JP6477083B2 - Optical distance measuring device - Google Patents

Description

  The present invention relates to an optical distance measuring device, and more particularly to an optical distance measuring device mounted on a moving body such as a vehicle.

  There is a laser radar as one of sensors for the environment outside the vehicle for use in a safety system such as a collision prevention system. The laser radar is an optical distance measuring sensor (optical distance measuring device) using TOF (Time of Flight), and light irradiated from the sensor is reflected by the measurement object and returns to the optical distance measuring device. This is a sensor that measures the distance between the optical distance measuring device and the measurement object based on the time difference up to, that is, the TOF.

  In an optical distance measuring device, a large number of measurement points (point groups) are generally obtained by scanning irradiation light in a measurement object space or arraying a light receiving unit that receives reflected light. The road surface and obstacles are recognized by analyzing the spatial distribution of the point cloud.

  In order for the safety system to determine whether it is necessary to brake or avoid the moving body, it is important to determine whether the measurement point is a road surface. If the measurement point is a road surface, the safety system determines that there is no need for braking or avoidance. On the other hand, when the measurement point is a non-road surface, that is, an obstacle or the like, the safety system may assist the operation of the moving body by warning, automatic braking, automatic avoidance, or the like.

  On the other hand, in an optical distance measuring device using TOF, a photodiode may be used as a light receiving element for reflected light. Among photodiodes, a photodiode that applies a high voltage in the vicinity of a breakdown voltage (breakdown voltage) and uses a current internal amplification function is called an avalanche photodiode (APD).

  When photons enter the APD, electron-hole pairs are generated, and the generated electrons and holes are accelerated by a high electric field due to the high voltage applied to the APD, causing collisions and ionization one after another like an avalanche. New electron / hole pairs are generated. This is the APD internal amplification effect by the avalanche effect, and the sensitivity is enhanced by this internal amplification effect. Therefore, the APD is often used particularly in an optical distance measuring device that is required to detect a measurement object at a long distance.

  The operation mode of the APD includes a linear mode that operates with a reverse bias voltage smaller than the breakdown voltage and a Geiger mode that operates with a reverse bias voltage larger than the breakdown voltage.

  In the linear mode, the proportion of electron / hole pairs that disappear (out of the region of high electric field) is larger than the proportion of electron / hole pairs that are generated, and avalanche amplification stops naturally. The output current of the APD in the linear mode is substantially proportional to the amount of light incident on the APD.

  On the other hand, in the Geiger mode, avalanche amplification does not stop naturally, but it must be stopped by lowering the voltage applied to the APD to a breakdown voltage or lower. This forcibly stopping avalanche amplification is called quenching, and a circuit that brings a quenching action to APD is called a quenching circuit.

  The simplest quenching circuit is a quenching circuit using a resistor connected in series to the APD, and this series resistance is called a quenching resistor. When an avalanche current is generated, the voltage between the terminals of the quenching resistor increases, and the reverse bias voltage applied to the APD decreases by this increase. When the reverse bias voltage decreases to the breakdown voltage, avalanche amplification stops and the avalanche current does not flow. When the avalanche current stops, the voltage between the terminals of the quenching resistor decreases, and a reverse bias higher than the breakdown voltage is again applied to the APD, and the avalanche effect can be caused again.

  By taking out the change in potential between the APD and the quenching resistor through, for example, a buffer, it is possible to output photon incidence as a voltage pulse. In the Geiger mode, no information is included in the amplitude of the output signal itself, and the frequency of voltage pulses represents the amount of incident light. Since Geiger mode APDs can cause an avalanche effect even when a single photon is incident, they are also referred to as single photon avalanche diodes (SPAD).

  Here, with reference to FIG. 8, a description will be given of the time until the irradiation light irradiated to the measurement object is reflected by the measurement object and returned as reflected light, that is, the TOF. FIG. 8 is a diagram for explaining the output of the linear mode and the Geiger mode when the horizontal axis is time and the irradiation light pulse Pt and the reflected light pulse Pr are generated as shown in FIG. 8C. . In an apparatus that measures the distance to the measurement object using the TOF, since the light other than the light emitted by itself is noise, it is necessary to remove the influence of disturbance light such as sunlight as much as possible. Hereinafter, the output of each mode including the removal of the disturbance light will be described.

In APD linear mode, the output current I _L is approximately proportional to the amount of incident light, as shown in FIG. 8 (a), comparing processing with the threshold current I _th in after removing the direct current component included in the received current Thus, it is possible to extract the timing at which the reflected light arrives while reducing the influence of disturbance light. As a result, a correct TOF can be obtained even in the presence of disturbance light.

On the other hand, the SPAD Geiger mode, because the output voltage pulse V _P as shown in FIG. 8 (b) to the incident photon creates a histogram by repeatedly measuring the arrival time of the voltage pulse V _P, By extracting the peak value in the histogram, the correct TOF can be obtained even in the presence of disturbance light. FIG. 8D is a diagram showing an example of a histogram and the TOF obtained from the histogram, and the vertical axis shows the number of times of measurement (frequency) N.

  Regardless of which mode of APD is used, in order to obtain the TOF with high accuracy, the ratio of the disturbance light, more generally noise, to the signal that is the reflected light with respect to the light emitted by the received light signal, It is important to improve the S / N ratio.

  Next, a conventional technique for detecting a road surface from a large number of measurement points measured by an optical distance measuring device, that is, a point group will be described.

  For example, Patent Document 1 discloses a method for detecting the height of a road surface from a distance image (point cloud) obtained by a stereo camera. With reference to FIG. 9, the technique for detecting the height of the road surface disclosed in Patent Document 1 will be described. In Patent Document 1, as shown in FIG. 9A, a row L of corresponding points arranged in the height direction of the building 8 on the roadside 7 is extracted, and the lowest corresponding point having the lowest height from the corresponding point row L. To extract. Then, as shown in FIG. 9B, the height of the road surface of the traveling path 6 is detected by performing a curve approximation or a straight line approximation on the lowest corresponding point sequence arranged in the front-rear direction.

  Patent Document 2 discloses a method for determining whether or not a point cloud obtained by a laser radar is a road surface. In Patent Document 2, a pseudo target that is located at a distance equal to or smaller than a preset short distance threshold and whose reflected wave reception intensity is less than the threshold is tracked as a road surface candidate. A pseudo target recognized as a single object is detected over a plurality of divided areas, a depth width is equal to or less than a preset depth threshold value, and a moving speed is equal to or less than a preset stop determination threshold value. In this case, the pseudo target is determined to be a road surface.

  Furthermore, Patent Document 3 discloses a method for determining whether a measurement point is a road surface from the waveform of reflected light of a laser radar. With reference to FIG. 10, the method disclosed in Patent Document 3 for determining whether or not a measurement point is a road surface will be described. FIG. 10 shows the waveform of the irradiation pulse 23 and the waveform of the reflection pulse 25 when the irradiation beam 29 irradiated from the laser scanner 13 enters the reflection surface 35.

  As shown in FIG. 10A, when the reflecting surface 35 is a part of an obstacle such as a building, the incident angle of the irradiation pulse to the reflecting surface 35 is close to a right angle, but as shown in 10 (b). In addition, the incident angle of the irradiation pulse on the reflecting surface 35 when the reflecting surface 35 is a part of the road surface becomes small. Therefore, as shown in FIG. 10A, when the reflecting surface 35 is a part of an obstacle, the pulse width W1 of the reflected pulse 25 is relatively narrow. However, when the reflecting surface 35 is a road surface, a difference occurs in the distance to the reflecting surface 35 according to the spatial distribution of the irradiation beam 29. Therefore, as shown in FIG. Spread. In Patent Document 3, it is determined whether or not the measurement point is a road surface using the difference in the pulse width of the reflected pulse.

