JPH0582601B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a control device for a self-propelled robot, and relates to a scene recognition method that can obtain accurate scene recognition and simplify travel control. It is.

[Background of the invention]

The scene recognition method of a self-propelled robot that runs throughout a limited room, as seen in conventional automatic vacuum cleaners, is
As described in Japanese Patent Application Laid-Open No. 55-97608, this is a method of detecting obstacles using an ultrasonic transmitter and a receiver.
If there is an obstacle, the vehicle will stop and change direction by 180 degrees. However, in reality, if there are obstacles in the room, there will be areas where the vehicle has not traveled, as shown in Figure 2 of the main text, and the vehicle must search for and travel to those untraversed areas. In the prior art, this point has not been taken into consideration.

To explain FIG. 2, a self-propelled robot travels in a rectangular room ABCD from point A to point B by repeatedly going straight and making U-turns. Inside room abcd,
Assume that there is an obstacle in the shaded area efgh. Therefore, in order to travel throughout room ABCD, search for untraveled areas HGCI that have not been traveled yet,
The next target point to move from point B, point C, must be set. In the prior art, this point is not taken into consideration.

As a method for easily searching for the untraversed area, there is a method of teaching the control device in advance the size and shape of the room in which the vehicle should travel, as shown in FIG. 3 of the main text. This teaching method will be explained with reference to FIG. In Figure 3, abcd is the rectangular room in which the vehicle actually travels (in this example, the shape is the same as in Figure 1 for ease of explanation.) Also, inside the room abcd, there is an obstacle efgh. shall be.
The self-propelled robot then travels from point A to point B by repeatedly going straight and making U-turns. In this case, if the positions of walls and obstacles in the room are detected using the ultrasonic transmitter and receiver shown in Figure 2 of the drawings of the conventional Japanese Patent Application Laid-Open No. 55-97608, and those positions are combined, the main text Walls and obstacles in the room are recognized along lines ab, bc, id, da, and fe in Figure 3. Note that the ultrasonic transmitter and receiver are sensors that can measure the direction and distance of a reflecting surface of an obstacle perpendicular to the ultrasonic traveling direction. As described above, the rectangle OPQR shown in FIG. 3 is taught to the control device in advance as the size and shape of the room in which the vehicle should travel. Here, of course, the size of the teaching room OPQR is as shown in Figure 3.
It must be set fairly large so that it can accommodate any room in ABCD. Normally, if you think about it, you would have no choice but to set it 3 to 5 times as wide. Under these conditions, the self-propelled robot that has traveled to point B in FIG. 3 must determine whether there is any area that it has not yet traveled.

In this case, the robot must search to see if there is an area in which it can run within the entire range of OPQR, which has been set in advance as a room in which it should run. Then, set the target point to which you should move next from point B (for example, target point C when searching for the untraveled area as hgci in Figure 3, or target point D when searching for the untraveled area as OPQR). Must.

However, the size of the room you should run in advance,
If you use the method of teaching the shape, you need to set the size of the room in which you are teaching to be quite large so that it can accommodate any room. Therefore, when the self-propelled robot is surrounded by obstacles as at point B in FIG. 3, there is a drawback that the calculation processing time required to search for the untraveled area and the next moving target point is extremely long. For this reason, there was a drawback that the time required to completely travel through the room was extended.

[Purpose of the invention]

An object of the present invention is to provide a self-propelled robot that does not require the user to teach the range of the room in which the robot is to run, in other words, allows the self-propelled robot to recognize the range of the room by itself, and to shorten the time required to search for untraveled areas. Our goal is to provide robots.

[Summary of the invention]

Rather than teaching the control device the size and shape of the room in which the vehicle should travel in advance, the x and y coordinates of the positions of obstacles (including the walls of the room) detected by an ultrasonic transceiver are used to determine the left and right range of the room. Set the minimum value Xmin and maximum value Xmax of the x coordinate in the system,
Also, set the upper and lower range of the room using the maximum value Ymax and minimum value Ymin of the y-coordinate. Then, the untraveled area is defined by four straight lines x=Xmin, x=Xmax, y=
We devised a method to search for an area in which a robot can run within a rectangular area surrounded by Ymin and y = Ymax. Therefore, according to this method, there is no need for the user to teach the self-propelled robot in advance the range of the room in which it should travel, so that the self-propelled robot can be completely automated.

