JPH04238267A - Acceleration sensor - Google Patents

Abstract

PURPOSE:To obtain an acceleration sensor with a simple structure and a wide frequency band of the detectable acceleration. CONSTITUTION:The coherent light outputted from a coherent light source 21 is inputted to a multi-mode light guide path 25 to concurrently propagate multiple modes, the light with the preset light intensity distribution is outputted from the multi-mode light guide path 25, and the light intensity distribution is changed in response to the displacement of a measured object 24 arranged in an optical path. The light intensity distribution is detected by a light detector array 26, thereby the position of the measured object is obtained by a neural network 27, and the acceleration applied to the measured object 24 is obtained from the relation between the position and time of the measured object 24 by an arithmetic section 28.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improvements in acceleration sensors.

0002

[Prior Art] Fig. 2 shows an example of a conventional acceleration sensor, and here it is shown in "Acceleration sensor using photoelastic effect" by Tai et al. )pp.
55) is shown. Light output from the optical transmitter 1 passes through an optical fiber 2, is collimated by a rod lens 3, and then input to a polarizer 4, where a specific linearly polarized light is extracted. The light output from the polarizer 4 is input to the photoelastic element 5. Here, if acceleration is applied in the direction of arrow 6, stress is applied to photoelastic element 5 by weight 7. The photoelastic element 5 is an element that exhibits birefringence due to stress, and when acceleration is applied, polarized light rotates in the photoelastic element 5. The light output from the photoelastic element 5 passes through a quarter-wave plate 8 and is input to an analyzer 9, where only a specific linearly polarized light is extracted. The light output from the analyzer 9 passes through the rod lens 10
The signal is input into the optical fiber 11 and guided to the optical receiver 12. Here, the combination of the quarter-wave plate 8 and the analyzer 9 converts the polarization rotation angle due to the stress into the light intensity that reaches the optical receiver 12. Therefore, the polarization rotation angle is determined from the light intensity, and the stress is obtained. Since stress is given by the product of weight and acceleration, if the weight of the weight 7 is known, the acceleration can be found.

0003

[Problems to be Solved by the Invention] However, the above-mentioned acceleration sensor requires a large number of optical parts such as a polarizer, a photoelastic element, a quarter-wave plate, an analyzer, etc., and there is a problem that the configuration becomes complicated. Also, in order to increase the sensitivity, it is possible to increase the weight of the weight, but on the other hand, if the weight is increased too much, it will not be able to follow changes in stress, and the frequency band in which it can operate, that is, detect acceleration, will be narrow (500 Hz). There was a problem that the degree of

SUMMARY OF THE INVENTION In view of the above-described conventional problems, it is an object of the present invention to provide an acceleration sensor that has a simple configuration and can widen the frequency band of detectable acceleration.

[0005]

[Means for Solving the Problems] In order to achieve the above object, the present invention provides a coherent light source that generates coherent light, a multimode optical waveguide that can propagate a plurality of modes simultaneously, a photodetector array that detects the light intensity distribution of light; and a photodetector array that inputs the light output from the coherent light source to one end of the multimode optical waveguide, and inputs the light output from the other end of the multimode optical waveguide to the optical an optical system that guides the light to the detector array; an object to be measured that is disposed between the coherent light source and one end of the multimode optical waveguide and is displaced in response to acceleration applied from the outside; and an object that is detected by the photodetector array. Acceleration comprising: a neural network for determining the position of the object to be measured from the light intensity distribution; and a calculation means for calculating the acceleration applied to the object from the relationship between the position of the object to be measured and time determined by the neural network. In addition, as a second aspect of the sensor, an optical fiber is used for at least a part of an optical system that inputs light output from a coherent light source into one end of a multimode optical waveguide, and an end of the optical fiber on the side of the multimode optical waveguide. The acceleration sensor according to claim 1, wherein the part is a free end and the object to be measured is attached, and the object to be measured is disposed near one end of the multimode optical waveguide, and the acceleration sensor is provided with an The present invention proposes an acceleration sensor according to claim 1, further comprising an optical system that inputs light from the other end of the multimode optical waveguide to the other end of the multimode optical waveguide and guides the light output from the other end of the multimode optical waveguide to a photodetector array.

[0006]

According to claim 1 of the present invention, the output light from the coherent light source is input to the multimode optical waveguide via the optical system, and propagates while exciting a large number of modes. When acceleration is applied to the object to be measured and its position is displaced, the excitation conditions of the multimode optical waveguide change and the mode distribution changes, but the output light is guided to a photodetector array and the light intensity distribution is detected. . The light intensity distribution is processed by a neural network to determine the position of the object to be measured, and further the acceleration applied to the object to be measured is determined by the calculation means.

According to claim 2, when acceleration is applied to the object to be measured and its position is displaced, the position of the end of the optical fiber on the multimode optical waveguide side changes, thereby changing the position of the multimode optical waveguide. Excitation conditions change.

