JP2023158207A - optical equipment - Google Patents

Abstract

An object of the present invention is to suppress a decrease in output accuracy of a laser radar device.
A map data structure 110 of map data stored in a map data storage unit 11 includes a matched filter 6 that performs noise reduction processing on an output signal of a light receiving unit 3 of a lidar 1 that includes a light emitting unit 2 and a light receiving unit 3. A map data structure 110 of map data used in the map data structure 110, in which waveform information 112 of a reference signal used in the matched filter 6 is set for each predetermined area 111, and the matched filter 6 is configured based on the waveform information 112 of the reference signal. Process the output signal.
[Selection diagram] Figure 1

Description

TECHNICAL FIELD The present invention relates to optical equipment.

Conventionally, laser radar devices (LiDAR; also called Light Detection and Ranging) have been known, which irradiate a detection space with a pulse of laser light and detect an object within the detection space based on the level of the reflected light. ing. Among such laser radar devices, there is one that performs filtering on the output signal of the light receiving section to improve the signal-to-noise ratio (signal-to-noise ratio). For example, Patent Document 1 describes the use of a matched filter for the output signal of a light receiving section in a laser radar device.

When using the matched filter described in Patent Document 1, a known impulse response to be provided to the matched filter is prepared in advance, and the output signal of the light receiving section is filtered using the prepared impulse response. The problem arises that the output accuracy of the laser radar device decreases as a result of not being able to cope with individual differences among individuals, changes over time, and environmental changes such as temperature.

In order to deal with such problems, Patent Document 2 includes a reflector that reflects the emitted light, and when the APD receives the return light reflected by the reflector, the returned light is transmitted to the APD using a matched filter based on the output signal of the APD. The reference received pulse to be used is estimated.

Japanese Patent Application Publication No. 2007-225318 JP 2018-9831 Publication

However, in the method described in Patent Document 2, the reference received pulse (also referred to as a reference signal) is estimated without considering the characteristics of the object to which light is actually irradiated, so the shape of the object, the irradiation angle, etc. In some cases, the reference signal cannot be accurately estimated, and the output accuracy of the laser radar device is reduced.

An example of the problem to be solved by the present invention is to suppress a decrease in output accuracy of a laser radar device.

In order to solve the above problem, the invention according to claim 1 includes: an irradiation section that irradiates light; a light reception section that receives the return light of the light reflected by an object; and an output signal of the light reception section. It includes a signal processing unit that performs processing related to improving the SN ratio, and a storage unit that stores map data used in the signal processing unit, and the map data includes a reference signal used in the signal processing unit. Waveform information is set for each predetermined area on the map, and the signal processing section processes the output signal based on the waveform information of the reference signal.

1 is a schematic configuration diagram of a lidar equipped with map data having a map data structure according to a first embodiment of the present invention; FIG. FIG. 2 is a system configuration diagram when the rider shown in FIG. 1 downloads map data from a server device. FIG. 2 is a configuration diagram of the matched filter shown in FIG. 1. FIG. 2 is an example of a map data structure of map data stored in the map data storage section shown in FIG. 1. FIG. This is a modification of FIG. 4. This is a modification of FIG. 3. 7 is an example of filter processing in the low-pass filter shown in FIG. 6. This is another example of the reference signal waveform. It is a map data structure according to a second example of the present invention.

Hereinafter, a map data structure according to an embodiment of the present invention will be explained. A map data structure according to an embodiment of the present invention improves the signal-to-noise ratio of an output signal of a light receiving section of an optical device that includes an irradiating section that irradiates light and a light receiving section that receives return light of the light reflected by an object. A map data structure of map data used in a signal processing unit that performs processing, in which information regarding a reference signal used by the signal processing unit is set for each predetermined area on a map indicated by the map data, and information regarding the reference signal is Based on this, the signal processing unit processes the output signal. By doing this, information regarding the reference signal can be set in the map data in advance, so that a reference signal that takes into account the characteristics of the object to be irradiated with light can be selected and used. Furthermore, since the information regarding the reference signal is set for each predetermined area, the information regarding the reference signal can be set according to the characteristics of the area to which light is irradiated, such as a city with many buildings, a field, or a forest.