In addition, with this method, the size of the room in which the robot should run can be set to a size close to the actual size, and if the self-propelled robot becomes unable to run due to being surrounded by walls or obstacles in the room, it will be possible to set the size of the room it should run in to close to the actual size. The calculation processing time for searching can also be reduced.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. An embodiment of the self-propelled robot of the present invention will be explained using an example of a self-propelled robot for cleaning.

FIG. 1 is a block diagram showing the configuration of a self-propelled cleaning robot according to an embodiment of the present invention. 1st
In the figure, 1 is an arithmetic processing unit that controls the self-propelled robot, 2 is a storage device that stores programs and data of the arithmetic processing unit 1, 3 is an input port for receiving signals from an input device to the arithmetic processing unit 1, and 4 is an input port for receiving signals from an input device to the arithmetic processing unit 1. An output port for outputting a signal from the processing unit 1 to each driving device; 5 is an ultrasonic transmitter that emits directional ultrasonic waves; and 6 is an ultrasonic transmitter that emits ultrasonic waves that hit an obstacle and are reflected. 7 is an ultrasonic transmitter/receiver tachometer that measures the ultrasonic transmitting/receiving direction of the ultrasonic transmitter 5 and receiver 6, and 8 is a left and right drive wheel of the self-propelled robot. 9 is a wheel drive device; 10 is a rotation drive device for an ultrasonic transceiver; 1
1 is a dust suction drive device for cleaning.

Next, the operation will be explained. If there is an obstacle efgh in a rectangular room as shown in Figure 2 above, the self-propelled robot starts from starting point A, moves straight and repeatedly makes U-turns, and runs surrounded by the obstacles. Drive to point B where it is no longer possible, search the untraveled area hgci, and then find the target point C to which you should move next.
An example of searching will be explained using the processing flowcharts shown in FIGS. 4 and 5 and the plan views showing the traveling trajectory of the self-propelled robot, the room shape, and the position of obstacles in FIGS. 6, 7, and 8. do.

First, with reference to FIG. 4, a travel control method performed by the arithmetic processing device 1 of FIG. 1 will be described. The running pattern of this embodiment is similar to that shown in FIG. 2, in which the self-propelled robot moves straight if there is no obstacle in front of it in the direction of travel, and if there is an obstacle, it makes a U-turn. In FIG. 4, the arithmetic processing _unit 1 first inputs rotation speed data of the left and right wheels from the wheel tachometer 8 shown in FIG _. It is calculated and stored in the storage device 2. Figure 6 shows an example of the self-position (x _i , y _i and the _traveling direction θ ₎ . The situation in which the robot moves to point E (x _i , y _i ) for the first time is shown.

Since the rotation speed of the left and right wheels is measured from the wheel tachometer 8 shown in Fig. 1, the robot's traveling direction θ can be easily calculated from the distance traveled from point A to point E and the difference in the rotation speed of the left and right wheels (calculation method). (explanation omitted). In FIG. 6, the value of θ is expressed as an angle between a line segment FE passing through point E and parallel to the y-axis and the robot's advancing direction EE'.

Then, based on the self-position (x _i , y _i ) of the robot calculated above, the cleaning area to be cleaned by the dust suction drive device 11 shown in FIG. 1 is calculated, and this data is also stored in the storage device 2. FIG. 6 shows an example of the cleaning area. In FIG. 6, it is assumed that the range to be cleaned by the robot is a rectangle STUV as indicated by the starting point A. Therefore, when the self-propelled robot travels from the starting point A to point E, the area cleaned during that time is the shaded area of the rectangle STWZ.

Next, the processing unit 1 inputs the direction data α of the obstacle as seen from the robot from the ultrasonic transmitter/receiver tachometer 7, and at the same time, the ultrasonic transmitter 5 and receiver 6
Distance data l to the obstacle, which is measured by the time T from when the ultrasonic wave is transmitted until it is reflected from the obstacle and received, is input. Then, the coordinates (x _s , y _s ) of the obstacle are calculated and stored in the storage device.

