of other parts. In that respect, the human wall proximity control is performed.

In this control, the parallax is calculated from a pixel different from the pixels [i, j] shown in Table 4, the distance is detected, and the positions of the front wall and the left and right walls are recognized first.

That is, using the imaging sensor units 24 and 26, the parallax of the pixel corresponding to the front of a substantially horizontal direction is calculated first, the distance to the front wall is measured, and a distance number is calculated | required. Further, the parallax of the pixel corresponding to the left side in the substantially horizontal direction is calculated, the distance to the left wall is measured, the distance number is obtained, and the distance number of the right wall is thus obtained as well.

In addition, it demonstrates with reference to FIG. Fig. 29 is a view of the room in which the indoor unit is mounted, and shows a case in which the front wall WC, the left wall WL, and the right wall WR exist in front, left, and right sides as viewed from the indoor unit, respectively. In addition, the number on the left side of FIG. 29 has shown the distance number of the corresponding grating | lattice, and Table 12 has shown the distance from the indoor unit to the near point and the origin point corresponding to the distance number.

As mentioned above, the "obstacle" used in this specification assumes furniture, such as a table and a sofa, TV, audio, etc., and it is not detected in the angle range of 75 degrees of elevation, considering the normal height of such an obstacle. In this embodiment, the distance to the front, left and right ends of the indoor unit is detected at an elevation angle of 75 degrees or more, and it is assumed that there is a wall on the extension including the position.

In the horizontal viewing angle, the left wall WL is at a position of -80 degrees and -75 degrees, the front wall WC is at a position of -15 degrees to 15 degrees, and the right wall WR is an angle of 75 degrees and 80 degrees. Since it can be estimated that each exists in the position of FIG. 3, among the pixels shown in Table 3, the pixel corresponding to the said viewing angle of the said horizontal direction within 75 degrees of elevation is as follows.

Left: [14, 15], [18, 15], [14, 21], [18, 21], [14, 27], [18, 27]

Front: [66, 15]-[90, 15], [66, 21]-[90, 21], [66, 27]-[90, 27]

Velvet: [138, 15], [142, 15], [138, 21], [142, 21], [138, 27], [142, 27]

When determining the distance number from the indoor unit to the front wall WC, the left wall WL, and the right wall WR, wall surface data is first extracted from each pixel as shown in Table 13.

Next, as shown in Table 14, the front wall WC, the left wall WL, and the right wall WR are removed based on the wall data obtained by removing the upper and lower limit values of each wall surface data to eliminate unnecessary wall surface data. Determine the distance number to).

As the distance numbers to the front wall WC, the left wall WL, and the right wall WR, the maximum values WC: 5, WL: 6, and WR: 3 in Table 14 can be adopted. In the case of adopting the maximum value, air is controlled in a room (large room) in which the distance from the indoor unit to the front wall WC, the left wall WL, and the right wall WR is far, so that a larger space can be set as an object of air conditioning control. Can be. However, it is not necessarily the maximum value and an average value may be adopted.

In this way, after determining the distance numbers to the front wall WC, the left wall WL, and the right wall WR, the wall is placed in the obstacle position determination region belonging to the human position determination region determined by the human body detecting means. If it is determined by the obstacle detecting means, and it is determined that there is a wall, it is considered that there is a person in front of the wall. Therefore, during heating, temperature setting slightly lower than the set temperature set by the remote controller is performed.

Hereinafter, this human wall proximity control is demonstrated concretely.

A. The person is in the near or medium distance area

Since the near area and the medium range area are located near the indoor unit and the area area is also small, the degree of room temperature rises is increased, so that the set temperature set by the remote controller is set slightly lower by the first predetermined temperature (for example, 2 ° C). .

B. The person is in the far zone

Since the remote area is located far from the indoor unit and the area area is also large, the room temperature rises is lower than the near area or the medium distance area. For example, it is set as low as 1 ℃).

In addition, since the remote area has a large area area, even if a person and a wall are detected in the same person location determination area, the person and the wall may be separated from each other. The temperature shift is performed in accordance with the positional relationship between the person and the wall.