Further, the information regarding the reference signal may be set for each cell that divides the space indicated by the map into predetermined sizes. By doing this, the city, fields, etc. can be further subdivided into cells. Therefore, more detailed information regarding the reference signal can be set.

In addition, when the space indicated by the map is divided into cells of a predetermined size, multiple cells are treated as one area according to the similarity of the characteristics of each cell with respect to sensors such as lidar, and the number of cells is divided into one area. Information regarding the reference signal may also be set. For example, when a road surface is divided into cells of a predetermined size and subdivided into multiple cells, the same reference signal may become the appropriate reference signal because the cells including the division lines have similar reflection characteristics. be. In such a case, detailed information regarding the reference signal can be set while reducing the amount of information.

Further, the information regarding the reference signal may be set for each weather. By doing this, it is possible to select and use an appropriate reference signal even when the reflection characteristics of an area change depending on the weather or when objects with different reflection characteristics appear or disappear. . For example, changes in temperature, humidity, etc. change the reflection characteristics of the area, and the returned light and the background light incident on the optical device together with the returned light change. It is possible to respond to such changes.

Further, the information regarding the reference signal may be set for each season. By doing this, it is possible to take into account the effects of changes in the state of the area depending on the season. For example, in winter, in areas where snow accumulates, the reflection characteristics change as the snow accumulates. Alternatively, the reflective characteristics of street trees change depending on the season, such as the presence or absence of leaves. Further, for example, the reflection characteristics of the area change due to changes in temperature, humidity, etc. due to seasonal changes. Due to such changes in reflection characteristics, returned light and background light change. It becomes possible to respond to such changes.

Further, the information regarding the reference signal may be set for each time period. By doing this, the background light that enters the optical device along with the returning light may change depending on the time of day, the reflection characteristics of the area may change, or objects with different reflection characteristics may appear or disappear. Even in such a case, an appropriate reference signal can be selected and used. For example, during the day and night, background light changes due to changes in light sources such as the sun and street lights, and changes in temperature and humidity due to changes in time of day change the reflection characteristics of an area, causing changes in return light and background light. do. It becomes possible to respond to such changes.

Further, the storage medium according to an embodiment of the present invention stores map data having the map data structure described above. By doing so, map data having a map data structure can be stored and distributed in a hard disk, optical disk, memory card, or the like.

Furthermore, the storage device according to an embodiment of the present invention stores map data having the map data structure described above. By doing so, map data having a map data structure can be stored and distributed, for example, in a server device or the like.

An optical device according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 8. FIG. 1 is a functional configuration diagram of a lidar as an optical device according to this embodiment. The lidar 1 includes a light emitting section 2, a light receiving section 3, an optical system 4, an AD converter (ADC) 5, a matched filter 6, and a control section 8. The lidar 1 generates information about the object 100 by emitting a light pulse (hereinafter referred to as irradiation light) and receiving a light pulse reflected by an external object 100 as a target (hereinafter referred to as return light). It is a laser radar device, also called LiDAR.

The light emitting unit 2 outputs light pulses in response to a trigger signal etc. from the control unit 8. The light emitting section 2 includes a laser diode (LD), a driver circuit for driving the LD, and the like.

The light receiving unit 3 outputs a voltage signal (light receiving signal) proportional to the intensity of the return light from the object 100. Generally, a photodetecting element such as a photodiode outputs a current, so the receiving section converts this current into a voltage (I/V conversion) and outputs it. The light receiving section 3 includes an APD (avalanche photodiode) as a photodetecting element, the above-mentioned I/V conversion circuit, and the like.