According to claim 3, the output light of the coherent light source is input to the other end of the multimode optical waveguide via the optical system, excites and propagates a large number of modes, and is output from one end. . The output light is reflected by the object to be measured, inputted again into one end of the multimode optical waveguide, propagated through this, and inputted into the photodetector array from the other end via the optical system.

[0009]

[Embodiment] FIG. 1 shows a first embodiment of the acceleration sensor of the present invention. In the figure, 21 is a coherent light source, 22,
23 is a lens, 24 is an object to be measured, 25 is a multimode optical waveguide, 26 is a photodetector array, 27 is a neural network, 28 is a calculation section, and 29 is a clock generation section.

The coherent light source 21 is composed of, for example, a semiconductor laser, and generates coherent light. lens 2
2 couples the output light of the coherent light source 21 to the multimode optical waveguide 25, and the lens 23 couples the output light of the coherent light source 21 to the multimode optical waveguide 25.
The output light of 5 is coupled to a photodetector array 26. The object to be measured 24 is made of any solid material having a size that does not block all of the light focused by the lens 22. The multimode optical waveguide 25 transmits multiple modes simultaneously. The photodetector array 26 is formed by arranging a large number of photodetecting elements in a matrix, and detects the light intensity distribution of incident light and converts it into an electrical signal. The neural network 27 is configured to determine the position of the object to be measured 24 from the light intensity distribution detected by the photodetector array 26. The calculation section 28 calculates acceleration (including direction) from the position information from the neural network 26 and the time information from the clock generation section 29.

FIG. 3 shows a method of holding the object to be measured 24. As shown in FIG. 3(a), two movable parts 33 and 33 made of springs, rubber, etc. are placed between two fixed parts 31 and 32, respectively. 34 or as shown in FIG. 3(b)
As shown in FIG. 3, it is held cantilevered by a fixed part 35 via a movable part 36, and can be displaced in any direction.

Next, the operation of the device will be explained.

Light output from the coherent light source 21 is input into a multimode optical waveguide 25 through a lens 22 . Several modes are excited in the multimode optical waveguide 25, and a complex interference pattern (speckle pattern) is output at the output end face. Here, if the object to be measured 24 is displaced, the excitation conditions of the multimode optical waveguide 25 change, the excited mode distribution changes, and the amplitude and phase of the light output from the multimode optical waveguide 25 change. changes, and the speckle pattern, that is, the light intensity distribution also changes.

Therefore, an optical signal in the form of converting the displacement of the object to be measured 24 into light is outputted from the multimode optical waveguide 25 and inputted to the photodetector array 26 through the lens 23. Photodetector array 26 performs optical-to-electrical conversion, and electrical signals are input to neural network 27 .

Here, since the relationship between the light intensity distribution and the position of the object to be measured 24 cannot be expressed by a simple equation, it takes a lot of time to calculate the position. Therefore, in the present invention, a neural network is used to determine the position of the object to be measured 24.

[0016] As is well known, neural networks are
By appropriately setting internal parameters, the relationship between the input signal to the neural network and the output signal from the neural network can be freely set. Also, as seen in functions such as association and recognition, neural networks can make the output signal unchanged in response to slight changes in the input signal, and conversely, can be made sensitive to the input signal. It is also possible to vary the output signal. Furthermore, as seen in the learning function, by exemplifying several combinations of input and output signals to the neural network, internal parameters can be automatically changed to obtain the desired input-output relationship. You can also. Therefore, the position of the object to be measured 24 can be determined from the output light intensity distribution of the multimode optical waveguide 25.

The neural network 27 determines the position of the object to be measured 24 from the input electrical signal. Further, position information is sampled at the timing of the clock generated by the clock generator 29 and input to the arithmetic unit 28 . At this time, if the amount of change in position at two certain times, for example t1 and t2, is known, the average speed between those two times can be found. Also, two times, for example, time t
The time T1 between 1 and t2 and the time t2 and t
3. If the change in speed during the time T2 is known, the average acceleration during those two times can be determined. In this way, if position information at three times is given, acceleration can be determined.

[0018] The speed of neural networks can be increased by using high-speed elements. For example, as a neural network LSI that operates at high speed, Koga
There is one described in another book, ``High-speed analog neural network LSI for neuroprocessing type optical demultiplexer'' (IEICE Neurocomputing Study Group, November 1990), which operates at 150 MHz.

FIG. 4 shows a second embodiment of the present invention, in which the object to be measured 24 is held using an optical fiber. That is, in the figure, 40 is a multimode optical fiber, which is arranged between the lens 22 and the multimode optical waveguide 25. Multimode optical waveguide 25 of the optical fiber 40
The side is fixed to a fixed part 41 at a portion slightly apart from the end, and the object to be measured 24 is fixedly attached to the free end. In the above configuration, light from the coherent light source 21 is input to one end of the optical fiber 40 by the lens 22, and light from the other end of the optical fiber 40 to which the object to be measured 24 is attached is directly input to the multimode optical waveguide 25. be done. Due to the temporal displacement of the object to be measured 24, the optical fiber 40 is also displaced at the same time, and the excitation conditions of the multimode optical waveguide 25 change, and the multimode optical waveguide 2
The output light intensity distribution from 5 changes. Note that the other configurations and operations are the same as in the first embodiment.

