JPS60241177A - Form recognizer - Google Patents

Abstract

PURPOSE:To recognize the 3-dimensional form of a subject at a high speed and in simple constitution by detecting an echo arrived from the subject of a chirp ultrasonic wave, and amplifying and processing the echo. CONSTITUTION:When a chirp signal is produced, a chirp ultrasonic wave is radiated to a subject 4 and a table 3 from an ultrasonic wave transducer 2. Then the echos arrived from the subject 4 and the table 3 are detected by the transducer 2. These echos are amplified at the peak value by an AGC amplifier 51. The amplifier 51 detects the peak value between the start of the echo and the time T of the chirp signal. Then the gain of the amplifier 51 is increased if the detected peak value is smaller than a prescribed level and then reduced when the peak value exceeds the prescribed level respectively. Then the echo peak value is set at the prescribed level since the AGC action is repeated. Then the standardized echos are obtained and the wave is detected through a detector 6 to be supplied to a signal processing circuit 7.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a shape recognition device that recognizes the three-dimensional shape of a target object using ultrasonic waves.

[Prior art]

As a conventional visual sensor used in various robots, etc.
In addition to capturing multiple images using a TV camera and processing the information obtained, a system has been developed that recognizes the shape of a target object. However, the above T
The information obtained by the V-camera is only two-dimensional information that shows the shape of the target object, so in order to recognize the three-dimensional shape of the target object, it is necessary to use special power illumination or install two TV cameras. As a result, the conditions of use are limited, the device itself becomes complex, and the processing time required to recognize the shape of the target object is long. There were some practical drawbacks, such as:

[Summary of the invention]

This invention was made with the aim of improving the drawbacks of the conventional ones as described above, and uses chirp ultrasound as a means of obtaining information indicating the shape of a target object, and uses echoes of the chirp ultrasound from the target object. The object of the present invention is to provide a shape recognition device that can recognize the three-dimensional shape of a target object with a simple configuration and at high speed by detecting, amplifying, and processing the shape.

[Embodiments of the invention]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a shape recognition device according to an embodiment of the present invention. In the figure, a chirp signal generator 2 that generates a chirp signal every fixed period is driven by the chirp signal, emits RC ultrasonic waves in the direction of the target object 4, and detects the echoes of the target object 4). Ultrasonic transducer, 3 is table, 4 is table 3
5 is an amplifier that amplifies the echo signal detected by the ultrasonic transducer 2; 6 is a detector that detects (AM detection) the output of the amplifier 5; 7 is a
A signal processing device processes the detected signal and determines the shape of the target object 4.

@21W is an explanatory diagram of a chirp signal used in the shape recognition device of FIG. 1.

Next, the operation of the shape recognition device shown in FIG. 1, which is an embodiment of the present invention, will be described. First, the chirp signal generator 1 generates a chirp signal at a predetermined repetition period. This chirp signal is an FM wave whose frequency is linearly swept from fl to f2 for a predetermined time T, as shown in FIG. 2(a), and for example, as shown in FIG. 2(b). It is a signal with various waveforms. This signal is a well-known signal that is also used for pulse compression in radar, and is expressed by the following equation.

Here, '91rl=2gr, l w, ==2zf2
a &=constant (coefficient). The above signal is given to the ultrasonic transducer 2, and the ultrasonic transducer 2 emits ultrasonic waves. Chirp signal generator 1
If necessary, a power amplifier (not shown) is inserted between the ultrasonic transducer 2 and the ultrasonic transducer 2.・Ultrasonic transducer 2#: t is arranged so that multiple ultrasonic waves are emitted from directly above the target object 4 placed on the table 3. Ultrasonic waves are shown in Table 3. Target object 40
It is reflected from the surface and returns to the ultrasonic transducer 2. Now, the target object 4 has one surface that reflects ultrasonic waves as shown in FIG. 1, and if the height from the table 3 is The reflected waves of are respectively expressed by the following equations.

eo == a. ,,jl ・(t---)+wl
)・t...(2)T eH=aH-t"-・(t-""10") +Wl)
-tT CC (3) Here, C is the speed of sound, and h is the height of the ultrasonic transducer 2 from the table 3. Further, 'DIILI is a constant (coefficient) of a reflecting surface representing the degree of reflection of ultrasonic waves from the table 3 and the target object 4, respectively.