The optical system 4 irradiates the light pulses input from the light emitting unit 2 as irradiation light while changing the irradiation direction, and also irradiates the light pulses inputted from the light emitting unit 2 as irradiation light while changing the irradiation direction. Light is guided to the light receiving section 3. The optical system 4 includes optical members such as a rotating mirror, a collimator lens, and a condensing lens. That is, the light emitting section 2 and the optical system 4 function as an irradiating section that irradiates light while changing the irradiation direction, and the light receiving section 3 and the optical system 4 function as an irradiating section that emits light while changing the irradiation direction, and the light receiving section 3 and the optical system 4 function as an irradiating section that emits light while changing the irradiation direction. It functions as a light receiving section that receives the returned light. In this embodiment, a lidar that emits light while changing the irradiation direction will be described, but a flash lidar or the like that does not change the irradiation direction may be used.

The AD converter 5 is a well-known circuit that converts an analog signal (light reception signal) output from the light receiving section 3 into a digital signal.

The matched filter 6 is a well-known filter that has a pass characteristic in which the pass rate is 1 for a signal having the same waveform as the pulse waveform to be transmitted and received, and is also called a matched filter. The matched filter 6 convolves the output (light reception signal) of the AD converter 5 with a reference signal, which will be described later, as a predetermined impulse response, and performs filtering to improve the SN ratio of the light reception signal. This SN ratio improvement processing refers to processing that maximizes the SN ratio, for example. In other words, the matched filter 6 constitutes the signal processing device according to this embodiment.

The control unit 8 is composed of, for example, a microcomputer having a CPU (Central Processing Unit), memory, etc., or dedicated hardware, and manages the overall control of the rider 1. The control unit 8 sends a trigger signal to the light emitting unit 2 to instruct it to output a light pulse. Further, the control unit 8 calculates the distance from the rider 1 to the object 100 based on the sending timing of the trigger signal and the timing of the output signal from the matched filter 6.

The map data storage unit 11 is a memory or the like in which map data of a range such as a route traveled by a vehicle on which the rider 1 is mounted is stored. The map data stored in the map data storage unit 11 may be map data for the entire country stored in advance, or as shown in FIG. It may also be downloaded via the network N. Furthermore, map data may be updated from the server device 10 without being limited to the necessary range.

As is well known, the GPS receiver 12 periodically receives radio waves transmitted from a plurality of GPS satellites to detect current position information and time.

Next, details of the matched filter 6 having the above-described configuration will be explained with reference to FIG. 3. The matched filter 6 includes a filter 6a and a control section 6b. Filter 6a is the matched filter described above.

The control unit 6b determines the area (area) in the irradiation direction of the lidar 1 from the map data stored in the map data storage unit 11 based on the current position detected by the GPS receiver 12 and the movement direction obtained from the movement trajectory. Obtain information about the reference signal of. Then, a reference signal is generated based on the information regarding the obtained reference signal and is supplied to the filter 6a.

Here, an example of the data structure of the map data stored in the map data storage section 11 will be explained with reference to FIG. 4. In the map data structure 110 of FIG. 4, reference signal waveform information 112 is set for each area 111 in the map data. Here, as shown in FIG. 4(a), the area 111 indicates the type of area (land) such as an urban area or a vegetation area (of course, it is not limited to the two types shown in FIG. 4). . Note that the area 111 may be a unique area, such as a place name, rather than a type of area. For example, it may be set in units of Ginza, Kushiro Marsh, etc. In this case, it is preferable that the map data structure 110 also include location information indicating the area.

Further, as shown in FIG. 4(b), the area 111 may be divided according to the size of return light or background light. The example in FIG. 4B shows a region with high return light and low background light, and a region with low return light and high background light. The high return light, low background light region indicates, for example, a region where black objects are dominant and there is little return light, making distance detection difficult.The low return light, high background light region indicates a region where the output signal of the light receiving section is saturated. shows. In this case, it is preferable that the magnitude of the returned light be determined based on properties such as reflection characteristics of each area 111.

That is, waveform information 112 of the reference signal as information regarding the reference signal is set for each area 111 as a predetermined region on the map indicated by the map data.

Moreover, the waveform information 112 of the reference signal is set to, for example, the waveform information of the reference signal having the best SN ratio as a result of processing in advance by the filter 6a using a plurality of types of reference signals. For example, in the city, a plurality of types of reference signals are processed by the filter 6a in advance, and the waveform information of the reference signal output from the filter 6a with the best S/N ratio is set as the waveform information 112 of the reference signal of area A.