  On the other hand, as an example of a conventional receiving circuit for improving the S / N ratio of a light reception signal of an optical distance measuring device, one disclosed in Patent Document 4 is known. The receiving circuit disclosed in Patent Document 4 includes two systems of receiving units for receiving signals from the outside, amplifiers connected to the receiving units of each system, a switch having one end connected to the amplifier, and an output end of the amplifier. A comparator connected in parallel with the switch and an adder circuit connected to the other end of each of the two switches are provided.

Each receiving unit receives an external signal such as a pulse laser and outputs it to each amplifier. Each comparator compares the threshold voltage with the voltage of the signal output from each amplifier.
When the voltage of the signal output from each amplifier is higher than the threshold voltage, each comparator outputs a HIGH level signal to each switch, and electrically connects each amplifier and the addition circuit. . On the other hand, each comparator outputs a LOW level signal to each switch when the voltage of the signal output from the amplifier is lower than the threshold voltage, and electrically connects each amplifier and summing circuit. To separate.

  In the receiving circuit disclosed in Patent Document 4, the amplitude of each of a plurality of amplifiers is compared with a predetermined threshold, and when the amplitude of the signal output from the amplifier is equal to or greater than the threshold, the amplifier and the adding circuit Are electrically connected, and when the amplitude of the signal output from the amplifier is less than the threshold, the amplifier and the adding circuit are electrically separated. In Patent Document 4, if the amplitude of a signal output from an amplifier is less than a threshold value, noise included in the signal is not subjected to addition processing by an addition circuit, thereby preventing a decrease in S / N ratio caused by the addition processing. It is said that the S / N ratio can be improved more appropriately.

JP 2009-230709 A JP 2014-089114 A EP1286178A2 JP2011-239329A

  In general, when a road surface is detected from a point group, statistical estimation is performed in consideration of continuity and linearity of a large number of measurement points, as in the method of detecting a road surface disclosed in Patent Document 1. Therefore, in order to estimate the road surface more accurately, it is preferable to obtain a large number of measurement points farther away. However, the laser beam power density decreases as the distance increases. Furthermore, as described above, since the incident angle to the road surface decreases as the distance increases, the pulse width of the reflected pulse increases, so that the peak power of the reflected pulse decreases. For this reason, it becomes difficult to measure the distance of the road surface as the distance increases.

  In Patent Document 2, a measurement point with a low reflection intensity among the measurement points at a short distance is used as a road surface candidate. However, although the method disclosed in Patent Document 2 is effective for detecting a road surface at a short distance, it is difficult to improve the detection performance of a road surface at a far distance.

  Further, in Patent Document 3, a road surface candidate is a point where the pulse width of the reflected pulse is wide and therefore the intensity is small. However, even if the surface is perpendicular to the road surface, such as a wall surface, the incident angle of the irradiated light to the measurement object may be reduced depending on the position and orientation, and there is a risk that an obstacle will be erroneously detected as a road surface. . Therefore, it is difficult to improve the detection performance of a distant road surface in the case of Patent Document 3 as well.

  On the other hand, regarding the S / N ratio of the receiving circuit, in the method of Patent Document 4, when the output of the receiving unit is equal to or greater than the threshold value, the switch is turned on and the output of the receiving unit is added by the adding circuit. That is, it is based on the assumption that the signal component is a signal component when the output of the receiver is equal to or greater than a threshold value. Therefore, if the threshold value is set large, only a large peak value, that is, a signal that does not require improvement of the S / N ratio is added. On the other hand, if the threshold value is set small, a small peak value is added, but a noise component is also added, and the S / N ratio is lowered, which may cause erroneous detection.

  In other words, if the signal component and the noise component can be easily and clearly separated by the threshold processing, the TOF can be easily obtained, so there is no need to improve the S / N ratio. When the signal component and the noise component cannot be easily separated by the threshold processing, the S / N ratio needs to be improved. However, since the magnitudes of the signal component and the noise component are close to each other, it is not easy to set the threshold. Therefore, in the S / N ratio improvement method of the receiving circuit disclosed in Patent Document 4, the S / N ratio cannot be improved according to the noise level, and the detection performance of the optical distance measuring device is improved. There is room for improvement from the viewpoint.

  The present invention has been made in view of the above problems, and realizes an optical distance measuring device with improved detection performance of a measurement object, particularly a measurement object located far away or a measurement object with low reflectance. For the purpose.

In order to achieve the above object, the optical distance measuring device according to claim 1 includes a light projecting unit that projects an irradiation light pulse on a measurement object, and a reflected light pulse reflected by the measurement object. A light receiving unit that is disposed on the imaged imaging surface and has a plurality of light receiving pixels that receive the reflected light pulse; and a pixel of interest among a plurality of light receiving signals output from each of the plurality of light receiving pixels; A selection unit that selects a light reception signal output from one or a plurality of pixels in the vicinity of the target pixel and outputs it as a plurality of individual light reception signals, and a synthesis unit that outputs a combined light reception signal obtained by adding the plurality of individual light reception signals And calculating a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a reception time of the reflected light pulse using the combined light reception signal, and the measurement based on the flight time. The distance to the object Seen including a distance calculating unit for output, wherein the combining unit is configured to correct the difference between the light receiving time of the reflected light pulse by the observation direction of the light receiving pixels are different by adding the plurality of individual light signals, wherein Outputs the combined light reception signal .

According to a second aspect of the present invention , the light projecting unit that projects the irradiation light pulse onto the measurement object and the imaging surface on which the reflected light pulse reflected by the measurement object is formed are arranged. And a light receiving unit having a plurality of light receiving pixels that receive the reflected light pulse, and one or more of the target pixel and the vicinity of the target pixel among the plurality of light receiving signals output from each of the plurality of light receiving pixels. A selection unit that selects a light reception signal output from each pixel and outputs it as a plurality of individual light reception signals, a combination unit that outputs a combined light reception signal obtained by adding the plurality of individual light reception signals, and the combined light reception signal Distance calculation that calculates a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a reception time of the reflected light pulse, and calculates a distance to the measurement object based on the flight time and parts, signal and miscellaneous The individual reliability, which is the reliability of the peak value detection of the individual light reception signal, is calculated based on the ratio of the received light signal. A reliability calculating unit that calculates a degree , and the combining unit outputs the combined light reception signal when the combined reliability is greater than the individual reliability, and the combined reliability is equal to or less than the individual reliability. In this case, the individual light reception signal corresponding to the target pixel is output as the combined light reception signal.

According to a third aspect of the present invention, in the first aspect of the present invention, an individual reliability that is a reliability of peak value detection of the individual received light signal is calculated based on a ratio between a signal and noise, And a reliability calculating unit that calculates a combined reliability that is a reliability of peak value detection of the combined light reception signal based on a ratio of noise to noise, and the combining unit includes the combined reliability that is higher than the individual reliability. When it is larger, the combined light reception signal is output, and when the combined reliability is equal to or less than the individual reliability, the individual light reception signal corresponding to the target pixel is output as the combined light reception signal.

According to a fourth aspect of the present invention, in the third aspect of the present invention, when the composite reliability is greater than the individual reliability, the measurement object is determined to be a road surface, and the composite reliability is determined. When the degree is equal to or less than the individual reliability, the apparatus further includes a determination unit that determines that the measurement object is a non-road surface.