FIG. 6 shows an example of detecting the obstacle.
In Figure 6, when the robot travels to point E (x _i , y _i ), if an obstacle G is detected at a distance l in the direction of angle α to the left from the robot's traveling direction θ, then The position (x _s , y _s ) in the xy obstacle system is x _s
= x _i −l _sio (θ+α), y _s =y _i +l _cps (θ+α). Note that the distance l to the obstacle is determined by measuring the time T from when the ultrasonic transmitter 5 in FIG. From T, if the propagation velocity of the ultrasonic wave is V _ONP , it is calculated as l=T/2·V _ONP .

The position (x _s , y _s ) of the obstacle G is calculated using the above method. Therefore, by the above calculation, the self-propelled robot has stored data on its own position, direction of movement, position of obstacles, and cleaning area in the storage device 2 shown in FIG.

Next, the arithmetic processing unit 1 makes a running judgment based on the data of the robot's own position and direction of movement, the position of obstacles, and the cleaning area stored in the storage device 2, and drives the wheel drive device 9 and the ultrasonic rotary drive device 10.
and outputs travel commands and control commands to the dust suction drive device 11.

The processing for determining whether the vehicle is running is shown below in FIG.
The arithmetic processing unit 1 determines whether the robot should go straight or make a U-turn based on the data stored in the storage device 2 about the robot's own position, direction of movement, and position of obstacles. Fig. 8 shows the actual running method of the robot.

In FIG. 8, obstacles are detected scattered along a rectangle ABCD and a line segment EF, as represented by point J. In such a situation where there are obstacles, the self-propelled robot starts traveling from the starting point A. Then, as shown below in FIG. 4, it is determined whether there is an obstacle ahead. If there is no obstacle ahead, the processing unit 1 outputs a command to the wheel drive unit 9 to go straight. If there is an obstacle ahead, stop, judge whether a right U-turn or left U-turn is possible, and if a right U-turn is possible, issue a right U-turn command, and if a left U-turn is possible, issue a left U-turn command. Output to the wheel drive device 9. However, as shown in Fig. 8, right U-turns and left U-turns are determined by switching alternately. When the robot travels to point B in FIG. 8, where it can no longer make a right U-turn or a left U-turn, the processing unit 1 determines whether the robot has not traveled yet based on the data of the obstacle position and cleaning area stored in the storage device 2. Determine whether the area exists. If there is no untraversed area, the robot will stop and the control will end. If there is an untraversed area, as shown below in Figure 4, search for the untraversed area and find the next target point from point B in Figure 8, where the robot is surrounded by obstacles and cannot travel. C is searched, and then a travel route from point B to point C to the target point C shown in the example of BB ₁ B ₂ C in FIG. 8 is searched. However, this travel route BB ₁ B ₂ C is an example assuming that the vehicle travels along the x and y axes. Then, the arithmetic processing device 1 outputs a command to the wheel drive device 9 to run along the searched travel route. After reaching the target point C, the vehicle repeats going straight and making U-turns again, and outputs a command to travel throughout the untraveled area hgci.

Next, a method of recognizing the size and shape of a room in which a self-propelled robot should run throughout, which is a feature of the present invention, will be explained.

In the present invention, the size of the room to be traveled through is determined by x=Xmin in the xy coordinate system, as shown in FIG.
It is represented by a rectangle abcd surrounded by x=Xmax, y=Ymin, and y=Ymax. And the above Xmin,
Each time an obstacle is detected, the values of Xmax, Ymin, and Ymax are compared with the x and y coordinates of the obstacle's position, and Xmin is the minimum value of the x-coordinate of the obstacle's position, and Xmax is the minimum value of the x-coordinate of the obstacle's position. The maximum value of the x-coordinate of the position,
Correct Ymin to the minimum value of the y-coordinate of the obstacle position, and Ymax to the maximum value of the y-coordinate of the obstacle position. Therefore, the range in the x-axis direction of the room in which you should run (left and right direction in Figure 7) is from straight line x = Xmin to straight line x = Xmax.
In the range of y-axis direction (vertical direction in Figure 7)
is grasped in the range from straight line y=Ymin to straight line y=Ymax.

Next, Xmin, Xmax, which indicates the range of the above room,
We will explain how to set the initial values of Ymin and Ymax and how to correct them each time an obstacle is detected while the robot is running. The initial values of Xmin, Xmax, Ymin, and Ymax are the starting point A (x _p , y _p ) in FIGS. Set to x and y coordinates. That is, let Xmin=Xmax=x _p and Ymin=Ymax=y _p .