FIG. 5 shows a third embodiment of the present invention, in which the object to be measured 24 is placed at the end of the apparatus. That is, in the figure, 42 is an optical directional coupler, which is arranged between the lens 23 and the photodetector array 26. Further, the coherent light source 21 and the lens 22 are arranged on one side of the optical directional coupler 42. In the above configuration, the output light from the coherent light source 21 passes through the lens 22, the optical directional coupler 42, and the lens 23, and then enters the multimode optical waveguide 25. The output light from the multimode optical waveguide 25 is reflected by the object to be measured 24 and input into the multimode optical waveguide 25 again. The output light from the multimode optical waveguide 25 passes through the lens 23 and the optical directional coupler 42, and then enters the photodetector array 26. Note that the other configurations and operations are the same as in the first embodiment. Furthermore, according to this embodiment, the light source section and the detection/processing section can be integrated, allowing for further miniaturization.

FIG. 6 shows a fourth embodiment of the present invention, in which another example is shown in which the object to be measured 24 is placed at the end of the apparatus. That is, in the figure, 43 is an optical circulator, which is provided with four input/output ports 43-1 to 43-4. The light input to the input/output port 43-1 is output from the input/output port 43-2;
The light input to the input/output port 43-3 is outputted from the input/output port 43-3. In the above configuration, the coherent light source 2
The output light from 1 is transmitted through a lens 22, an optical circulator 43,
After passing through the lens 23, it is input into the multimode optical waveguide 25. The output light from the multimode optical waveguide 25 is reflected by the object to be measured 24 and input into the multimode optical waveguide 25 again. The output light from the multimode optical waveguide 25 passes through the lens 23 and the optical circulator 43, and then enters the photodetector array 26. Note that the other configurations and operations are the same as in the first embodiment. Further, according to this embodiment, the light source section and the detection/processing section can be integrated as in the third embodiment, allowing for further miniaturization.

[0022]

As explained above, according to claim 1 of the present invention, conventional polarizers, photoelastic elements, quarter-wave plates, analyzers, etc. are not required, and the configuration is simplified. In order to take advantage of the fact that the light intensity distribution of light output from a multimode optical waveguide changes in response to slight changes in the light input to the multimode optical waveguide, the object to be measured is transferred to the multimode optical waveguide. It only needs to have a size that can slightly change the input light, so it is possible to use a light object to be measured, and a neural network can be used to determine the position of the object from the light intensity distribution. Since acceleration is determined, the frequency band of detectable acceleration can be widened without reducing sensitivity.

Further, according to claim 2 of the present invention, the object to be measured can be held using an optical fiber that is a part of the optical system, and the configuration can be simplified.

Furthermore, according to claim 3 of the present invention, the light source section and the detection/processing section can be integrated, making it possible to further reduce the size.

[Brief explanation of the drawing]

FIG. 1 A configuration diagram showing a first embodiment of an acceleration sensor of the present invention.

[Figure 2] Block diagram showing an example of a conventional acceleration sensor

[
Figure 3: Configuration diagram showing how to hold the object to be measured

[Figure 4]
A configuration diagram showing a second embodiment of the acceleration sensor of the present invention

[Fig. 5] A configuration diagram showing a third embodiment of the acceleration sensor of the present invention.

[Fig. 6] A configuration diagram showing a fourth embodiment of the acceleration sensor of the present invention.

[Explanation of symbols]

21... Coherent light source, 22, 23... Lens, 24...
Measured object, 25... Multimode optical waveguide, 26... Photodetector array, 27... Neural network, 28... Arithmetic unit, 29... Clock generator, 40... Optical fiber, 42... Optical directional coupler, 43... Light circulator.

Claims (3)

[Claims] 1. A coherent light source that generates coherent light, a multimode optical waveguide capable of simultaneously propagating a plurality of modes, a photodetector array that detects a light intensity distribution of incident light, and an output from the coherent light source. an optical system that inputs light to one end of the multimode optical waveguide and guides light output from the other end of the multimode optical waveguide to the photodetector array; a neural network that determines the position of the object from the light intensity distribution detected by the photodetector array; An acceleration sensor comprising: calculation means for calculating acceleration applied to the object to be measured from the relationship between the position of the object to be measured and time determined by . 2. An optical fiber is used as at least a part of the optical system that inputs the light output from the coherent light source into one end of the multimode optical waveguide, and the end of the optical fiber on the multimode optical waveguide side is a free end. 2. The acceleration sensor according to claim 1, further comprising: an object to be measured; 3. The object to be measured is placed near one end of the multimode optical waveguide, and the light output from the coherent light source is input to the other end of the multimode optical waveguide. The acceleration sensor according to claim 1, further comprising an optical system that guides the output light to a photodetector array.