In addition, although the stereo method as a distance detection means was employ | adopted in this embodiment, the method using the light projection part 28 and the imaging sensor unit 24 can also be employ | adopted instead of the stereo method. This method will be described.

As shown in FIG. 30, the main body 2 of this embodiment has the imaging sensor unit 24 and the light projection part 28. As shown in FIG. The light projecting unit 28 includes a light source and a scanning unit (not shown), and the light source may be an LED or a laser. In addition, the scanning unit can arbitrarily change the projection direction using a galvano mirror or the like. 31 is a schematic diagram illustrating the relationship between the imaging sensor unit 24 and the light projecting portion 28. Originally, the projection direction is two degrees of freedom and the imaging plane is a two-dimensional plane in the vertical and horizontal directions. Here, the light projecting portion 28 projects light in the light projection direction ρ with respect to the optical axis direction of the imaging sensor unit 24. The imaging sensor unit 24 performs the difference processing between the frame image immediately before the light projecting portion 28 transmits and the frame image being projected, thereby causing the image sensor unit 24 to reflect the light projected by the light projecting portion 28 on the image of the point P. Get u coordinate u1. If the distance from the imaging sensor unit 24 to the point P is X, the following relationship is established.

therefore,

That is, by detecting the reflection point P of the light while changing the light emission direction p of the light projection part 28, distance information in the air control space can be obtained.

I and j in Table 16 indicate addresses to be scanned by the light-transmitting part 28, and the angle in the vertical direction and the angle in the horizontal direction are measured from the elevation angle α and the indoor unit described above from the front reference line to the right direction. Angle (beta) is shown, respectively. That is, from the indoor unit, each address is set in the range of 5 degrees to 80 degrees in the vertical direction and -80 degrees to 80 degrees in the horizontal direction, and the light projecting unit 28 measures each address to scan the living space.

Next, the distance measurement to an obstacle is demonstrated, referring the flowchart of FIG. In addition, since the flowchart of FIG. 32 is very similar to the flowchart of FIG. 25, only another step is demonstrated below.

First, in step S48, when it is determined that there is no person in the area (any one of the areas A to G shown in FIG. 13) corresponding to the address [i, j] for transmitting light, the process proceeds to step S49. On the other hand, if it is determined that there is a person, the flow proceeds to step S43. That is, since the person is not an obstacle, the pixel corresponding to the area where the person is judged to correspond to the area where no person is determined using the previous distance data (not updating the distance data) without performing the distance measurement. Distance measurement is performed only for the pixel to be used, and the distance data newly measured is used (to update the distance data).

In step S49, the distance to an obstacle is estimated by acquiring the above-mentioned lightcasting process and a reflection point from the imaging sensor unit 24. FIG. Of course, as mentioned above, the distance number determination process may be used to perform the process using the distance number.

Moreover, you may use a human body detection means as distance detection means. This consists of a human body distance detection means using a human body detection means, and an obstacle detection means using a human body detection means. This processing will be described.

As shown in FIG. 33, the main body 2 of this embodiment has a single imaging sensor unit 24. As shown in FIG. 34 is a flowchart which shows the flow of a process of the human body distance detection means using a human body detection means. In this figure, the same steps as in Fig. 5 are denoted by the same reference numerals, and detailed description thereof is omitted here.

In step S201, the human body distance detecting means detects the pixel existing at the top of the image among the pixels with the difference in each region where the human body detecting means divides the region, and acquires the v-coordinate as v1.

Further, in step S202, the human body distance detecting means estimates the distance from the imaging sensor unit to the person using v1, which is the v-coordinate at the top of the image. 35 is a schematic diagram for explaining this process. FIG. 35A is a schematic diagram of a scene in which two persons 121 and 122 exist near and far from the camera, and FIG. 35B is a differential image of an image captured by the imaging sensor unit in the scene of FIG. Indicates. In addition, the areas 123 and 124 in which the difference occurs correspond to the people 121 and 122, respectively. Here, the height h1 of the person is already known, and the heights of all the people in the air conditioning space are roughly the same. As described above, since the imaging sensor unit 24 is provided at a height of 2 m, as shown in Fig. 35A, the imaging sensor unit performs imaging from the top of the person. At this time, the closer the person is to the imaging sensor unit, the more the person is picked up on the lower part of the image, as shown in Fig. 35B. In other words, the v coordinate v1 at the top of the person's image and the distance from the imaging sensor unit to the person correspond to one to one. From this, the correspondence between the v coordinate v1 at the top of the person and the distance from the imaging sensor unit to the person can be obtained in advance, so that the human body distance detecting means using the human body detecting means can be performed. Table 17 shows an example in which the average height of a person is used as h1, and the correspondence between the v coordinate v1 of the uppermost image of the person and the distance from the imaging sensor unit to the person is obtained in advance. Here, an imaging sensor unit having a resolution of VGA was used as the imaging sensor unit. From this table, for example, when v1 = 70, the distance from the imaging sensor unit 24 to the person is estimated to be approximately 2 m.

