JPH06149363A - Blow-off air inspection robot for air conditioning - Google Patents

Abstract

(57) [Summary] [Purpose] To provide a robot that automatically discovers an air conditioning air outlet and moves while recognizing its position, and automatically measures the air volume, temperature, etc. of the air blown out. [Structure] A visual system for detecting an air outlet on a ceiling, a moving system for moving in a predetermined direction by a predetermined distance, a sensor system for measuring the air volume and temperature of blown air, a lifting system for raising and lowering a sensor, A control system that stores the position information of the air outlets, obtains information from each system, calculates and transmits the information. [Effect] Since it is not necessary to install a mark for moving the robot, work efficiency is improved. Since there is an accurate drawing of the air outlet, the position can be accurately grasped and the measurement accuracy is improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a navigation technique for a mobile robot and a quality control technique for indoor air-conditioning equipment. The vision system automatically finds an air-conditioning air outlet and moves while recognizing its position. However, the present invention relates to a robot that measures and records the air flow rate and temperature of blown air. This robot sequentially discovers a large number of air outlets arranged in the room one by one by image processing by a visual system. Next, the distance and the direction to the position directly below it are calculated. Then, according to the map information given in advance, it autonomously moves to just below the air outlet to be inspected, and measures and records the air volume and temperature of the air outlet.

[0002]

2. Description of the Related Art At present, just before a new building is completed and completed, it is inspected whether or not air having a design value is supplied from an air-conditioning air outlet. The inspection items in this case are mainly the air volume and temperature. When manually measuring the air volume, for example, the following procedure is performed. First, 3
A group of people look up at the ceiling while looking at the drawing, and look for the outlet to measure. Next, bring the stepladder and place it directly below it, and one person goes up to the stepladder with the anemometer and approaches the outlet. The other extends the measurement hood from below and lifts it up to the outlet to collect all the blown air in the hood. Then, a person on the stepladder puts an anemometer in the hood, waits until the display of the measured value becomes stable, and measures the wind speed at 4 to 9 points. Finally, the measurement result is recorded by a third person, and after that the average value is calculated and multiplied by the hood cross section to obtain the air volume.

As described above, in addition to taking a long time for measurement,
Since there are so many inspection points, we have to take an unreasonable posture for a long time, which is a very difficult work. In addition, the inspection result becomes a huge amount, and the data organization becomes a very complicated work. Furthermore, the data obtained in this way is also unreliable because measurement errors easily occur due to individual differences such as the orientation of the measuring instrument.

A device for mechanically measuring only the air volume has also been devised. For example, the product name "Aero Checker (outlet air volume measuring device)" of Wetmaster is JIS A.
It is said that it is made in conformity with 1431 "Method of measuring air volume of air conditioning and ventilation equipment". This device has a structure in which a person discovers a blowout port, and a person pushes and moves this device by hand, and the wind collecting hood is moved up and down just below the blowout port for measurement. Since this apparatus uses an AC power source, the power cord must be routed around, and setting directly below the outlet is complicated. Therefore, the work for the air flow rate inspection involves physical fatigue, the reliability of the measurement accuracy is low, and there is a problem in the work efficiency for the measurement.

On the other hand, a technique for attaching a so-called "mechanical eye" such as a TV camera to a self-moving robot to obtain image information, processing it, and moving the robot in a predetermined direction is already widely known. , 1990, for example.
"Robot Vision," published by Shokodo in October, 2010, gives a detailed description of a system using a mobile robot and an image pickup tube or a solid-state sensor, and an image processing method thereof.

However, generally, in the conventional mobile robot technology which has been put into practical use, it is necessary to prepare a special mark indicating the movement route on the floor surface or measure the position in advance and install a prism or the like at the reference point. is there. That is,
Prior to inspection, it is necessary to prepare the environment settings beforehand. Further, the conventional mobile robot detects its own moving distance and moving direction by a sensor called an internal sensor. These sensors detect the turning angle of one's own like a gyro and calculate the moving distance from the rotation speed of the wheel to detect the self-position after the movement. However, these internal sensors have a drawback that slight errors in angle detection (drift) and errors in distance detection due to wheel slippage and the like accumulate little by little, resulting in a large position measurement error.