After calculating the obstacle position (x _s , y _s ) indicated by ~ in the processing flowchart of Fig. 4, and before making a decision on travel control, the room ranges Xmin, Xmax, Ymin, Ymax of the obstacle
Performs processing to compare and correct the xy coordinates and their sizes.
FIG. 5 shows a calculation method for .about. in the processing flowchart of FIG. 4. In Figure 5, first the coordinates of the obstacle (x _s , y _s )
Read, determine whether the x coordinate x _s is smaller than the previous Xmin, and if it is smaller, correct Xmin to x _s , similarly determine whether x _s is larger than Xmax, and if larger, change Xmax to x _s . Fix it. Similarly, for the y-coordinate, determine whether the y-coordinate y _s is smaller than the previous Ymin, and if it is smaller, correct Ymin to y _s , similarly determine whether y _s is larger than Ymax, and if it is larger, change Ymax to y Correct to _s .

Therefore, the situation of correction will be explained graphically in connection with the running of a self-propelled robot. A specific example of correction near the starting point A will be described in FIG. FIG. 6 is a diagram in which the robot starts from point A, travels to point E, and detects an obstacle at point G (x _s , y _s ) as described above. At the starting point A, as mentioned above, Xmin=
Xmax=x _p , Ymin=Ymax=y _p , that is, the room in which the vehicle should run is understood to be a point, not a rectangle. Then, at point E, make the correction according to Figure 5,
That is, since x _s <Xmin, it is corrected to Xmin=x _s , and since y _s >Ymax, it is corrected to Ymax=y _s . Therefore, at point E, the rooms to travel through are x = Xmin = x _s , x = Xmax = x _p , and y = Ymin.
It can be understood as a rectangle HAIG surrounded by = y _p and y = Ymax = y _s .

Next, a specific example of correction when the robot has traveled to a certain extent will be described with reference to FIG. FIG. 7 shows a situation in which the robot starts from point A, travels straight ahead and repeatedly makes U-turns to point H, and at point H, an obstacle J point (x _s , y _s ) is detected. Then, the size of the room in which you need to run until you reach point H is x=Xmin, x=
Suppose that it is grasped by a rectangle abcd surrounded by Xmax, y=Ymin, and y=Ymax. Then, since an obstacle J (x _s , y _s ) was detected at point H, corrections were made according to Figure 5, and since x _s > Xmax, Xmax = x _s
will be corrected. Therefore, at point H, the room to be traveled is changed from abcd to ab'c'd. Note that the shaded area SU ₂ U ₃ U ₄ U ₅ U ₆ in Fig. 7 is
The cleaning area when the robot travels from A to H is shown.

Next, FIG. 8 shows a method of searching for untraveled areas. Figure 8 shows a robot starting from point A and repeatedly moving straight and making U-turns until it reaches point B, where it is surrounded by obstacles and cannot move. (represented by J) is plotted on the same coordinates. The shaded area abfeid is the cleaning area of the robot when the robot travels from point A to point B. However, the robot was supposed to clean the STUV range using point A as an example. As mentioned above, the size of the room in which you should run at point B is x=
It can be understood as a rectangle abcd surrounded by Xmin, x=Xmax, y=Ymin, and y=Ymax. Therefore B
For points, if the outline of the self-propelled robot is STUV in the example of point A within rectangle ABCD, a search is made to see if there is an area in which that STUV can run. In FIG. 8, since there is an area in HGCI where the robot can run, that HGCI is set as a non-running area.

Further, at point B, an appropriate target point C within the untraveled area is selected. Then, a traveling route BB ₁ B ₂ C is searched for traveling from point B to the next target point C. This traveling route BB ₁ B ₂ C is an example of traveling along the x-axis and y-axis.

According to this embodiment, there is no need for the user to teach the robot in advance the range of rooms in which it should travel, so the self-propelled robot can be completely automated. Therefore, the self-propelled robot becomes easier to use and has the effect of increasing control flexibility.