Both reflected waves interfere with each other and reach the ultrasonic transducer 2, where the next signal is detected.

e””eo+J W, -w.

x output ((T t + w1) is 1 ψ) - (4) This signal is amplified by the amplifier 5 and then detected by the detector 6.

Then, the following signal is extracted from the above equation (4).

The above equation (5) expresses the characteristics (wl) of the chirp signal to be used.

Determine T) is a function of time determined by the constant &(1, &) of each reflecting surface and the height X of the target object 4. Generally, as shown in Fig. If n4m exists, a signal according to the following equation is extracted.

W, -W, * × stroke ((-t + w, ) t-ψ) ... (6) l
0=0.

In this case as well, the constant aOr al + 't of each reflective surface
+...an and the distance xI between the table 3 and each reflective surface of the target object 4.

``tw... is a function of time determined by xn.

Therefore, a signal with a waveform corresponding to the shape of the target object 4 is obtained as the output of the detector 6. Figure 4(a)-(i)
shows an example of a signal with the above waveform. Figure 4(a)
The waveform signals shown in ~(i) correspond to the target objects 4 having shapes as shown in FIGS. 5(a) to (i), and all of these target objects 4 are composed of cylinders. It is difficult to easily distinguish them just by looking at them from above. Each of the above-mentioned signals is processed by the signal processing device 7, and the shape of each target object 4 is recognized.

FIG. 6 is a block diagram showing in detail the signal processing device in the shape recognition device of FIG. 1. In the figure, 71
is an ADO (analog-to-digital converter), 72 is a buffer memory that temporarily holds the digitized signal, 73 is a normalization calculator for normalizing the contents of the backup memory 72 to the maximum value, and 74 is a target a pattern recording memory 75 for recording a waveform pattern corresponding to the shape of the object 4; a comparator 76 for comparing the pattern in the pattern recording memory 74 with the output signal of the normalization calculator 73; is a determiner that determines the shape of the target object 4 based on the result of the comparator 75. Now, in the signal processing device as shown in FIG. .

The above data is searched for the maximum value by the normalization calculator 73 and normalized accordingly. The result is sent to either the pattern recording memory 74 or the comparator 75, depending on the operating mode of the shape recognition device. The data is sent to the pattern recording memory 74 in the teaching mode, and prior to shape recognition, the shape of the target object 4 and the detected pattern are correlated through the operator's judgment. . That is, the operator places the target object 4 on the table 3 and presses a teach button on a controller (not shown).

At this timing, the output of the normalization calculator 73 is input to the pattern recording memory 74 and recorded.

On the other hand, the output of the normalization calculator 73 is sent to the comparator 75 in the recognition operation mode, and is compared with the content previously recorded in the pattern recording memory 74 in the teaching mode. Determine whether it corresponds to the object shape. That is, when the target object 4 is placed on the table 3, at this timing (given to a controller not shown by some method not directly related to this invention),
The result of the normalization calculator 73 is sent to the comparator 75, and the patterns recorded in the pattern recording memory 74 are given one after another to the comparator 75 and the two are compared.

The comparison result is given to the determiner 76, and if there is a match, it outputs a signal with a shape corresponding to the match.

In the above embodiment, in the signal processing device 7, the ADC 71
The buffer memory 72 and the normalization calculation unit 73 are provided immediately after the amplifier 5, and the amplifier 5 is an AGC (automatic gain) amplifier (see 51 in Fig. 7) based on the peak value, and the normalization calculation is performed on this amplifier. It can also have functions. With such a configuration, 72 buffer memories are not necessarily required.