According to a fifth aspect of the present invention, in the invention according to any one of the second to fourth aspects, the light receiving unit includes a light receiving pixel that receives only disturbance light that does not include the reflected light pulse. The disturbance light receiving unit is provided, and the reliability calculation unit calculates the individual reliability using the peak value of the light reception signal as the signal and the output of the disturbance light reception unit as noise.
According to a sixth aspect of the present invention, the light projecting unit for projecting the irradiation light pulse onto the measurement object and the imaging surface on which the reflected light pulse reflected by the measurement object is imaged are arranged. And a light receiving unit having a plurality of light receiving pixels that receive the reflected light pulse, and one or more of the target pixel and the vicinity of the target pixel among the plurality of light receiving signals output from each of the plurality of light receiving pixels. A selection unit that selects a light reception signal output from each pixel and outputs it as a plurality of individual light reception signals, a combination unit that outputs a combined light reception signal obtained by adding the plurality of individual light reception signals, and the combined light reception signal Distance calculation that calculates a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a reception time of the reflected light pulse, and calculates a distance to the measurement object based on the flight time And The light projecting unit repeatedly projects the irradiation light pulse on the measurement object over a predetermined measurement time, and the selection unit is configured to focus the target pixel and one or more pixels in the vicinity of the target pixel. Each of which is output as an individual histogram obtained by accumulating the received light signal of the irradiation light pulse repeatedly projected along the time axis, and the combining unit adds the individual histogram. Is output as the combined light reception signal, and the distance calculation unit calculates the distance to the measurement object using the time corresponding to the peak value of the combined histogram as the light reception time.

The invention according to claim 7 is the invention according to any one of claims 1 to 5 , wherein the light projecting unit is repeatedly applied to the measurement object over a predetermined measurement time. and projecting the irradiation light pulse, the selection unit is output from each of the one or more neighboring pixels of the target pixel as a target pixel, a light reception signal of the irradiated light pulses repeatedly projected time The individual histogram accumulated along the axis is output as the individual light receiving signal, the combining unit outputs a combined histogram obtained by adding the individual histograms as the combined light receiving signal, and the distance calculating unit is a peak value of the combined histogram The distance to the measurement object is calculated using the time corresponding to 1 as the light reception time.

  According to the present invention, there is an effect that it is possible to realize an optical distance measuring device in which detection performance of a measurement object, particularly a measurement object located far away or a measurement object having a low reflectance, is enhanced.

It is a side view which shows an example of a structure of the optical ranging apparatus which concerns on embodiment. It is a top view which shows an example of a structure of the light reception LSI which concerns on embodiment. It is a figure which shows the observation direction of the TOF pixel which concerns on embodiment, and the observation direction of a reference pixel. It is a block diagram which shows the TOF signal processing part and reference signal processing part which concern on embodiment. It is a figure for demonstrating the geometric relationship between the optical ranging apparatus which concerns on embodiment, its observation direction, and a road surface. It is a figure for demonstrating the separate histogram and synthetic | combination histogram which concern on 1st Embodiment. It is a figure for demonstrating the reliability which concerns on 2nd Embodiment. It is a figure for demonstrating the linear mode and Geiger mode of an avalanche photodiode which is a light receiving element. It is a figure for demonstrating the detection method of the road surface which concerns on a prior art. It is a figure for demonstrating the detection method of the other road surface which concerns on a prior art.

[First Embodiment]
Hereinafter, the present embodiment will be described in detail with reference to FIGS. 1 to 6. First, the basic configuration of the optical distance measuring device 10 according to the present embodiment will be described with reference to FIG. To do.
The optical distance measuring device 10 includes a light projecting unit 30, a light receiving unit 32, a hyperboloid mirror 20, a polygon mirror 22, and a housing 26.

  The light projecting unit 30 is a part that generates the light projecting light Lt that is the irradiation light that irradiates the measurement target space. The light projecting unit 30 includes a light projecting substrate 16 that includes an LD (Laser Diode) element 18 as a light source. Yes. In the present embodiment, the light projecting light Lt has a pulse-like waveform that is continuously projected, and the light projecting substrate 16 uses the continuous light emitted from the LD element 18 as a light pulse (pulse laser light). An LD element driving circuit (not shown) is included. In the present embodiment, an example in which an LD element is used as a light source will be described. However, another light source such as a solid-state laser may be used. As an example, the period of the pulse laser beam can be 4 μs (microsecond), and the pulse width can be about several ns (nanosecond).

  The light receiving unit 32 is a part that receives reflected light reflected by the measurement object as received light Lr, and includes a light receiving substrate 14 on which a light receiving LSI (Large Scale Integrated Circuits) 12 is mounted.

  The polygon mirror 22 is a rotating polygon mirror having six mirror surfaces. The polygon mirror 22 scans the projection light Lt to irradiate the measurement target space as irradiation light, and scans the reflected light from the measurement target to scan the hyperboloid. This is the part that leads to the mirror 20.

  Each mirror surface of the polygon mirror 22 is inclined with respect to the rotation axis A, and the polygon mirror 22 is rotated about the rotation axis A at a predetermined rotation speed. Since the angles formed by the mirror surfaces of the polygon mirror 22 and the rotation axis A according to the present embodiment are different from each other, the depression angles of the irradiation light and the reflected light change as the polygon mirror 22 rotates. .

  Therefore, by rotating the polygon mirror 22, the light projection Lt from the LD element 18 is scanned not only in the horizontal direction (scanning) but also in the vertical direction at different depression angles, that is, scanned in a planar shape, Irradiated toward the space to be measured. In the present embodiment, an example using a polygon mirror 22 having six surfaces will be described as an example. However, the present invention is not limited to this, and a configuration using polygon mirrors having other numbers of surfaces according to specific design conditions and the like. It is good.

  The hyperboloid mirror 20 is a part that collects the reflected light reflected by the measurement object and guides it to the light receiving LSI 12 as the received light Lr.

  The housing 26 includes a power supply board 24 and a control board 27 together with a support structure that supports each of the above-described structures, and the light projecting board 16 and the light receiving board 14 are connected to the power supply board 24. The power supply board 24 includes a circuit that generates a voltage to be supplied to the light projecting board 16, the light receiving board 14, and the control board 27 using power supplied from the outside. In addition, the control board 27 is provided with a control unit (not shown) that performs overall control of the entire optical distance measuring device 10, and the control unit executes control of the light projecting board 16 and the light receiving board 14. To do. The control unit includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like (not shown).

  Next, the operation of the optical distance measuring device 10 having the above configuration will be described.

  The pulsed laser light emitted from the LD element 18 is collimated by a collimator lens (not shown) (made parallel light) and guided to the polygon mirror 22 from the opening K provided in the hyperboloid mirror 20. The projection light Lt reflected by the polygon mirror 22 is projected toward the measurement target space, irradiated to the measurement target in the measurement target space, and the reflected light reflected by the measurement target is directed to the optical distance measuring device 10. And return again. The collimating lens defines a light projection space and may be used as necessary.

  The reflected light that has returned to the optical distance measuring device 10 is reflected again by the polygon mirror 22, further reflected by the hyperboloid mirror 20, and enters the light receiving LSI 12 as the received light Lr. The hyperboloid mirror 20 has an effect similar to that of a lens, and forms an image of the received light Lr on a light receiving surface (described later) of the light receiving LSI 12.

  As is clear from the above description, the optical system of the optical distance measuring device 10 according to the present embodiment is a coaxial optical system in which the optical axis in light projection coincides with the optical axis in light reception.

  Next, the light receiving LSI 12 according to the present embodiment will be described with reference to FIG. The light receiving LSI 12 is a semiconductor integrated circuit in which a light detection unit that receives and photoelectrically converts the received light Lr and a signal processing unit that processes a signal from the light detection unit are integrated. In the present embodiment, an example in which a silicon photomultiplier (SiPM) is used as the light detection unit will be described. The SiPM is a configuration in which a plurality of SPADs are arranged in a matrix and a large photodetector, that is, a pixel as a whole.

  By the way, in order to extract the peak value with high accuracy from the histogram in which the incident time of the photon is repeatedly measured as shown in FIG. 8D, many measurements are required, and as a result, measurement for obtaining the TOF. The time will be longer. When the measurement time becomes long, a measurement error occurs when the measurement object moves. On the other hand, the photodetector using SiPM is a method that can obtain the correct TOF while eliminating the influence of disturbance light in a short measurement time (for example, JP 2012-06001 A, C. Niclass, M .Soga, H. Matsubara, S. Kato, “A 100m-range 10-frame / s 340 × 96-pixel time-of flight depth sensor in 0.18μm CMOS”, Proceeding of ESSCIRC, pp.107-110, September, See 2011.).