Next, the obstacle detecting means using the human body detecting means will be described.

36 is a flowchart showing the flow of processing of the obstacle detecting means using the human body detecting means.

In step S203, the obstacle detecting means estimates the height v2 of the person on the image, using the distance information from the imaging sensor unit 24 estimated by the human body distance detecting unit to the person. FIG. 37 is a schematic diagram for explaining this process, and is a schematic diagram showing the scene as shown in FIG. Here, as described above, the height h1 of the person is already known, and the heights of all the people in the air conditioning space are roughly the same. As described above, since the imaging sensor unit 24 is provided at a height of 2 m, as shown in Fig. 34A, the imaging sensor unit performs imaging from the top of the person. At this time, the closer the person is to the imaging sensor unit 24, the larger the size on the image of the person as shown in Fig. 34B. That is, the difference v2 between the v coordinate of the uppermost image of the person and the v coordinate of the lowermost image corresponds to one-to-one with respect to the distance from the imaging sensor unit 24 to the person. In this regard, when the distance from the imaging sensor unit to the person is known, the size on the image can be estimated. This may be obtained in advance by a correspondence between the difference v2 between the v coordinate of the uppermost image of the person and the v coordinate of the lowermost image of the person and the distance from the imaging sensor unit to the person. Table 18 shows the correspondence between the v coordinate v1 of the uppermost image of the person, the difference v2 between the v coordinate of the uppermost image of the person and the v coordinate of the lowermost image, and the distance from the imaging sensor unit 24 to the person. This is an example of obtaining a dictionary. Here, an imaging sensor unit having a resolution of VGA was used as the imaging sensor unit. From this table, for example, when the distance from the imaging sensor unit 24 to the person is 2 m, it is estimated that the difference v2 = 85 between the v coordinate of the uppermost image of the person and the v coordinate of the lowest image.

In step S204, the obstacle detecting means detects the pixel in which the difference that exists in the uppermost image and the pixel in which the difference that exists in the lowermost image occur in each region of the difference image, and the difference v3 of this v-coordinate. Calculate

In step S205, the imaging sensor unit is compared by comparing the height v2 of the person on the image estimated using the distance information from the imaging sensor unit 24 to the person and the height v3 of the person obtained from the actual difference image. At 24, it is estimated whether an obstacle exists between the characters. 38 and 39 are schematic views for explaining this process. FIG. 38 shows a scene as shown in FIG. 35, and is a schematic diagram showing a scene in which no obstacle exists between the imaging sensor unit 24 and the person. 39 is a schematic diagram showing a scene in which an obstacle exists. In FIG. 38, when there is no obstacle between the imaging sensor unit and the person, from the height v2 of the person on the image estimated using the distance information from the imaging sensor unit 24 to the person and the actual difference image. The height v3 of the person 123 obtained is approximately equal. On the other hand, in FIG. 39, when an obstacle exists between the imaging sensor unit 24 and the person, part of the person is shielded, so that there is no difference in the shielded area. Here, it is considered that most of the shield in the air conditioning space is placed on the floor, so that the lower area of the person is shielded. This indicates that when the distance to the person is obtained using v1, which is the v-coordinate of the uppermost image of the person area, even if an obstacle exists between the image sensor unit 24 and the person, the distance is accurately determined. On the other hand, if an obstacle exists between the imaging sensor unit 24 and the person, the height v3 of the person 125 obtained from the actual difference image indicates the distance information between the imaging sensor unit 24 and the person. It is estimated that it becomes small compared with the height v2 of the person in the image estimated using. For this reason, when it determines with v3 being small enough compared with v2 in step S205, it transfers to step S206 and determines that there is an obstacle between an imaging sensor unit and a person. At this time, it is assumed that the distance between the imaging sensor unit and the obstacle is the same as the distance from the imaging sensor unit to the person obtained from the uppermost v coordinate v1.