FIG. 7 is a block diagram showing a shape recognition device according to another embodiment of the present invention. In the configuration shown in Figure 7, the signal is AD converted and recorded as digital data only in the teaching mode, and in the recognition operation mode, the signal is compared with an analog waveform for high-speed processing. The device itself can be simplified. The operation will be explained with reference to the figure. First, a chirp signal generator IK command is given to the controller 8 and a chirp signal is generated. 7. The ultrasonic transducer 2 receives a target object 4. A chirp ultrasonic wave is emitted toward the table 3, and these target objects 4. Echoes from the table 3 are detected by the ultrasound transducer 2. This echo line is amplified by the AGC amplifier 51 based on the peak value, but the AC, C
The amplifier 51 detects the peak value between the time T of the chirp signal from the start of the echo, and if this value is less than a predetermined value, the gain of the AGC amplifier 51 is increased, and if it exceeds a predetermined value, the gain is decreased. . Since the controller 8 gives commands to the chirp signal generator l at regular time intervals, the above-mentioned eaves operation is repeated nine times, so that the peak value of the echo becomes a predetermined value and is normalized. I get an echo. This echo is detected by a wave detector 6 and input to a signal processing circuit 7. In the signal processing device 7, in the teaching mode, the input signal is converted into digital data by the ADC 71 and stored in the pattern recording memory 74. In the recognition operation mode, the input signal sent from the detector 6 is input to the comparator 75. On the other hand, the contents of the pattern recording memory 74 are read out at the same timing as the input signal, converted to analog data by a DAC (digital-to-analog converter) 77, and input to the comparator 75. The comparator 75 compares the two and sends the result to the determiner 76, which determines whether or not they have corresponding shapes. Reading from the pattern recording memory 74 is performed by sequentially updating the data corresponding to each shape in accordance with the ultrasonic transmission/reception operation, and by sequentially comparing the data, the data corresponding to the target object 4 recorded in advance is read out. Can process all shapes. In addition, the turn recording memory 74, DAC 77, and comparator 75 can be configured to operate in parallel for all shapes of the target object 4 recorded in advance.Comparator 75
It can be easily constructed using a Mori subtraction circuit and an integration circuit.

Further, the signal processing device 7 can be composed of an ADC 71 and a microcomputer (not shown), and although the processing speed is slightly lowered by this, it is possible to obtain a highly flexible device.

In addition, in the above practical example, the case where the ultrasonic transducer 2 is used for both transmission and reception was explained, but two separate ultrasonic transducers are used for each purpose.
It goes without saying that transmission and reception may be performed separately.

By the way, as can be seen from equation (6) above, the waveform of the received echo includes information regarding the shape of the target object 4, but there is no one-to-one correspondence. Practically speaking, there is no problem as long as the shape of the target object 4 is limited. That is, there is a one-to-one correspondence between the target object 4 and the echo waveform.

However, if the number of types of shapes of the target object 4 increases, this condition may no longer hold true. In this case, as shown in FIG. 8, the determination can be made by obtaining echoes from two or more directions and processing them. As shown in Figure 8, 2
Reference numeral 1.22 denotes ultrasonic transducers, each of which emits Chirbian ultrasonic waves toward the target object 4 and detects echoes from the target object 4. This detected signal is
The signals are amplified by independent amplifiers 52 and 53, respectively detected by detectors 61 and 62, and then processed by the signal processing device 7.

Furthermore, if T is determined by the constant of the chirp signal generated from the chirp signal generator l, then there is a range in the dimensions of the target object 4 from which shape information of the target object 4 can be obtained. It is also possible to expand the range of the shape of the target object 4 by controlling the frequency sweep range of the chirp signal generator 1 and changing each of the above constants according to the processing results of the signal processing device 7.