  As shown in FIG. 2, the light detection unit 50 of the light receiving LSI 12 according to the present embodiment includes a signal light detection unit 130 and a reference light detection unit 230. The signal light detection unit 130 is a light detection unit that detects the received light Lr for measuring the distance, and the reference light detection unit 230 is a light detection unit that detects the intensity of disturbance light.

  The signal light detection unit 130 according to the present embodiment includes 16 TOF pixels GT1 to GT16 (hereinafter, collectively referred to as “TOF pixels GT”) arranged in a one-dimensional manner for detecting the received light Lr. It consists of Each of the TOF pixels GT is configured by SPAD 120 being arranged in an array of 4 rows × 6 columns.

  On the other hand, the reference light detection unit 230 according to the present embodiment detects 16 reference pixels GR1 to GR16 (hereinafter, collectively referred to as “reference pixels GR”) arranged in a one-dimensional manner to detect disturbance light. It is configured to include. Each of the reference pixels GR is configured by SPADs 220 arranged in an array of 4 rows × 6 columns. In the present embodiment, an element having the same structure as that of the signal light detection unit 130 is used as the reference light detection unit 230.

  The beam shape of the pulse laser beam emitted from the light projecting unit 30 according to the present embodiment is a vertically long beam shape in the vertical direction (Z direction in FIG. 1). The TOF pixel GT of the signal light detection unit 130 and the reference pixel GR of the reference light detection unit 230 are also arranged in the vertical direction (Z direction in FIG. 2). Since the reflected light also returns to the optical distance measuring device 10 with the vertically long beam shape, the received light Lr having the vertically long beam shape enters the signal light detection unit 130 along the arrangement direction of the TOF pixels GT. That is, the reflected light pulse corresponding to one pulse laser beam is collectively received by the signal light detection unit 130.

  Furthermore, in the present embodiment, the light receiving LSI 12 is arranged so that the light receiving region of the signal light detecting unit 130 and the beam spot of the reflected light pulse imaged on the light receiving surface of the signal light detecting unit 130 overlap. With this arrangement, the signal light detection unit 130 detects reflected light and disturbance light, whereas the reference light detection unit 230 detects only disturbance light without detecting reflected light.

  As described above, in the present embodiment, when the polygon mirror 22 rotates, the light beam of the projection light Lt (projection beam) and the beam of the reception light Lr (light reception beam) are simultaneously scanned in the horizontal direction. Is done. The polygon mirror 22 according to the present embodiment is composed of six surfaces, and each surface is formed so as to have a different depression angle, so that scanning is also performed in six vertical directions. By making the difference in depression angle of each surface of the polygon mirror 22 substantially coincide with the beam divergence angle of the projection beam, scanning can be performed in the vertical direction without any gap. Therefore, the number of scanning lines in the vertical direction of the optical distance measuring device 10 is 96 lines, which is the product of the number 16 of TOF pixels GT and the number 6 of surfaces of the polygon mirror 22 in the signal light detector 130.

  Next, the observation directions of the signal light detection unit 130 and the reference light detection unit 230 will be described with reference to FIG. As shown in FIG. 3, in this embodiment, the observation direction DT of the signal light detection unit 130 and the observation direction DR of the reference light detection unit 230 are configured to always observe different directions. An arrow S shown in FIG. 3 indicates the scanning direction of the light projecting unit 30.

  The angle difference φ in the observation direction is determined by the horizontal distance d (see FIG. 2) between the signal light detection unit 130 and the reference light detection unit 230 on the light receiving LSI 12 and the focal length of the hyperboloid mirror 20. The angle between the current observation direction and the next observation direction of the signal light detection unit 130 is determined by the rotation speed of the polygon mirror 22, the light emission time interval of the pulsed light, and the like. If the angle difference φ and the focal length of the hyperboloid mirror 20 are fixed, the observation direction of the reference light detection unit 230 is adjusted by adjusting the distance d between the signal light detection unit 130 and the reference light detection unit 230. Can be matched with the next observation direction of the signal light detection unit 130. In the present embodiment, by configuring as described above, the observation direction DT of the signal light detection unit 130 and the observation direction DR of the reference light detection unit 230 are always different directions.

  Next, the signal processing unit according to the present embodiment will be described with reference to FIG. The signal processing unit according to the present embodiment is a part that processes a light reception signal from the light detection unit 50, and a TOF signal processing unit 100 that obtains TOF using the light reception signal output from the signal light detection unit 130; And a reference signal processing unit 200 that obtains the magnitude of disturbance light using a light reception signal output from the reference light detection unit 230.

  As shown in FIG. 4A, the TOF signal processing unit 100 includes a front end unit 102 (front end units 102-1 to 102-24) and a pulse shaping circuit 104 (pulse shaping circuits 104-1 to 104-24). , An adder 106, a comparator 108, a TDC (Time to Digital Converter) 110, and a histogram processing unit 112.

  The front end unit 102 includes a SPAD 120, a resistor 122 connected in series to the SPAD 120, and a preamplifier 124 that inputs a connection point between the SPAD 120 and the resistor 122. A reverse bias voltage higher than the breakdown voltage is applied to each of the SPADs 120, and each of the SPADs 120 causes an avalanche effect with respect to the incidence of photons and outputs a voltage pulse (light reception pulse). The resistor 122 is the quenching resistor described above, and has a function of stopping the avalanche effect of the SPAD 120. The preamplifier 124 is a buffer that extracts a change in potential at a connection point between the SPAD 120 and the resistor 122.

  The SPAD 120 illustrated in FIG. 4A corresponds to each of the SPAD 120 illustrated in FIG. 2A, and each of the 24 front end units 102-1 to 102-24 is included in the TOF pixel GT. It is provided corresponding to each SPAD 120. Further, 16 sets of 24 front end portions 102 are provided corresponding to each of the TOF pixels GT, and a front end portion corresponding to the signal light detection portion 130 is configured. In the front end unit 102 according to the present embodiment, other circuits such as an amplifier for matching the signal level with the pulse shaping circuit 104 in the next stage may be provided in the subsequent stage of the preamplifier 124.

  The pulse shaping circuit 104 (pulse shaping circuits 104-1 to 104-24) is connected to each of the 24 front end units 102 and shapes the waveform of the received light pulse output from the front end unit 102. The pulse width of the received light pulse output from the front end unit 102 corresponds to a time constant determined by the parasitic capacitance of the SPAD 120 and the quenching resistor 122 (as an example, several tens to several hundred ns). The pulse shaping circuit 104 performs shaping to narrow the pulse width of the received light pulse to about the pulse width of the pulse laser beam output from the light projecting unit 30 (as an example, about several ns).

  The adder 106 is a circuit that adds (synthesizes) the outputs from the 24 pulse shaping circuits 104 and outputs the addition result to the comparator 108. The adder 106 adds the received light pulses output from the plurality of SPADs 120 at the same timing. In the present embodiment, the addition result is output as a digital signal. For example, if a light reception pulse is simultaneously output from two of the plurality of SPADs 120, the output signal from the adder 106 is a digital signal “2” (decimal number).

  The comparator 108 compares the added value output from the adder 106 with a predetermined threshold, and when the added value is equal to or greater than the threshold, the light pulse of the received light Lr (reflected light) corresponding to the pulsed laser light of the light projection light Lt. The determination result indicating that the (pulse) has arrived is output. In this embodiment, the threshold is set to 2, and when the added value output from the adder 106 is 2 or more, the comparator 108 outputs a pulse signal.

  In the present embodiment, the threshold value is 2 as an example, but the threshold value is used to determine when a predetermined number or more of SPADs simultaneously output light reception pulses, that is, when a large number of photons have arrived at the same time. The threshold value need not be 2, and an optimal value may be set in consideration of specific design conditions and the like. The output of the comparator 108 becomes a trigger pulse for triggering the TDC 110 at the next stage.