As described above, the distance detection means is realized using the detection result by the human body detection means.

The present embodiment has been described as having the image sensor unit 24 fixed with sufficient viewing angle. However, when the horizontal field of view of the image sensor unit 24 is narrow, it may be operated to reciprocate horizontally to widen the field of view. . Similarly, when the vertical field of view of the imaging sensor unit 24 is narrow, it may be operated to reciprocate vertically to widen the field of view. When the horizontal and vertical visual fields are both narrow, the horizontal and vertical visual fields can be widened by scanning and operating the imaging sensor unit 24.

Also in this case, each image processing may use the whole image obtained by the drive of an imaging device, and only the number of pixels differs, and the idea is the same as that of the fixed imaging device.

In this operation, it is possible to avoid discomfort to the occupants by operating the imaging sensor unit 24 as in the present invention.

In addition, in this embodiment, although the imaging device 25 arrange | positioned under the indoor unit was covered with a part of indoor unit, it goes without saying that even if it is arrange | positioned above the indoor unit, it is the same. In addition, even when the imaging device 25 is not covered by a part of the indoor unit, and the imaging sensor unit is exposed even when the operation is stopped or protected by a transparent cover, the driving method of the imaging sensor unit to be driven can perform the operation described in the present invention. Needless to say, by doing so, it may not cause discomfort to the residents.

Moreover, of course, the structure which arrange | positions one imaging device 25 near the center of an indoor unit may be sufficient.

(Industrial availability)

The air conditioner according to the present invention is particularly useful as a general home air conditioner because it can suppress the deterioration of the recognition performance of the imaging sensor and can provide a sense of security to the occupants.

2: indoor unit main body 2a: front opening
2b: surface opening 4: movable front panel
6: heat exchanger 8: indoor fan
10: air vent 12: up and down wings
14: left and right wing 16: filter
18, 20: arms for front panel 24, 26: imaging sensor unit
28: light projecting portion 25: imaging device
51: circuit board 52: lens
53: imaging sensor 54: support (sensor holder)
55: rotating shaft for horizontal (horizontal) rotation 56: rotating shaft for vertical (vertical) rotation
57: motor for horizontal rotation 58: motor for vertical rotation

Claims (6)

The indoor unit is provided with a human body detecting means for detecting the presence of a person and an obstacle detecting means for detecting the presence or absence of an obstacle, and is provided in the indoor unit based on a detection result of the human body detecting means and a detection result of the obstacle detecting means. As an air conditioner to control the direction change wing,
The human body detecting means and the obstacle detecting means are realized by a fixed imaging device or a driven imaging device,
At the time of stopping operation of the air conditioner, the imaging device is covered with a part of the indoor unit.
Air conditioner, characterized in that.
The method of claim 1,
And the imaging device is set to face the same direction at the start of operation of the air conditioner when the imaging device is driven.
The method according to claim 1 or 2,
Said same direction is the front of an indoor unit, The air conditioner characterized by the above-mentioned.
The method according to claim 1 or 2,
In the same direction, the air conditioner of the imaging device is substantially perpendicular to the installation surface in front of the indoor unit as viewed from above.
The method according to any one of claims 1 to 4,
The imaging device is configured to be able to change its direction in a predetermined angle range in a vertical direction and a horizontal direction,
Said same direction is an upper limit position or a lower limit position in the angular range of a vertical direction, It is characterized by the above-mentioned.
Air conditioner.
6. The method according to any one of claims 1 to 5,
Part of the indoor unit is air, characterized in that the movable front panel that can open and close the front opening of the indoor unit, or the up and down wind direction change wing to open and close the air vents for blowing air into the room, and to change the air blowing direction up and down Harmonizer.

Images