In Japanese Patent Laid-Open No. 57-108777 "Position Detecting Device", there is proposed a method for measuring the self-position by utilizing the unevenness of a wall or a pillar without using a special mark for position recognition. ing. However, the walls and pillars used for self-position measurement are prone to position errors due to construction on site, and the installation accuracy is not so high originally,
The self-position measuring method using this has a drawback that the measurement accuracy cannot be expected so much. In addition, an object placed on the floor of a wall or floor is often obstructed by an object placed on the floor, which is not suitable for a target.

Japanese Unexamined Patent Publication No. 61-221910 "Method for guiding unmanned moving body and AC light source detecting device" proposes a method for guiding a moving body using a fluorescent lamp on the ceiling as a guide mark. Since it depends on the illuminance of the fluorescent lamp,
It has the drawback of being vulnerable to disturbance and difficult to detect stably.

[0009]

SUMMARY OF THE INVENTION An object of the present invention is to move a robot to reach directly below a predetermined air outlet without installing a special mark for position recognition, and An object of the present invention is to provide a blowout air inspection robot for air conditioning, which is capable of performing temperature inspection. Another object of the present invention is to provide a robot capable of autonomously moving by itself and continuously inspecting a large number of air outlets without dragging a power cable.

[0010]

The above-mentioned object of the present invention is to provide a visual system for detecting an air outlet on a ceiling, a moving system for moving a predetermined distance in a predetermined direction, and a moving system for moving the air from the outlet. A sensor system that measures the air volume and temperature of the air that comes out, a lifting system that raises and lowers the sensor system, and position information of the air outlets and stores information from each of the systems to obtain and calculate each system. And a control system for transmitting a predetermined signal to the air-conditioning blow-out air inspection robot.

That is, this robot is composed of the following five subsystems. (A) Vision system (Vision system) This is the part that finds the air outlet on the ceiling and calculates the distance and direction from the robot's current point to the point directly below the outlet. Main CCD camera, remote control lens, electric turning It consists of a stand, a small CCD camera, an image processing device, and the like.

(B) Moving System (Vehicle System) This is a left-right wheel independent drive type moving mechanism, which accurately moves by the data (moving distance and rotation angle) calculated by the vision system. It consists of a DC servo motor, a speed reducer, and the like. The movement method is in-situ rotation-straight movement-in-situ rotation pattern.

(C) Sensor system (wind speed / temperature sensor) This is a system for measuring the wind speed and temperature coming out of the outlet, converting the air volume and recording the data, such as a thermal anemometer and thermometer. Consists of.

(D) Lifting system (elevating mechanism) Raise the sensor system to approach the air outlet,
After the measurement is completed, the system is lowered so as not to hinder the movement, and is composed of a lifting motor, a link mechanism and the like.

(E) Control system The position information of the air outlet is stored, the position of the air outlet is calculated from the position and the direction of the air outlet, and the information from each system is obtained and calculated to each system. Send a predetermined signal. It is composed of one or more CPUs, memories, operation panels, and the like.

[0016]

According to the robot of the present invention having the above structure, the air outlet on the ceiling is used for self-moving navigation. Therefore, it is necessary to prepare a mark indicating the movement route in the room or measure the position in advance. It eliminates the need to set prisms and reflectors at the reference point, which significantly improves work efficiency. In addition, since the robot itself is equipped with the visual system, it is possible to successively move to a large number of air outlets successively, which eliminates the need for manpower and further improves work efficiency.

Generally, since the air outlet for air conditioning is installed with high positional accuracy based on a reliable construction drawing, it is very good as map information for navigation unlike the case where a wall or a pillar is used. Therefore, the self-position measurement error is small. Moreover, since the construction drawings can be easily obtained, an efficient path plan can be immediately established, and it is easy to instruct the robot about the moving direction and the moving distance. further,
Since the target of the movement is the air outlet at the work target, the visual feedback works unlike the movement amount detection (dead reckoning) only by the rotation speed sensor of the wheel that is disturbed by slips. The accumulated error of the internal sensor is reset, and the advantage that the movement error is not accumulated is obtained.

As a preferred embodiment of the present invention, a battery for supplying electric power to each system is mounted on the robot. As a result, the robot can move freely and perform a predetermined work without dragging the power cable.