  The TDC is a circuit that generally measures a time difference between pulse signals and converts the time difference into a digital numerical value and outputs the digital value. The TDC 110 according to the present embodiment measures the arrival time of the reflected light pulse using the trigger pulse output from the comparator 108. That is, the TDC 110 outputs the time from the light projecting time when the pulse laser beam is projected from the light projecting unit 30 to the light receiving time when the reflected light pulse is detected by the comparator 108 as a TOF measurement result.

The histogram processing unit 112 further accumulates the TOF measurement results obtained by the TDC 110 over a predetermined measurement time, and generates a histogram H (see FIG. 6). The histogram H is generated by repeatedly irradiating a pulse laser beam for measuring TOF from the light projecting unit 30 over a predetermined measurement time and accumulating the measurement results of TOF obtained corresponding to each pulse laser beam. Is done. That is, for the time _t r _-t t from the light projecting time _{t t} of the pulsed laser beam from the light projecting unit 30 to the light receiving time _{t r} of the reflected light pulse in the light receiving portion 32 (= TOF), a comparator in a given measurement time Reference numeral 108 denotes the number of times (frequency) at which the pulse signal is detected. In the present embodiment, every time the TOF is measured, the output of the adder 106 is added to the corresponding bin value of the histogram H. Thereby, the number of SPADs that simultaneously output pulses, that is, the amount of incident light is accurately reflected by the frequency of the histogram H.

  Since the histogram H generated by the above measurement over a predetermined measurement time is statistically accumulated, even a weak reflected light pulse from a distant road surface can be captured. Thus, by correctly measuring the arrival time of the reflected light pulse to generate the histogram H and extracting the peak value of the generated histogram H, it is possible to perform correct TOF measurement even in the presence of disturbance light. .

  In the TOF calculation process using the histogram H described above, when a reflected light pulse with respect to the pulse laser light projected from the light projecting unit 30 enters the signal light detecting unit 130, a large number of photons arrive at the same time, When the disturbance light is incident on the signal light detection unit 130 while being received by the SPAD 120, since the photons arrive at random timing, processing based on the property that the probability of simultaneous reception by the plurality of SPADs 120 is low. It is.

  According to the TOF signal processing unit 100 configured as described above, since the response to photons caused by ambient light can be reduced, the TOF can be correctly measured with a small number of measurements.

  Next, reference signal processing section 200 according to the present embodiment will be described. As shown in FIG. 4B, the reference signal processing unit 200 includes a front end unit 202 (front end units 202-1 to 202-24) and a pulse shaping circuit 204 (pulse shaping circuits 204-1 to 204-24). , An adder 206, a comparator 208, and a counter 210.

  The reference signal processing unit 200 is obtained by replacing the TDC 110 and the histogram processing unit 112 in the TOF signal processing unit 100 with a counter 210. Accordingly, the operations of the front end unit 202, the pulse shaping circuit 204, the adder 206, and the comparator 208 are the same as the operations of the front end unit 102, the pulse shaping circuit 104, the adder 106, and the comparator 108, respectively, and thus description thereof is omitted. .

  As described above, since only the disturbance light is incident on the reference signal processing unit 200, the disturbance light (background light) is measured by measuring the number of output pulses of the comparator 208 over a predetermined time. ) Can be measured. The reflected light pulse information obtained by the processing in the TOF signal processing unit 100 includes a net TOF signal component and a background light component due to disturbance light. Therefore, when extracting the peak value of the histogram H, the threshold value can be appropriately determined by using the information processed by the reference signal processing unit 200.

  Next, referring to FIG. 5, the difference in distance between the optical distance measuring device 10 between the TOF pixels and a point on the road surface when the road surface is measured by the optical distance measuring device 10 mounted on the vehicle. That is, the relationship between the depression angle and the difficulty of detecting the road surface will be described.

As shown in FIG. 5, the optical distance measuring device 10 according to the present embodiment is mounted on the front upper portion of the vehicle 1 at a position where the height from the road surface R is H. From the light projecting unit 30 of the optical distance measuring device 10, assuming that the portion received by the i-th TOF pixel GT in the light Lt continuously scanned by the polygon mirror 22 is Lt _i. The light beam Lt _i is projected to a point S _i on the road surface R where the horizontal distance to the optical distance measuring device 10 is X _i . The optical axis of the projected light Lt _i, an angle between a road surface R and theta _i, is defined as depression of the angle theta _i. As described above, in the present embodiment, since the signal light detection unit 130 and the reference light detection unit 230 are arranged in the vertical direction (that is, the vertical direction) with respect to the road surface R, the depression angle θ _{i is set} to the i th This is also referred to as the depression angle in the observation direction of the TOF pixel GT or the reference pixel GR.

When the road surface R is assumed to be a horizontal plane, the relationship between the distance D _i from the optical distance measuring device 10 to the point S _i on the road surface R and TOFτ _i is as follows from the geometric relationship shown in FIG. It is represented by (Equation 1) shown.


Here, c is the speed of light, and c = 3 × 10 ⁸ m / sec (second).

Relationship between the degree of difficulty of detecting the point _{S i} on the depression angle theta _i and the road surface R, will be described with reference to the rate of change dτ _i / dθ _i of TOFtau _i for depression theta _i. From (Expression 1) and (Expression 2), dτ _i / dθ _i is given by (Expression 3) shown below.

As an example, when H = 1.4 m and the road surface R 5 m ahead (that is, the horizontal distance X _i = 5 m) is measured, the depression angle θ _i is θ _i = 15.6 degrees. Substituting θi = 15.6 degrees, H = 1.4 m, and c = 3 × 10 ⁸ m / sec into (Equation 3), the rate of change dτ _i / dθ _{i with} respect to the depression angle θ _i of TOFτi is dτ _i / dθ _i = 1.23 × 10 ⁻⁷ (sec / degree). If the vertical field angle of one TOF pixel GT is 0.1 degree, the change rate Δτ _i of TOFτ _i in one TOF pixel GT is Δτ _i = 0.22 nsec (nanosecond).

On the other hand, when H = 1.4 m and the road surface R 30 m ahead (that is, the horizontal distance X _i = 30 m) is measured, the depression angle θ _i is θ _i = 2.67 degrees, and Δτ _i in this case is , Δτ _i = 7.5 nsec. Further, when H = 1.4 m and the road surface R 50 m ahead (that is, the horizontal distance X _i = 50 m) is measured, the depression angle θ _i is θ _i = 1.60 degrees, and Δτ _i in this case is , Δτ _i = 20 nsec.

That is, it can be seen that even when the pixel angle of view is relatively narrow, such as 0.1 degrees, the rate of change Δτ _i of TOFτ _i within one TOF pixel GT increases in the observation of a distant road surface R. When the rate of change Δτ _i of TOFτ _i in one TOF pixel GT increases, the pulse width of the reflected light pulse widens and the peak value decreases, which makes it difficult to measure TOFτ _i .

Next, with reference to FIG. 6, a method for improving the detection performance up to a far road surface in the present embodiment for solving the above problem will be described. As is apparent from the above description, the TOF based on the reflected light pulse received by each TOF pixel GT varies depending on the angle of depression of the projection light Lt _i for each TOF pixel GT. Therefore, when a light reception signal for each TOF pixel GT, that is, a histogram is used, the peak value has a certain limit, and the far road surface detection performance naturally has a limit.

  Therefore, in the present embodiment, a technique is adopted in which the amount of TOF shift between the TOF pixels GT arranged in the vicinity is corrected and the light reception signals from the plurality of TOF pixels GT are combined. As a result, when the object to be measured is a road surface, the peak values of the received light signals are added and the received light signal is further increased, so that the S / N ratio of the received light signal can be further increased. On the other hand, when the measurement object is a non-road surface such as an obstacle, the peak values are not added to each other, so that the S / N ratio does not increase.