In addition, in the above embodiment, the case where the shape of the target object 4 placed in the air is recognized, but in addition to this, when the ultrasonic flaw detection device recognizes the shape of a defect in the object to be tested, can also be fully applied.

〔Effect of the invention〕

As explained above, the present invention provides a shape recognition device that
Chirped ultrasound is used as a means of obtaining information indicating the shape of the target object, and the configuration is configured to detect, amplify, and process the echoes of the chirp ultrasound from the target object, making it extremely simple and fast. The present invention has the excellent effect of providing a shape recognition device that can recognize the three-dimensional shape of a target object, and has high sensitivity and reliability, especially for recognizing the shape of a target object in the depth direction. .

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram showing a shape recognition device which is an embodiment of the present invention, FIG. 2 is an explanatory diagram of a chirp signal used in the shape recognition device of FIG. 1, and FIG. Figure 4 (→~(i) is a diagram for explaining the operating principle of the shape recognition apparatus in Figure 1. Figure 5 (intimidation~(i) 6 is a block diagram showing in detail the signal processing device in the shape recognition device 1tK of FIG. , FIG. 7, and FIG. 8 are block diagrams respectively showing shape recognition devices according to other embodiments of the present invention. In the figures, l...chirp signal generator, 2.21°22... ffl sound wave transducer, 3... table, 4-... target object, 5.52.53... amplifier,
6,61°62...Detector, 7...Signal processing device,
8... Controller, 51... AGC amplifier, 71...
ADC (analog-digital converter), 72... Buffer memory, 73 - Normalization calculator, 74... Memory for pattern recording, 75... Comparator, 76... Determiner, 7
7... DAC (digital-to-analog converter). In each figure, the same reference numerals indicate the same or equivalent parts. Agent Masuo Oiwa Figure 1 Figure 2 Cause Figure 3 Day 2 =; Representative: Hitoshi Katayama, Department 4, Agent address: 5, Mitsubishi Electric Corporation, 2-2-3 Marunouchi, Chiyoda-ku, Tokyo, Japan. Subject of amendment: ``Detailed description of the invention'' column 6 of the @ letter. Contents of amendment (1) "This result," in the 4th line of page 10 of the specification is amended to "this result is," (2) "Circuit 7" on page 12, line 20 of the same book is replaced with "device 7".
” he corrected.

Claims (8)

[Claims] (1) A chirp signal generator that generates a chirp signal; a transmitting ultrasonic transducer that is driven by the chirp signals generated by the chirp signal generator and generates chirp ultrasonic waves toward a target object; A reliable ultrasonic transducer includes a receiving ultrasonic transducer that detects the echoes of the generated chirp ultrasonic waves from the target object, an amplifier that amplifies the signal detected by the receiving ultrasonic transducer, and a digital signal that is connected to the amplifier. Shape recognition comprising: a detector that detects an amplified signal; and a signal processing device that processes the signal detected by the detector and recognizes the shape of the target object. Device. (2) The shape recognition device according to claim 1, wherein the transmitting ultrasonic transducer and the receiving ultrasonic transducer are constituted by one ultrasonic transducer. (3) The signal processing device according to claim 1 is characterized in that the signal processing device is composed of an AD converter, a buffer memory, a normalization calculator, a pattern recording memory, a comparator, and a determiner. Shape recognition device. (4) The shape recognition device according to claim 1, wherein the amplifier is configured to have an AGC function that maintains a maximum value of echoes from the target object at a constant value. (5) The shape recognition device according to claim 1, wherein the signal processing device includes multiple components including an AD converter, a buffer memory, a pattern recording memory, a comparator, and a determiner. (6) The shape recognition device according to claim 1, wherein the signal processing device includes an AD converter and a microcomputer. (7) The shape recognition device according to claim 1, characterized in that the frequency sweep range of the chirp signal generator can be changed according to a command of the signal processing device. . (8) The shape recognition device according to claim 1, characterized in that two or more sets of the ultrasonic transducer, the amplifier, and the detector are provided.