Furthermore, an image processing device is provided in the visual system, and the image processing device detects the presence of the air outlet because a plurality of squares or circles having the same center of gravity are concentrated. Can be configured. In this way, if the air outlet is recognized by the degree of concentration of a plurality of quadrangular or circular components, it becomes stronger against disturbance and the air outlet is not mistaken for another object such as a fluorescent lamp.
Further, since it is not a processing method with a heavy load such as pattern matching, recognition can be performed in a short time, and work efficiency is improved. Hereinafter, the present invention will be described in more detail with reference to the embodiments of the accompanying drawings.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the entire air-conditioning blowout air inspection robot 10 according to the present invention. The robot 10 has five subsystems, which are air outlets 9 provided on the ceiling.
A visual system 11 for detecting 0, a moving system 12 for moving a predetermined distance in a predetermined direction, a sensor system 13 for measuring the air volume and temperature of the air coming out of the air outlet 90, and lifting for lifting the sensor system 13. A system 14 and a control system 15 for storing position information of the air outlet 90, obtaining information from each of the systems, calculating the information, and transmitting a predetermined signal to each system are provided.

The visual system 11 includes a ceiling air outlet 90.
Is a part for calculating the distance and the direction from the present point of the robot to the point directly below the outlet, and the main CCD camera 24, the remote control lens 25, the electric swivel base 2
6, a small CCD camera 27, an image processing device, etc. (see FIG. 4). Further, a servo amplifier 23 for compensating the servo operation is arranged below the electric swivel base 26.

The moving system 12 is a left-right wheel independent drive type moving mechanism, and a DC servo motor 22 having a built-in reducer so that the moving system 12 can move exactly by the data (moving distance and rotation angle) calculated by the visual system 11. ,, etc. Power is supplied from a battery 21 mounted on the robot. The method of movement is performed in the pattern of in-situ rotation-straight movement-in-situ rotation.

The sensor system 13 is a system for measuring the wind speed and temperature coming out from the air outlet 90, converting the air volume and recording the data, and is arranged at equal intervals on the inner surface of the hood 35 which is enlarged upward. It consists of a thermal anemometer consisting of 24 wind sensor probes 16 and a thermometer.

The lifting system 14 is a system that raises the sensor system 13 to bring it closer to the air outlet 90 and lowers it after the measurement is completed so as not to obstruct the movement of the drive shaft 28. It is composed of an elevating motor 29 and a link mechanism 30 which are connected to each other via a link mechanism 30.

The control system 15 is composed of two CPUs 19 and 20 for performing parallel processing to increase the processing speed, an operation panel 18, a built-in memory and the like. Further, the robot 10 is provided with a wind speed / air temperature monitor 17 for displaying inspection data as needed. Control panel 1
A color liquid crystal display or the like can be attached to 8 if necessary.

FIG. 2 shows a robot 10 according to the present invention, which is used to inspect 31 air outlets 90 arranged on the ceiling of a room in a building from a start point S to an end point E according to a predetermined route. Represents a path plan for carrying out.

FIG. 3 shows a typical shape of an air outlet (anemostat type outlet) installed on the ceiling.
The shapes of the quadrangles having the same center of gravity are concentrated at one point. As another shape, there is a shape in which concentric circles are concentrated at one point. Therefore, in the robot of the present invention, the presence of the air outlet is detected by recognizing that a plurality of squares or circles having the same center-of-gravity position are concentrated by using an image processing apparatus as a preferable mode. It has become.

FIG. 4 shows a CCD camera of the vision system 11,
The remote control lens and the electric swivel base are enlarged and shown. The main CCD camera 24 can be tilted (swinged) in the T direction, and the remote control lens 25 can be moved forward in the F direction or backward in the R direction. The electric swivel base 26 can be rotated in the P direction to pan the direction of the main CCD camera. Hereinafter, the process of inspecting the air volume and temperature of the square anemostat type air outlet using this robot will be described.

First, as a precondition, (1) the position coordinates of all the outlets in the room are given to the robot in advance as map information. Further, each of the air outlets 90 is numbered for distinction. (2) The inspection order and route are also given as shown in FIG. (3) The outlet to be inspected has a shape as shown in FIG.

First, the robot 10 is guided into a room as shown in FIG. 2 and stopped at an appropriate position so that the air outlet 90 to be inspected first comes into the visual field of the main CCD camera 24 of the vision system. Operate manually. The main CCD camera 24 on the robot is fixed to the electric swivel base 26 as shown in FIG.
By tilting, the direction can be changed, and the remote control lens 25 can change the focus (focus), zoom (enlargement / reduction), and iris (aperture) to capture the state of the ceiling surface as an image. Then, the screen is captured so that the air outlet 90 is the most clear. This manual operation is only for finding the air outlet to be inspected first, and for the second and subsequent ones, the electric swivel base is moved toward the direction in which the air outlet to be inspected next can be clearly seen according to the given map information. And the remote control lens is set automatically.