  As a specific method for the above correction, in this embodiment, a histogram of a target pixel and a histogram of one or a plurality of pixels in the vicinity of the target pixel (hereinafter referred to as “individual histogram”) are added, integrated, and combined. By obtaining the histogram and calculating the TOF using the time corresponding to the peak value of the combined histogram, the detection performance of the road surface farther away is enhanced.

FIG. 6A shows an individual histogram H _{i of} a TOF pixel GT _i (hereinafter referred to as “pixel _i ”) of interest in the signal light detection unit 130, and pixels in the vicinity of the pixel _i , that is, two pixels before and after the pixel _i. , Pixel _i-2 , pixel _i-1 , pixel _{i + 1} , and pixel _{i + 2} are shown as individual histograms H _i-2 , H _i-1 , H _{i + 1} , and H _{i + 2} . Since depression for each pixel is different, different _{times, t i-2, t i} -1, t i, in _{t i + 1, t i +} 2, the peak PK are formed by the reflected light pulse.

Since the installation position and orientation of the optical distance measuring device 10 in the vehicle 1 are known and the depression angle θ _i of each pixel _i is also known, assuming the shape of the road surface R, the times t _i−2 and t _i− Calculations of ₁ , t _i , t _{i + 1} , and t _{i + 2} are possible. In this embodiment, each pixel, pixel _i-2, pixel _i-1, pixel _i, pixel _{i + 1,} and the peak value formation time of the pixel _{i + 2,} based on the peak value formation time of the pixel _{i, t i-2 -t i, t i-1 -t} i, 0, t i + 1 -t i, summed shifted by _{t i} + 2 -t _i, integrates obtain a combined histogram _{H T} as shown in Figure 6 (b). If the observation target of each pixel is the above-described assumed road surface R, the time when the peak value of the individual histogram H _i corresponding to each pixel is overlapped is added and integrated, so that the combined histogram H _T has a larger peak value. Form. As a result, the road surface R detection performance is enhanced.

  As described above in detail, according to the optical distance measuring device according to the present embodiment, it is possible to improve the detection performance of a measurement object, particularly a measurement object located far away. When the object to be measured farther can be detected, the measurable measurement points can be increased. When a larger number of measurement points are obtained, statistical estimation of the measurement object in consideration of continuity and linearity becomes more accurate and more robust.

[Second Embodiment]
In the present embodiment, an index called “reliability” is introduced in order to determine whether or not the peak value detection performed on a histogram using the above-described method, that is, the measurement of TOF is reliable. In addition, the road surface detection performance is further enhanced. Even when the correction according to the above-described embodiment is performed, it cannot be said that there is no case where an erroneous TOF is measured by the corrected received light signal. This embodiment is a mode for the purpose of avoiding this erroneous measurement.

  With reference to FIG. 7, a method of calculating the reliability of the TOF measured using the outputs of the TOF signal processing unit 100 and the reference signal processing unit 200 will be described. FIG. 7 is an example of a histogram created by the TOF signal processing unit 100, where the horizontal axis indicates time and the vertical axis indicates frequency (count value). In FIG. 7, P represents a peak value of the intensity of the reflected light pulse (indicated as “P: peak level” in FIG. 7), and N represents an average value of the intensity of disturbance light (“N: noise level in FIG. 7). ”).

As shown in FIG. 7, the histogram calculating the TOF, peak value P is formed at the time indicated the TOF t _{0 (time} corresponding to the light receiving time t _r). This peak value indicates the intensity of the reflected light pulse. However, this peak value P includes a disturbance light component as an offset. Therefore, PN obtained by subtracting the average value N of the disturbance disturbance light component from the peak value P becomes a net TOF signal component when viewed from the peak value.

In the present embodiment, the reliability of the histogram shown in FIG. 7 is defined using the S / N ratio of the histogram shown in FIG. If the fluctuation amount of the disturbance light component is regarded as a noise component, the noise component of the histogram shown in FIG. 7 can be expressed by the square root of the average value N of the disturbance light component. Therefore, since the S / N ratio of the histogram shown in FIG. 7 is obtained by dividing the net TOF signal component PN by the square root of the average value N of the noise components, the reliability C is shown below (Formula 4). Can be defined in

The reliability C can be calculated without distinction between the individual histogram H _i and the combined histogram H _T.

Here, in the following, calculating the composite histogram H _T is referred to as “correction”. Thus, the individual histograms H _i of before combining the correction is not performed. In the present embodiment, by comparing the reliability with correction and the reliability without correction, whether or not the measurement object is a road surface (whether it is an obstacle or the like). This makes it possible to determine more reliably.

The disturbance light component after correction, that is, after synthesis is obtained by calculating the sum of the outputs of the reference pixels GR corresponding to the TOF pixels GT to be synthesized. TOF signal component after synthesis can be obtained from the peak value of the combined histogram H _T.

As an example, consider a light reception signal of two pixels in which the peak value of the TOF signal component before synthesis is P, the average value of the disturbance light component is N, and the noise component due to the disturbance light is the square root of N. In this case, since the average value of the disturbance light component after synthesis is 2 · N, the noise component is the square root of 2 · N. When the observation target of the pixel is the road surface, the peak value of the combined TOF signal component is 2 · P. Therefore, the reliability C ^* after synthesis, that is, after correction, can be expressed as (Equation 5) shown below.

Since (Equation 5) can be transformed as shown in (Equation 6) below, it can be seen that the reliability C ^* after synthesis, that is, after correction, is a square root of 2 times the reliability C before synthesis, that is, before correction. .

For a three-dimensional object whose observation target is not a road surface, for example, a three-dimensional object having a surface perpendicular to the road surface, when considering the synthesis of two pixels as described above, the peak value of the TOF signal component including the disturbance light component after correction is P + N. The peak value of the net TOF signal component after correction is the same PN as before correction. On the other hand, since the corrected noise component is the square root of 2 · N, the reliability C ′ in this case is given by (Equation 7) shown below.

Since (Expression 7) can be transformed as shown in (Expression 8) below, the reliability C ′ after correction is obtained by dividing the reliability C before correction by the square root of 2.

  From the above examination results, in the present embodiment, when the reliability after correction becomes larger than the reliability before correction, it is determined that the observation target is a road surface, and the reliability after correction is the reliability before correction. In the following cases, it is determined that the observation target is a non-road surface (obstacle). When the road surface is determined, the corrected composite histogram is used to detect the peak value, and when the road surface is determined to be a non-road surface, the uncorrected histogram is used to detect the peak value.

As described above in detail, according to the present embodiment, the reliability based on the presence or absence of correction is compared, and the object whose peak value is detected differs depending on whether the observation target is a road surface or a non-road surface. False detection can be prevented. As a result, detection performance is enhanced road R, also, because becomes more accurate than the measurement of the light receiving time t _r, the distance to the road surface located more distant, it is possible to more accurately measure.

[Third Embodiment]
In the above-described embodiment, the configuration in which the detection performance of the road surface R is mainly enhanced by using the reliability C of the histogram peak value detection based on the S / N ratio of the histogram is shown. In the present embodiment, the S / N ratio of the histogram is used to improve the S / N ratio of the receiving circuit (specifically, the TOF signal processing unit 100) according to the noise level.

Here, for convenience of explanation, the reliability of the individual histogram H _i is referred to as “individual reliability C _i ”, and the reliability of the combined histogram H _T is referred to as “composite reliability C _T ”. In the above embodiment, the main purpose is to detect the road surface R. Therefore, the depression angle is different for each pixel _i , and the peak PK is formed by the reflected light pulse at different times (the observation target is at a different distance). The existing form) has been described as an example. On the other hand, this embodiment is a mode in which the S / N ratio of the receiving circuit is improved regardless of whether the observation target is at a different distance or the same distance. Therefore, in the following, a case where the observation target exists at the same distance will be described as an example. The case where the observation target exists at the same distance is, for example, a case where an obstacle having a plane substantially perpendicular to the road surface R is observed.