In FIG. 4, the waveform information 112 of the reference signal is set as information regarding the reference signal, but reflection characteristics and the like may be set for each area. Then, the control section 6b may generate a reference signal based on the reflection characteristics and the like.

Further, the area 111 may be a cell (grid) divided by a predetermined size instead of the type of area as shown in FIG. 4. Further, this cell may be a voxel divided also in the height direction. By doing so, the urban area, vegetation area, etc. can be further subdivided into cells. Therefore, more detailed information regarding the reference signal can be set.

In addition, when subdividing into cells or voxels in this way, if multiple cells that use a common reference signal are adjacent to each other, the amount of data can be reduced by considering those cells as one area. be able to.

Furthermore, as shown in FIG. 5A, the waveform information 112 of the reference signal may be set for each weather in response to changes in background light depending on the weather, such as sunny, cloudy, and rainy weather.

Further, the waveform information 112 of the reference signal may be set for each season such as spring, summer, fall, and winter, as shown in FIG. 5(b). For example, since the light reflection characteristics change in areas covered with snow, it is preferable to change the information regarding the reference signal in winter compared to other seasons in an area indicating a snowfall region. Further, in an area where there are street trees or plants, the light reflection characteristics change depending on the season as the presence or absence of leaves changes, so it is preferable to change the information regarding the reference signal depending on the season. Alternatively, in a region representing a rice field, the light reflection characteristics change depending on whether or not water is filled and the growth of the rice, so it is preferable to change the information regarding the reference signal depending on the season. In this way, in areas where the reflection characteristics change depending on the season, it is preferable to set the waveform information 112 of the reference signal for each season.

Further, the waveform information 112 of the reference signal may be set for each time period such as morning, noon, and night, as shown in FIG. 5(c). In this case as well, changes in reflection characteristics etc. can be accommodated. Note that the time is not limited to morning, noon, or night, and a specific time such as 6:00 a.m. to 10:00 a.m. may be specified.

The control unit 6b of the matched filter 6 identifies the area where the lidar 1 is irradiating the light pulse based on the current position information and traveling direction detected by the GPS receiver 12, and sets standards according to the area. Signal waveform information is read from map data having the map data structure 110 shown in FIG. 4 and the like.

Then, the control unit 6b generates a reference signal based on the read waveform information and supplies it to the filter 6a. The filter 6a to which the reference signal is supplied filters the received light signal based on the reference signal and outputs the filtered signal to the control unit 8.

Further, the configuration of the matched filter 6 is not limited to that shown in FIG. 3. The matched filter 6 may include a plurality of filters each having characteristics based on different reference signals. In this case, the waveform information 112 of the map data structure 110 stores selection information indicating which of a plurality of filters to select. Then, the matched filter 6 selects one of the plurality of filters based on the selection information and performs the signal-to-noise ratio improvement process.

Further, for example, as shown in FIG. 6, the matched filter 6 may include a low-pass filter (LPF) 6c that generates a reference signal and outputs it to the filter 6a.

In this case, the filter characteristics of the low-pass filter 6c are set for each area as information regarding the reference signal, and the control unit 6b uses the current position detected by the GPS receiver 12 and the moving direction obtained from the moving trajectory. By acquiring the filter characteristics of the low-pass filter 6c and setting the acquired characteristics in the low-pass filter 6c, a reference signal according to the area may be given to the filter.

The low-pass filter 6c includes a plurality of filters (for example, Gaussian filters) each having different characteristics with respect to an ideal reference signal. The low-pass filter 7 outputs the filtered result to the matched filter 6 as a reference signal. An ideal reference signal is, for example, when the light receiving section 3 receives the return light that is reflected from a reflector with known reflection characteristics placed in an environment where the irradiation light output from the light emitting section 2 is not affected by external light. This is an output signal of the AD converter 5. This ideal reference signal may be supplied from a memory that stores the signal waveform of the signal, or may be supplied from an oscillation circuit that generates the signal.