The shape of the ceiling captured on the screen is a grayscale image at first, but the contour is sharpened as much as possible through image processing such as edge enhancement, smoothing, binarization, and isolated point removal. Emphasize the line as shown in. These image processing methods themselves are well-known techniques.

Next, a straight line portion is extracted from this line by linear approximation. The intersection of the extracted straight lines is calculated and found, and in it, the 90 ° key-shaped “<” shown in FIG.
The intersections 91 and 92 in the shape of are picked up. Then, two arbitrary intersections are connected only when the following condition is satisfied. A) The slopes of the straight lines forming the intersections facing each other and the slopes of the straight lines connecting the two intersections are almost equal. B) The parallel lines forming the intersections have the same direction. These conditions are satisfied. In such a case, the two intersecting points are connected to each other to form the groove shape 93 which is open on one side.

By continuing this work, a shape close to a quadrangle is finally produced. Then, quadrangle approximation is performed to calculate the center of gravity of the figure regarded as a quadrangle, and the point where the center of gravity is concentrated is recognized as the center 94 of the air outlet. This state is shown in FIG. Note that these image processing and image recognition processing are performed by a device having a parallel processing function, for example, an arithmetic processing device called a computer, a computer having a multitask function, or the like to reduce the processing time. it can. As an example of the computer, the model number TRPM-401 or TRPM-401A manufactured by Concurrent Systems Co., Ltd. can be used, and as the image processing board, the model number TRP-IMG manufactured by the same company can be used.

Next, a method of calculating the distance and the direction to the air outlet 90 will be described with reference to FIGS. 8 and 9. The pan angle θc (horizontal rotation angle) and the tilt angle Φc (vertical rotation angle) of the main CCD camera 24 in the coordinate system of the robot are detected by an angle sensor (not shown) attached to the electric swivel base 26. To be detected.

On the other hand, the center coordinates 94 of the air outlet obtained by image recognition are represented by a coordinate system on the photographing screen, and this is represented by g (Xg, Yg). Further, assuming that the image resolution is DOTX × DOTY, the outer shape of the air outlet is SX × SY, which is the size shown on the CCD element of the camera, and the focus of the camera is f, the center position coordinates of the air outlet in the robot coordinate system ( x, y, z) is expressed as follows.

[Formula 1]

Then, tan δθ = y / x θ = θc + δθ tanΦ = √ (x ² + y ² ) / z. Here, θ indicates the direction to just below the outlet. Further, from FIG. 9, the distance L from the robot to immediately below the air outlet is represented by L = (H−Hc) tanΦ. Here, both Hc and H are known from the height of the CCD element of the robot and the ceiling height. In this way, the distance and angle to the air outlet are calculated.

The calculation result is transmitted to the mobile system and D
The C servo motor 22 is driven to move in a predetermined direction by a servo mechanism by a predetermined distance. This amount of movement is detected as a feedback amount by the built-in internal sensor, but at this time, it is necessary to move a certain distance and recognize the target outlet again, and calculate the same distance and direction. If this is done, the error in measurement by the rotation speed sensor of the wheel portion does not accumulate as an error in position, and the accuracy of movement is further improved.

Next, the fact that the robot is located immediately below the air outlet 90 is confirmed by the small CCD camera 27 installed vertically upward in the hood 35 of the sensor system 13.
Perform the same process and check. If necessary, make a fine movement to correct the position and direction, and adjust the outlet 90 and sensor hood 3
Match the orientation of 5.

Here, as shown in FIG. 1, the entire sensor system 13 for measuring the air flow rate and temperature is raised by the lifting system 14 and brought close to the air outlet to measure the air flow rate and temperature. The measurement is performed with the wind speed sensor probe 16 and the temperature sensor probe which are installed in the hood 35 at equal intervals. The measurement result is stored in correspondence with the number of the outlet for later processing. Thereafter, by repeating the same steps, all the air outlets in the room as shown in FIG. 4 are automatically inspected according to the given inspection route, and the operation of the robot is completed. The inspection results are tabulated and tabulated for each floor and output from the printer.

FIG. 10 is a control block diagram for operating the robot of the present invention. The elements in each block are the elements described above or known parts. The “servo fixer” is an interface that compensates the operation of the servo mechanism, and is not a unique element of the present invention.