As described above, the individual reliability C _i is the S / N ratio of the individual histogram H _i , and the combined reliability C _T is the S / N ratio of the combined histogram H _T. Therefore, in the present embodiment, after obtaining the individual histogram H _i and the combined histogram H _T by the above-described method, the individual reliability C _i and the combined reliability C _T are calculated. Then, the calculated individual reliability C _i and the combined reliability C _T are compared, and it is determined which of the combined histogram H _T and the individual histogram H _i is used as the output of the TOF signal processing unit 100.

Hereinafter, the method for improving the S / N ratio of the receiving circuit according to the present embodiment will be described more specifically. First, the individual reliability C _i and the combined reliability C _T can be calculated in the same manner as in the above embodiment. That is, pixels near the individual histograms _{H i,} and pixel _i of the pixel _i is a TOF pixel GT _i of interest in the signal light detection unit 130, the pixel _i-2, pixel _i-1, pixel _{i + 1,} and pixel _{i + 2} separate The histograms H _i-2 , H _i−1 , H _{i + 1} , and H _{i + 2} can be obtained as shown in FIG. However, in the present embodiment, since the observation objects exist at the same distance, H _i−2 , H _i−1 , H _i , H _{i + 1} , and H _{i + 2} are observed at the same time.

The combined histogram H _T is obtained by adding and integrating the individual histograms H _i−2 , H _i−1 , H _i , H _{i + 1} , and H _{i + 2} without correcting the time difference as shown in FIG. 6B. be able to. If the observation target of each pixel is the above-mentioned obstacle or the like, the time at which the peak value of the individual histogram H _i corresponding to each pixel is added is added and integrated, so the combined histogram H _T has a larger peak. Form a value.

Now, as in the above embodiment, the pixel of interest as a pixel _i, the pixel _i, consider the case of synthesizing the pixel _{i + 1} in the vicinity of the pixel _i. The disturbance light component after the synthesis can be obtained from the sum of the histogram outputs of the pixel _i and the pixel _{i + 1} by the same method as in the above embodiment. The combined signal component can be obtained from the peak value of the sum of the histogram outputs of pixel _i and pixel _{i + 1} . As an example, consider a case where the peak value of the individual histogram of pixel _i and pixel _{i + 1} is P, the disturbance light component is N, and the noise component is the square root of N. The individual reliability C _{i in} this case is given by (Equation 9) shown below using (Equation 4).

On the other hand, the disturbance light component after the synthesis of the individual histograms H _i is 2N, the noise component is the square root of 2N. When the pixel _i and the pixel _{i + 1} observes an object of the same distance, since the peak of the individual histograms H _i before combining as described above are formed at the same time, the peak value after synthesis becomes 2P is added. Therefore, synthetic reliability C _TO when the observation target is present in the same distance is given by the following with reference to Equation (5) (Equation 10).

That is, synthetic reliability C _TO when the observation target are in the same distance, the square root of two times the individual reliability C _i.

On the other hand, when the pixel _i and the pixel _{i + 1} observe objects of different distances, the peaks of the individual histograms H _i before synthesis are formed at different times, so that the peak values after synthesis are not added, but the magnitude P + N Two peaks are formed at different times. However, in this case as well, the disturbance light component is added to 2N, and the noise component is the square root of 2N. Signal component is a P-N obtained by subtracting the disturbance light component 2N from the peak value P + N, the same as the individual histogram H _i.
Therefore, the combined reliability _CTR when the observation target exists at different distances is given by (Equation 11) shown below.

That is, the combined reliability _CTR when the observation target is present at different distances is a value obtained by dividing the individual reliability C _i by the square root of 2.

As described above, the present embodiment uses the fact that the value of the combined reliability C _{T with} respect to the individual reliability C _i , that is, the value of the S / N ratio, differs according to the distance of the observation target. That is, synthetic reliability C _TO when the observation target is in the same distance is larger than the individual reliability C _i, synthetic reliability C _TR when the observation target is in a different distance be less than the individual reliability C _i Is used.

In the present embodiment, the case where the individual histograms H _i at the same time are combined when the observation target is present at the same distance has been described as an example. However, the individual reliability C _i and the combined reliability C _T are described above. The same relationship holds true even when the individual histograms _Hi are synthesized by correcting the time difference based on the difference in depression angle when the observation objects exist at different distances. However, as described above, the magnitude relationship between the combined reliability _CTO and _{CTR in} this case and the individual reliability C _i is reversed.

In the present embodiment, when the synthetic reliability C _T is less than the individual reliability C _i outputs the individual histograms H _i. By performing such determination processing, the S / N ratio of the signal output from the receiving circuit does not become at least smaller than the individual reliability C _i . On the other hand, when the composite reliability C _T is larger than the individual reliability C _i , the composite histogram H _T is output. By performing such a determination process, it is possible to determine a case where a larger S / N ratio can be expected, and to further improve the S / N ratio.

As described above in detail, according to the method for improving the S / N ratio of the receiving circuit according to the present embodiment, the reliability of peak value detection, that is, the S / N ratio, when the individual histograms are combined or not. by comparing the ratio, it is possible to properly determine the appropriateness of synthesis of individual histograms H _i, received signal always optimum S / N ratio can be obtained.

Here, in the present embodiment, the mode of synthesizing the target pixel and one neighboring pixel has been described as an example. However, the present invention is not limited to this, and a plurality of neighboring pixels may be synthesized. In this case, in pairs of the pixel of interest and neighboring pixels, it may be determined the appropriateness of synthesis of individual histograms H _i, also possible to determine the suitability of the synthesis of the target pixel and all the neighboring pixels at the same time it can.

  Further, in the present embodiment, an example in which a plurality of different pixels on the light receiving LSI 12 are combined has been described as an example. However, the present invention is not limited to this, and outputs of the same pixel on the light receiving LSI 12 at different times can be combined. . The pixels on the light receiving LSI 12 are arranged in a row in the vertical direction in the observation target space, whereas outputs at successive different times during scanning in the horizontal direction are arranged in the horizontal direction. Therefore, the above synthesis corresponds to the synthesis of outputs of adjacent pixels in the horizontal direction.

In each of the above-described embodiments, a horizontal plane road surface is assumed as the road surface, and the embodiment in which the present invention is applied is described for the case where the depression angle does not vary with time. However, the present invention is not limited to this, and the gradient and curvature are not limited thereto. Assuming a plurality of road surface patterns having different angles, the present invention may be applied to a case where the depression angle varies with time. In this case, based on the peak value formation time t _i obtained for each of the plurality of road patterns assumed, summing the individual histogram of neighboring pixels of the individual histograms and target pixel of the target pixel, integrating to generate a composite histogram. A plurality of additions corresponding to the plurality of road surface patterns are compared with the reliability of the integrated composite histogram, and the road surface pattern having the highest reliability is adopted. As a result, it is possible to detect road surfaces flexibly corresponding to road surfaces of various shapes.

Furthermore, it is possible to adopt an embodiment of the present invention that takes into consideration the change in depression angle associated with the vertical movement of a moving body (vehicle or the like) on which the optical distance measuring device according to each of the above embodiments is mounted.
In this case, based on the peak value formation time t _i obtained for each of the plurality of depression angle obtained by adding the plurality of offset values for positive and negative to depression theta _i, may be employed depression conditions the same highest reliability is increased.

  Further, in each of the above-described embodiments, the configuration in which the light detection unit is configured by the pixels (TOF pixel GT, reference pixel GR) using SiPM composed of 24 SPADs has been described, but the present invention is not limited thereto. The light detection unit may be configured by pixels using SiPM in which the number of SPADs is another appropriate number. Furthermore, a single SPAD can constitute a pixel, and even in this case, generation of an individual histogram and generation (that is, correction) of a combined histogram are possible.

  Further, in each of the above-described embodiments, the configuration in which the light receiving unit is configured by the Geiger mode SPAD has been described as an example. However, the present invention is not limited thereto, and the light receiving unit is configured by a linear mode APD, The time series signal obtained from the above may be used.