The low-pass filter 6c is, for example, a Gaussian filter whose standard deviation σ is a predetermined real number a (a×1), a Gaussian filter whose standard deviation σ is a predetermined real number a×16, or a Gaussian filter whose standard deviation σ is a predetermined real number a×64. The filter includes a Gaussian filter whose standard deviation σ is a predetermined real number a×128. Note that the reference signal is not limited to the Gaussian filter, and may be generated using other filtering methods. That is, the reference signals are generated by being processed differently based on the irradiation light emitted by the light emitting section (irradiation section).

FIG. 7 shows an example of filtering results by the low-pass filter 6c. The left side of the figure is a light emission signal indicating irradiation light, and the right side shows the result of filtering the light emission signal using a low-pass filter. The filtering results are, from top to bottom, a Gaussian filter with a standard deviation σ of a predetermined real number a (a×1), a Gaussian filter with a standard deviation σ of a predetermined real number a×16, and a Gaussian filter with a standard deviation σ of a predetermined real number a×64. , the standard deviation σ is a Gaussian filter with a predetermined real number a×128. Furthermore, the function for generating the reference signal is not limited to one that shows a symmetrical distribution as shown in FIG. 7, but may be one that shows an asymmetrical distribution as shown in FIG. 8, for example.

In the above description, the low-pass filter 6c generates the reference signal based on a predetermined fixed light emission signal, but it may also generate the reference signal based on the light emission signal used by the light emitting unit 2.

According to this embodiment, the map data structure 110 of the map data stored in the map data storage section 11 performs processing for improving the SN ratio of the output signal of the light receiving section 3 of the lidar 1, which includes the light emitting section 2 and the light receiving section 3. A map data structure 110 of map data used in the matched filter 6 performed, in which waveform information 112 of a reference signal used in the matched filter 6 is set for each predetermined area 111, and based on the waveform information 112 of the reference signal. A matched filter 6 processes the output signal. By doing this, the waveform information 112 of the reference signal can be set in the map data in advance, so that a reference signal that takes into consideration the characteristics of the object 100 can be selected and used. Further, since the waveform information 112 of the reference signal is set for each predetermined area 111, the information regarding the reference signal can be set depending on the characteristics of the area to be irradiated with light, such as an urban area with many buildings or a vegetation area. I can do it.

Furthermore, when the area 111 is divided into cells or voxels, a plurality of cells or voxels that use a common reference signal may be treated as one area, and information regarding the reference signal may be set. By doing so, it is possible to set information regarding the reference signal according to the characteristics of the area while reducing the amount of information.

Further, the waveform information of the reference signal may be set for each weather. By doing so, even if the reflection characteristics of the area change depending on the weather, an appropriate reference signal can be selected and used.

Further, the waveform information of the reference signal may be set for each season. By doing this, it is possible to take into account the effects of changes in the state of the area depending on the season.

Further, the waveform information of the reference signal may be set for each time period. By doing so, it is possible to take into account the influence of changes in the state of the area depending on the time period.

Additionally, map data having a map data structure 110 is stored in the server device 10 . By doing so, map data having the map data structure 110 can be stored in an external device such as the server device 10 and distributed.

Note that the map data having the map data structure 110 may be stored in a storage medium such as an optical disk or a memory card.

Next, a map data structure according to a second embodiment of the present invention will be explained with reference to FIG. Note that the same parts as those in the first embodiment described above are given the same reference numerals, and the description thereof will be omitted.

In this embodiment, the configuration of the rider 1 is the same as that in the first embodiment. An example of the data structure of map data according to this embodiment will be explained with reference to FIG. In the map data structure 110A of FIG. 9, waveform information 112 of a reference signal is set in the map data for each object 113 indicating a feature. In other words, as shown in FIG. 9, reference signal waveform information 112 is set for each object such as a building, a street tree, a road, etc. (Of course, the waveform information 112 is not limited to the three types shown in FIG. ).

In addition, in FIG. 9, the object is described by the type of feature, but it may also be a specific name, such as ○○ building, for example. In that case, it is preferable that the location (latitude, longitude) of the feature is also included in the map data structure 110A.