Furthermore, according to this embodiment, the range of the room that the self-propelled robot must travel through is corrected to the minimum and maximum values of the coordinates of obstacles, so the size and shape of the room can be made close to the actual size. Since the robot can be set, if the robot is surrounded by walls or obstacles in the room and cannot move, the calculation processing time to search for areas where it has not yet moved will be reduced compared to the conventional room size shown in Figure 3. This has the effect of shortening the length of the shape rather than teaching it in advance.

Specifically, if we compare the calculation processing time required for one search for an untraveled area, we can teach the data of a 10m x 10m room and search for the untraveled area hgci or OP'Q'R' in Figure 3. This takes 5 to 10 seconds on a 16-bit microprocessor. However, if the method of this embodiment is used to determine the size and shape of the room based on the positions of obstacles, the time can be reduced to 1 to 2 seconds, which is about 1/5 of the conventional method. Figures 6, 7, and 8 of the embodiment
In the figure, we explained a simple environment with one obstacle in the room, but if a self-propelled robot were to run in an actual room, it is expected that the number of times the robot would become unable to move due to being surrounded by obstacles is expected to be quite large. Therefore, the effect of this embodiment is great.

Therefore, if the calculation time for searching the untraversed area can be shortened, the time it takes for the self-propelled robot to travel throughout the room, that is, the time it takes for the self-propelled robot to travel throughout the room, or the time it takes for the self-propelled robot to travel throughout the room, can be reduced. When used as a vacuum cleaner, the time required for cleaning can be shortened and it is also economical.

Although this embodiment has been explained using a self-propelled robot for cleaning as an example, it is also applicable to a self-propelled robot for painting, or for excavating earth and sand, which is required to run all over the room. It can be applied to self-propelled robots, etc.

〔Effect of the invention〕

According to the present invention, there is no need for the user to teach the robot in advance the range of rooms in which it should travel, so the self-propelled robot can be completely automated, and the self-propelled robot is easier to use.
This also has the effect of increasing control flexibility.

Furthermore, according to the present invention, each time an obstacle is detected, the range of the room in which the vehicle must travel is corrected by the minimum and maximum coordinates of that obstacle, so the size and shape of the room can be adjusted. Since it can be set to a similar shape, it has the effect of reducing the amount of processing equipment required to search for areas where the vehicle has not yet traveled. Therefore, the robot runs smoothly, and at the same time,
It also has an economical effect because the travel time can be shortened.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the configuration of a self-propelled cleaning robot that is an embodiment of the present invention, Fig. 2 is a plan view of the travel route of the self-propelled robot, and Fig. 3 is a conventional method for searching areas where the robot has not traveled. A plan view showing the method, Fig. 4,
Figure 5 is a flowchart of calculation processing in the embodiment;
7 and 8 are plan views showing a method of recognizing the size and shape of a room in which the vehicle must travel and a method of searching for an untraversed area, according to the embodiment. DESCRIPTION OF SYMBOLS 1... Arithmetic processing unit, 2... Storage device, 3... Input port, 4... Output port, 5... Ultrasonic transmitter, 6...
Ultrasonic receiver, 7... Ultrasonic transmitter/receiver tachometer, 8...
Wheel rotation meter, 9... Wheel drive device, 10... Ultrasonic transmitter/receiver rotation drive device, 11... Dust suction drive device.

Claims (1)

[Claims] 1 An ultrasonic transmitter, an ultrasonic receiver, an ultrasonic transmitter/receiver tachometer that measures the direction and distance of obstacles, a measuring device that measures the robot's own position and direction of movement, and arithmetic processing that processes the measurement data. and a storage device for storing the calculation results, and the calculation processing device calculates the minimum value Xmin and maximum value Xmax of the x-coordinate values and the y-coordinate value of the position coordinates of all obstacles based on the calculation results. Minimum value Ymin and maximum value
Ymax is calculated, and the robot's traveling range is defined as two straight lines parallel to the x-axis in the xy orthogonal coordinate system, x=Xmin,
x = Xmax and two straight lines parallel to the y axis y = Ymin, y
=Ymax, and the arithmetic processing unit outputs a running command to the robot's driving device in relation to the position coordinates of the obstacle detected by the ultrasonic transmitter and receiver. robot x
A self-propelled robot configured to modify the range in which it should travel by moving two straight lines x=Xmin, x=Xmax parallel to an axis and two straight lines y=Ymin, y=Ymax parallel to a y-axis.