As described above, in the image processing according to the present invention, the air outlet is recognized by the degree of concentration of the quadrangular or circular components, so that the air outlet is strong against disturbance and the air outlet or the like may be mistaken for another object such as a fluorescent lamp. There is no. Further, since it is not a processing method with a heavy load such as pattern matching, if the configuration is such that parallel processing is performed by the transformer as in the above-described embodiment, it is possible to recognize the outlet in a short time. Of course, it is natural that the time required for image processing can be shortened by using a computer having a high operation speed.

In the present invention, the air outlet for air conditioning has been described as the work target, but the robot of the present invention can also be applied to work for the air inlet for air conditioning.

[0043]

As described above in detail, according to the robot of the present invention, since the air outlet on the ceiling is used for the navigation of self-sustaining movement, it is necessary to purposely prepare a mark indicating the movement route in the room, or in advance. This eliminates the need to measure the position and set the prism or reflector at the reference point, thus significantly improving work efficiency. In addition, since the robot itself is equipped with the visual system, it is possible to successively move to a large number of air outlets successively, which eliminates the need for manpower and further improves work efficiency. The air outlet that serves as the position reference is installed with high positional accuracy based on a reliable construction drawing, so there is little error in measuring the self-position, and the air outlet that is the work target can be visually moved. Since the target is set, the movement error due to the wheel speed sensor does not accumulate. Further, an advantage that an object placed on the floor hardly interferes with the self-position measurement is obtained, and the technical effect is extremely remarkable.

[Brief description of drawings]

FIG. 1 is a front view showing the entire robot according to the present invention.

FIG. 2 is a plan view showing an arrangement of air outlets installed on a ceiling in a room.

FIG. 3 is a bottom view showing an air outlet.

FIG. 4 is a front view showing a vision system of a robot.

FIG. 5 is a front view showing an image by a CCD camera.

FIG. 6 is a principle diagram showing an image processing method.

FIG. 7 is a front view showing an image after being processed by a CCD camera.

FIG. 8 is a schematic diagram for explaining a distance and a direction from a CCD camera to an air outlet on a three-dimensional coordinate.

FIG. 9 is a schematic diagram for explaining a distance and a direction from a CCD camera to an air outlet on a two-dimensional coordinate.

FIG. 10 is a block diagram showing control of a robot according to the present invention.

[Explanation of symbols]

 10 robot 11 vision system 12 moving system 13 sensor system 14 lifting system 15 control system 16 wind speed sensor probe 19, 20 CPU 21 battery 22 DC servo motor 24 main CCD camera 25 remote control lens 26 electric swivel 27 small CCD camera 29 lifting Motor 30 link mechanism 35 hood

Front page continuation (72) Inventor Koetsu Tanaka 41 Central Wadai, Tsukuba City, Ibaraki Shinryo Corporation (72) Inventor Yoshio Tanaka 41 Wadai, Tsukuba City, Ibaraki Shinryo Corporation (72) Inventor Toshio Fukuda 2-66 Yata-cho, Higashi-ku, Nagoya-shi, Aichi 122 Oyata Dormitory 122 (72) Inventor Fumito Arai 2-38 Kamenoi, Meito-ku, Nagoya-shi, Aichi 206 Inaka-cho, Inakamachi-cho

Claims (4)

[Claims] 1. A visual system for detecting an air outlet provided in a ceiling, a moving system for moving a predetermined distance in a predetermined direction, and a sensor for measuring an air volume and a temperature of air coming out of the outlet. A system, a lifting system that raises and lowers the sensor system, and a control system that stores the position information of the air outlets, obtains information from each system, calculates the information, and sends a predetermined signal to each system. An air-conditioning blowout air inspection robot characterized by being provided. 2. The air conditioner blown air inspection robot according to claim 1, wherein a battery for supplying electric power to each of the systems is mounted on the robot. 3. The vision system includes an image processing device,
2. The air-conditioning blowout air inspection robot according to claim 1, wherein the image processing apparatus is adapted to detect the presence of an air blowout port because a plurality of squares or circles having the same center of gravity are concentrated. 4. The control system, as a method for guiding an unmanned moving body, recognizes an air outlet on a ceiling surface, measures the position of the robot itself, and calculates a moving distance and a moving direction from map information given in advance. The blowout air inspection robot for air conditioning according to claim 1, wherein