The control unit 6b of the matched filter 6 identifies the object to which the lidar 1 is irradiating the light pulse based on the current position information detected by the GPS receiver 12, and generates waveform information of the reference signal corresponding to the object. is read from the map data having the map data structure 110A shown in FIG.

Then, the control unit 6b generates a reference signal based on the read waveform information and supplies it to the filter 6a. The filter 6a to which the reference signal is supplied filters the received light signal based on the reference signal and outputs the filtered signal to the control unit 8.

Note that the unit of object is not limited to the unit of features such as buildings and roads. For example, since the reflection characteristics of light differ depending on the walls and windows of a building, or the presence or absence of road markings such as white lines on the road, it is also possible to subdivide the components of a feature as an object.

Also in this embodiment, objects may be divided by the magnitude of return light or background light. In this case, the magnitude of the returned light is preferably determined based on the properties of each object, such as its reflection characteristics.

Further, the information regarding the reference signal set for each object may be set according to the similarity of the characteristics of the object to the rider 1. This is because the same reference signal can be used even if different objects have similar reflection characteristics, so it is possible to reduce flag information by assigning a common flag or the like. Therefore, it is possible to set information regarding the reference signal according to the characteristics of the object while reducing the amount of information.

Also in this embodiment, the information regarding the reference signal may be set for each weather, each season, or each time period.

Also in this embodiment, map data having the map data structure 110A may be stored in the server device 10 and distributed. Furthermore, the map data having the map data structure 110A may be stored in a storage medium such as an optical disk or a memory card.

According to this embodiment, the map data structure 110A of the map data stored in the map data storage section 11 performs processing for improving the SN ratio of the output signal of the light receiving section 3 of the lidar 1, which includes the light emitting section 2 and the light receiving section 3. In the map data structure 110A of map data used in the matched filter 6 performed, waveform information 112 of a reference signal used in the matched filter 6 is set for each predetermined object 113, and based on the waveform information 112 of the reference signal. A matched filter 6 processes the output signal. By doing this, the waveform information 112 of the reference signal can be set in the map data in advance, so that a reference signal that takes into consideration the characteristics of the object 100 can be selected and used. Further, since the waveform information 112 of the reference signal is set for each object 113, the waveform information 112 of the reference signal can be set according to the characteristics of a feature such as a building, tree, or road that irradiates light. be able to.

Further, the present invention is not limited to the above embodiments. That is, those skilled in the art can implement various modifications based on conventionally known knowledge without departing from the gist of the present invention. Of course, such modifications fall within the scope of the present invention as long as they still have the map data structure of the present invention.

1 Lidar (optical equipment)
2 Light emitting part (irradiation part)
3 Light receiving section 6 Matched filter (signal processing section)
6a filter 6b control unit 6c low-pass filter 8 control unit 11 map data storage unit 12 GPS receiver 100 object (target object)
110, 110A Map data structure 111 Area 112 Waveform information of reference signal (information regarding reference signal)
113 Object

Claims (6)

an irradiation unit that irradiates light, and a light reception unit that receives return light of the light reflected from a target object;
a signal processing unit that performs processing related to improving the SN ratio of the output signal of the light receiving unit;
a storage unit storing map data used in the signal processing unit;
In the map data, waveform information of a reference signal used by the signal processing unit is set for each predetermined area on the map,
The signal processing unit processes the output signal based on waveform information of the reference signal.
An optical device characterized by: 2. The optical device according to claim 1, wherein the waveform information of the reference signal is set for each cell that divides the space indicated by the map into predetermined sizes. 3. The optical device according to claim 2, wherein the waveform information of the reference signal is set with a plurality of cells as one region according to similarity of characteristics of each cell with respect to a sensor of the optical device. . 3. The optical device according to claim 1, wherein the waveform information of the reference signal is set for each weather. 3. The optical device according to claim 1, wherein the waveform information of the reference signal is set for each season. 3. The optical device according to claim 1, wherein the waveform information of the reference signal is set for each time period.

Images