  In each of the above embodiments, the signal light detection unit and the reference light detection unit have been described by way of example with the same element structure. However, the present invention is not limited to this, and the characteristics required for each light detection unit. For example, the signal light detection unit and the reference light detection unit may have different element structures. Further, the light receiving mode (Geiger mode or linear mode) may be made different between the signal light detection unit and the reference light detection unit. Furthermore, it is possible to configure the light detection unit using only the signal light detection unit without using the reference light detection unit. In this case, the disturbance light level can be obtained from the light reception signal from the signal light detection unit in the idle period when the irradiation light is not projected.

  Further, in each of the above embodiments, the configuration in which the light detection unit and the signal processing unit are integrated has been described as an example. However, the present invention is not limited to this, and the light detection unit and the signal processing unit are configured as separate semiconductor integrated circuits. It may be formed.

  In each of the above-described embodiments, the mode in which the observation direction of the reference light detection unit is set as the observation direction of the next signal light detection unit has been described as an example. It is good. Furthermore, the observation direction of the reference light detection unit may be the observation direction of the previous signal light detection unit.

  Further, in each of the above embodiments, the signal processing of the received light signal has been described by exemplifying the form executed by the TOF signal processing unit 100 and the reference signal processing unit 200. However, the present invention is not limited to this, and the function of each signal processing unit Alternatively, a part of the program may be described as a program by software, and the program may be executed by a CPU of a control unit (not shown). In this case, the program may be stored in a ROM (not shown), and the CPU may be loaded into a RAM (not shown) and executed.

DESCRIPTION OF SYMBOLS 1 Vehicle 6 Road 7 Roadside 8 Building 10 Optical distance measuring device 12 Light receiving LSI
13 laser scanner 14 light receiving substrate 16 light projecting substrate 18 LD element 20 hyperboloid mirror 22 polygon mirror 23 irradiation pulse 24 power supply substrate 27 control substrate 25 reflection pulse 26 housing 29 irradiation beam 30 light projecting unit 32 light receiving unit 35 reflecting surface 50 light Detection unit 100 TOF signal processing unit 102 Front end unit 104 Pulse shaping circuit 106 Adder 108 Comparator 110 TDC
112 Histogram processor 120 SPAD
122 resistor 124 preamplifier 130 signal light detection unit 200 reference signal processing unit 202 front end unit 204 pulse shaping circuit 206 adder 208 comparator 210 counter 220 SPAD
230 Reference light detector A Rotating axis DT, DR Observation direction GT TOF pixel GR Reference pixel K Aperture Lt Light projection light Lr Light reception light Pt Irradiation light pulse Pr Reflection light pulse PK Peak Si point Xi Horizontal distance

Claims (7)

A light projecting unit that projects an irradiation light pulse on the measurement object;
A light receiving unit that is disposed on an imaging surface on which the reflected light pulse reflected by the measurement object is imaged, and that has a plurality of light receiving pixels that receive the reflected light pulse;
Selection of outputting as a plurality of discrete light receiving signals of a plurality of selected light receiving signal output from one or more neighboring pixels of the pixel of interest and the target pixel of the light receiving signal outputted from each of the plurality of light receiving pixels And
A combining unit that outputs a combined light receiving signal obtained by adding the plurality of individual light receiving signals;
Using the synthesized light reception signal, a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a light reception time of the reflected light pulse is calculated, and the measurement object is calculated based on the flight time. a distance calculation unit for calculating the distance to, only including,
The synthesizing unit corrects a difference in light reception time of the reflected light pulse due to different observation directions of the light receiving pixels, adds the plurality of individual light reception signals, and outputs the composite light reception signal . A light projecting unit that projects an irradiation light pulse on the measurement object;
A light receiving unit that is disposed on an imaging surface on which the reflected light pulse reflected by the measurement object is imaged, and that has a plurality of light receiving pixels that receive the reflected light pulse;
Selection of a light reception signal output from one or a plurality of pixels in the vicinity of the target pixel and a plurality of light reception signals output from each of the plurality of light reception pixels, and output as a plurality of individual light reception signals And
A combining unit that outputs a combined light receiving signal obtained by adding the plurality of individual light receiving signals;
Using the synthesized light reception signal, a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a light reception time of the reflected light pulse is calculated, and the measurement object is calculated based on the flight time. A distance calculation unit for calculating the distance to
Based on the ratio between the signal and the noise, the individual reliability that is the reliability of the peak detection of the individual received light signal is calculated, and based on the ratio between the signal and the noise, the reliability of the peak detection of the combined received light signal is calculated. includes a reliability calculation unit which calculates a certain synthetic reliability and,
The combining unit outputs the combined light reception signal when the combined reliability is greater than the individual reliability, and when the combined reliability is equal to or less than the individual reliability, the individual corresponding to the target pixel. An optical distance measuring device that outputs a light reception signal as the combined light reception signal. Based on the ratio between the signal and the noise, the individual reliability that is the reliability of the peak detection of the individual received light signal is calculated, and based on the ratio between the signal and the noise, the reliability of the peak detection of the combined received light signal is calculated. A reliability calculation unit for calculating a certain composite reliability;
The combining unit outputs the combined light reception signal when the combined reliability is greater than the individual reliability, and when the combined reliability is equal to or less than the individual reliability, the individual corresponding to the target pixel. The optical distance measuring device according to claim 1 , wherein a light reception signal is output as the combined light reception signal. When the composite reliability is greater than the individual reliability, it is determined that the measurement object is a road surface, and when the composite reliability is equal to or less than the individual reliability, the measurement object is a non-road surface. The optical distance measuring device according to claim 3 , further comprising: a determination unit that determines that. The light receiving unit includes a disturbance light receiving unit including a light receiving pixel that receives only disturbance light not including the reflected light pulse,
The reliability calculation unit is configured and the signal peak value of the photodetection signal, as claimed in any one of claims 2 to 4 for calculating the individual reliability the output of the disturbance light receiving section as noise Optical distance measuring device. A light projecting unit that projects an irradiation light pulse on the measurement object;
  A light receiving unit that is disposed on an imaging surface on which the reflected light pulse reflected by the measurement object is imaged, and that has a plurality of light receiving pixels that receive the reflected light pulse;
  Selection of a light reception signal output from one or a plurality of pixels in the vicinity of the target pixel and a plurality of light reception signals output from each of the plurality of light reception pixels, and output as a plurality of individual light reception signals And
  A combining unit that outputs a combined light receiving signal obtained by adding the plurality of individual light receiving signals;
  Using the synthesized light reception signal, a flight time that is a difference between a projection time of the irradiation light pulse corresponding to the target pixel and a light reception time of the reflected light pulse is calculated, and the measurement object is calculated based on the flight time. A distance calculation unit that calculates the distance to
  The light projecting unit repeatedly projects the irradiation light pulse on the measurement object over a predetermined measurement time,
  The selection unit outputs an individual histogram obtained by accumulating light reception signals of the irradiation light pulse repeatedly projected along a time axis, which are output from the pixel of interest and each of one or a plurality of pixels in the vicinity of the pixel of interest. Output as the individual light reception signal,
  The combining unit outputs a combined histogram obtained by adding the individual histograms as the combined received light signal,
  The distance calculation unit calculates the distance to the measurement object using the time corresponding to the peak value of the combined histogram as the light reception time.
  Optical distance measuring device. The light projecting unit repeatedly projects the irradiation light pulse on the measurement object over a predetermined measurement time,
The selection unit outputs an individual histogram obtained by accumulating light reception signals of the irradiation light pulse repeatedly projected along a time axis, which are output from the pixel of interest and each of one or a plurality of pixels in the vicinity of the pixel of interest. Output as the individual light reception signal,
The combining unit outputs a combined histogram obtained by adding the individual histograms as the combined received light signal,
The optical distance measurement according to any one of claims 1 to 5 , wherein the distance calculation unit calculates a distance to the measurement object using a time corresponding to a peak value of the combined histogram as the light reception time. apparatus.