JPH047921B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a contact sensor for determining the shape of an object by contact with the object.

<Prior Art and Problems to be Solved by the Invention> As sensors used for the purpose of object recognition, sensors that utilize images using television cameras are considered important and are being actively researched.

However, video signals have a large amount of information and are complex to process, requiring sophisticated hardware and a huge amount of software. There are still many problems in terms of processing time and price.

On the other hand, a method of determining the shape of the object in contact may also be considered. For example, a sensor can be attached to the floor to determine the shape of an object placed on it, or a sensor can be attached to the grip of a robot hand to determine the shape of an object being grasped. Possible uses include inspection of products in factories and sorting processes. Furthermore, the planar contact sensor can also be applied as an input device for computers and other information processing equipment.

As an example of the structure of a sensor used for such a purpose, there is an example in which a planar sensor is constructed by arranging switches from which on/off signals can be obtained in a matrix.

In this method, switches with mechanical contacts are lined up, which naturally limits the number of pixels that can be configured. It also becomes difficult to make the sensor itself thinner and smaller.

In order to solve these problems, research is being carried out using pressure-sensitive conductive rubber.

Pressure-sensitive conductive rubber is made by uniformly mixing carbon particles into silicone rubber and forming it into a planar shape.When pressure is applied perpendicular to the surface, the vertical resistance of the area where the pressure is applied decreases, causing a localized It has the property of providing a conductive state, and is increasingly being used in computer input devices and keyboards for electronic desk calculators. In these devices, each switch is constructed using pressure-sensitive conductive rubber as a contact point. In order to obtain a thin and small contact sensor, arranging individual switches in a matrix does not solve the problem. Therefore, as shown in Figure 1, striped electrodes 2 are arranged so that the lattice points are arranged in a matrix.
A conceivable method is to construct a contact sensor by sandwiching the upper and lower surfaces of the pressure-sensitive conductive rubber 23 using the contact sensors 1 and 22. In this case, a certain voltage is sequentially applied to the electrodes 21 in the rows on the upper surface, and the output state of the electrodes 22 in the columns on the lower surface is observed. When pressure is applied to one lattice point, only that part of the pressure-sensitive conductive rubber 23 becomes conductive in the vertical direction, so that an output is obtained from the column electrode 22 corresponding to that lattice point.

However, when applying pressure to multiple lattice points and attempting to determine their shape, multiple lattice points become conductive, creating an unexpected current path, and output is also output from the column electrodes 22 at lattice points to which no pressure is applied. The resulting phenomenon occurs.

For example, as shown in Figure 2, even if a 2 x 2 matrix is configured and pressure is applied to only three points, pressure will also be applied to the grid points to which no pressure is applied because of the electrical path made up of those three points. It will be detected as if it were there.

In order to avoid this inconvenience, it is conceivable to insert a diode at each grid point to prevent backflow of current. However, in this case, the flexibility of the rubber itself is lost, making it impossible to use it by pasting it on a cylindrical surface, and making it impossible to construct a small matrix. A method has also been devised in which the resistance value of the pressure-sensitive conductive rubber is inversely calculated from the measurement of the voltage obtained from each column, but this method requires complicated calculations on a computer, so it takes time to process. There are some difficulties.

<Means for Solving the Problems> The present invention solves the problem of obtaining output even from grid points to which no pressure is applied when pressure-sensitive conductive rubber is sandwiched between striped electrodes to form a matrix-type contact sensor. Furthermore, the present invention provides a contact sensor that does not require complicated calculations by a computer without impairing the flexibility of rubber.

<Example> Now, let us consider a case in which pressure-sensitive conductive rubber is sandwiched between m striped electrodes on the upper surface side and n striped electrodes on the lower surface side to form an m×n matrix sensor as shown in Fig. 3. think. A voltage V is applied to one of the m electrodes on the upper surface side.

The output shall be read as a voltage by a constant grounding resistor attached to the lower electrode. The electrodes on the upper surface side are called rows, and the electrodes on the lower side are called columns.The resistance at the i-th row and j-th column lattice point of the pressure-sensitive conductive rubber sandwiched between the upper and lower electrodes is expressed as conductance _gij .
Similarly, let the grounding resistance be the conductance go. i
The output obtained from the j-th column electrode when V is applied to the j-th row electrode is expressed as V _ji . If the row electrodes other than those to which voltage is applied are grounded, the circuit consisting only of resistors shown in FIG. 3 can be rewritten as shown in FIG. 4. (In Figure 4, +V is applied to i and the remaining row electrodes are grounded.) By applying Kirchhoff's law in Figure 4, goV _ji = g _ij V- _n 〓 ^{k= 1} g _kj V _ji (1) Therefore, when voltage V is applied to the i-th row electrode, the voltage Vji obtained from the j-th column electrode is expressed by the following equation.

V _ji = g _ij V/(go+ _n 〓 ^k=1 g _kj )... (2) Looking at equation (2), the output voltage obtained from the output point is connected to the column electrode whose output is being measured. It can be seen that it is expressed by the total conductance and the conductance gij of the lattice point with the row electrode to which voltage is applied.

Pressure-sensitive conductive rubber changes its resistance when pressure is applied. FIG. 5 shows the relationship between the actually measured pressure and the resistance in the direction perpendicular to the surface. There is approximately r=P ^-s ... (3) There is a relationship. Therefore, if expressed as conductance, g= ^Ps ... (4).

When a load distributed within a plane is applied to a matrix sensor configured as shown in Figure 3, the conductance of each grid point changes depending on the load distribution, but in a steady state, the denominator of equation (2) is fixed.

Therefore, it is considered that the output voltage obtained from the j-th column when voltages are sequentially applied to the row electrodes 1 to m is determined by the conductance of the grid point.

From equation (4), the conductance of a lattice point increases as the applied pressure increases, so the output voltage from a lattice point to which a large pressure is applied is higher than the output from a lattice point to which a small pressure is applied. becomes.
Therefore, it is possible to compare the magnitude of force by comparing the magnitude of output. Also, V _ji (i=1~
m), m equations (2) corresponding to each V _ji can be found, g _ij (i=~m) can be inversely calculated, and the distribution of the force itself can be found from equation (4).

However, calculating all V _ji and performing the inverse operation requires the computer to perform complicated calculations, as described above, which is problematic in terms of processing time.

Furthermore, since each process must be processed through an AD converter, hardware complexity increases and restrictions arise.

When such a matrix sensor is used for the purpose of determining the shape of the applied pressure distribution, it is considered that it only needs to have the function of determining whether the pressure is suppressed above a certain level.

Therefore, as shown in FIG. 6, one row electrode is added, used as a reference electrode, and a certain resistance is connected between it and each column electrode. This resistance is selected with reference to FIG. 5 to be larger than the value corresponding to the pressure desired to be detected on the matrix sensor. First, a voltage is applied to the reference electrode and the other row electrodes are grounded. Measure the voltage V _jref obtained from each column electrode,
Remember it. V _jref is given by the following formula.

V _jref = g _ref V/(go+g _ref + _n 〓 ^k=1 g _ki )... (5) Next, V is applied to the row electrode i other than the reference electrode, and the other row electrodes are grounded. V _ji is
V _ji = g _ij V/(go+g _ref + _n 〓 ^k=1 g _ki )...(6). When comparing V _jref and V _ji , the magnitude depends on the magnitude of g _ij and g _ref . Lattice point (i,
If the force applied to j) is greater than P=s√g, then V _ji >V _jref .

Next, a block diagram of an actual circuit that realizes the above principle is shown in FIG. 7, and further explanation will be added. The figure shows an example in which the upper and lower surfaces of the pressure-sensitive conductive rubber are sandwiched between upper and lower electrodes to form a 3×3 matrix sensor. The number of row electrodes of the matrix sensor is increased by one for reference. These row electrodes are configured so that only one of them is connected to +V and the others are grounded according to a command from a computer using a scan circuit. This scan circuit can use a commercially available analog switch IC. First, +V is applied to the reference electrode (ref). The voltage appearing across the grounding resistor at this time is determined by equation (5) based on the pressure being applied to the pressure-sensitive rubber at that time and the reference resistor.

This voltage is temporarily stored using a sample and hold circuit. Then connect row electrode 1 to V,
Others are grounded. Assuming that the sample and hold circuit is controlled by the signal applied to the reference electrode line as shown in the figure, when +V is applied to row electrode 1, the voltage across the grounding resistor is applied to the sample and hold circuit. is not imported. The voltage at this time is given by equation (6). The reference voltage stored in the sample-and-hold circuit and this voltage are input to a comparison circuit to determine whether the voltage is large or small. If a load is applied to the corresponding grid point, the voltage at this point will be higher than the reference voltage, and an output will be obtained from the comparison circuit. The output is held in a register and input to the computer. If this is carried out sequentially with the row electrodes 2 and 3 and the data held in the register is taken in each time, the load distribution situation at each grid point of the 3.times.3 matrix can be known. FIG. 7 shows an example of a sample-and-hold circuit plus a comparison circuit configured using an operational amplifier. Also, dedicated sample and hold circuits and comparison circuit ICs.
It may be composed of

<Effects> The contact sensor using the pressure-sensitive conductive rubber described above has the following advantages.

1) By sandwiching pressure-sensitive conductive rubber between electrodes made of flexible printed circuit boards, for example, the advantage of rubber's flexibility can be utilized to the fullest. The resistors connected between the reference electrode and each column electrode do not need to be connected on the pressure-sensitive rubber matrix sensor, and may be placed on the circuit board that constitutes the processing circuit. Therefore, it does not become an obstacle that impairs the flexibility of the pressure-sensitive rubber.

This flexibility also makes it possible to attach it to a cylindrical robot arm, etc.

2) Since we have adopted a completely new method in which the voltage obtained through the reference resistor is temporarily stored and compared with the voltage obtained through the rubber resistance at other grid points, the grid points that do not apply pressure are This can be avoided except for the inconvenience of exhibiting the same behavior as the grid points being added.

Therefore, an output corresponding to the shape to which pressure is applied can be obtained, increasing reliability as a pressure-sensitive contact sensor.

3) Not only can it be used as a sensor to sense contact by attaching it to a robot hand or arm, but it can also be attached to a hand gripping part to determine the shape of a gripped object. It also has a wide range of applications, including as a sensor for automating processes such as product inspection, discrimination, and sorting in factories, and as an input device for information processing equipment such as computers.

[Brief explanation of drawings]

Fig. 1 is a perspective view showing the configuration of the pressure-sensitive rubber contact sensor, Fig. 2 is an explanatory diagram of the matrix in the same as above,
Fig. 3 is an explanatory diagram of a combination of resistors in a matrix, Fig. 4 is an equivalent circuit diagram of the same as above, Fig. 5 is a graph showing the relationship between pressure and resistance value of pressure-sensitive conductive rubber, and Fig. 6 is related to this invention. FIG. 7 is an explanatory diagram showing a ground sensor, FIG. 7 is a configuration diagram of the same as above, and FIG. 8 is a circuit diagram combining an operational amplifier circuit and a comparison circuit. 21, 22...Striped electrode, 23...Pressure-sensitive conductive rubber.

Claims (1)

[Claims] 1 A pressure-sensitive conductor that is formed into a planar shape and has an infinite resistance value in the direction perpendicular to the surface, but when localized pressure is applied perpendicular to the surface, the resistance value of that part decreases. The top and bottom sides of the conductive rubber are sandwiched between striped electrodes so that they are perpendicular to each other, and the pressure distribution status is determined by detecting the resistance change of the pressure-sensitive conductive rubber at the pole points of the upper and lower electrodes. In the pressure-sensitive sensor to be measured, one striped electrode on one side is used as a reference electrode, and a constant resistance is connected between each electrode on the other side.
The output voltage obtained through the reference resistor when voltage is applied to the reference electrode is temporarily stored as the reference voltage, and then the output power when voltage is applied to the second and subsequent electrodes of the one-side electrode is temporarily stored. A contact sensor characterized in that the in-plane distribution of force above a certain pressure is known by comparing it with a reference electrode and a comparator.