JP2018054426A - Information processing device, optical apparatus, control method, program, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device capable of suitably reducing noise superposed on an output signal from a light-receiving unit.SOLUTION: A lidar 1 includes: a scan unit 55 for emitting emission light Lo corresponding to a pulse trigger signal PT; an APD 41 for receiving return light Lr of the emission light Lo; and a DSP 16. The DSP 16 includes a synchronization obstruction estimation unit 71 that extracts a reception segment v where an amount of the return light Lr is estimated to be equal to or less than a prescribed value on the basis of current position information IP and map information IM, and calculates estimated synchronization obstruction w, i.e. a noise signal due to the pulse trigger signal PT or the like from the reception segment v.SELECTED DRAWING: Figure 11

Description

  The present invention relates to an information processing technique for an optical device using laser light.

  Conventionally, as a distance measuring device using light, it is configured to emit pulsed light to the measurement object and measure the distance to the measurement object based on the light reception timing of the pulsed light reflected by the measurement object What has been made is widely used. For example, in Patent Document 1, a trigger signal is generated by a control unit, pulse light is emitted from a trigger signal to a measurement object, and pulse light reflected by the measurement object is received by a light receiving unit. A distance measuring device that generates a signal is disclosed.

JP 2007-256191 A

  The trigger signal that defines the timing of emitting pulsed light is very large with respect to the output level of the light receiving unit, so noise caused by high frequency components included in the rising and falling edges of the trigger signal is superimposed on the output signal of the light receiving unit. Cause an error.

  The present invention has been made to solve the above-described problems, and a main object of the present invention is to provide an information processing apparatus that can suitably reduce noise superimposed on an output signal of a light receiving unit. And

  The invention described in claim is the light reception of an optical device comprising: an irradiation unit that irradiates a laser beam in response to a first signal; and a light receiving unit that receives the return light reflected from the object by the laser beam. An information processing apparatus that processes an output signal of a unit, comprising: an estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal .

  The invention described in claim 5 processes an output signal of the light receiving unit of an optical apparatus including an irradiation unit that emits laser light in accordance with a first signal and a light receiving unit that receives return light of the laser light. A first acquisition unit that acquires current position information and map information including position information of a feature, and the output signal based on the current position information and the map information. And an estimation unit that estimates a noise signal caused by one signal.

  The invention described in claim is an optical apparatus, wherein an irradiation unit that irradiates a laser beam according to a first signal, a light receiving unit that receives a return light reflected from an object, and the laser beam, An estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is equal to or less than a predetermined value in the output signal.

  The invention described in claim is executed by an optical device including an irradiation unit that irradiates a laser beam in response to a first signal, and a light receiving unit that receives the return light reflected from the object by the laser beam. A control method, comprising: an estimation step of estimating a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal.

  The invention described in claim is the light reception of an optical device comprising: an irradiation unit that irradiates a laser beam in response to a first signal; and a light receiving unit that receives the return light reflected from the object by the laser beam. An estimation unit for estimating a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal, which is a program executed by a computer that processes the output signal of the unit To make the computer function.

1 shows an overall configuration of a rider according to an embodiment. The structure of a transmitter and a receiver is shown. The structure of a scanning optical part is shown. The register setting example of the control signal which a synchronous control part produces | generates is shown. The time relationship of the control signal which a synchronous control part produces | generates is shown. It is a graph which shows the relationship between an ADC output signal and a gate. The time relationship of the pulse train of a rotary encoder is shown. Shows the time relationship between encoder pulse and segment slot in steady state The relationship between a pulse trigger signal and a received segment signal is shown. It is a bird's-eye view around the vehicle carrying a rider. The block diagram of the signal processing which DSP performs is shown. The waveform of the reference | standard reception pulse and impulse response in one segment period is shown. It is an example of a filtered segment. It is a bird's-eye view around the vehicle carrying a rider.

  According to a preferred embodiment of the present invention, an optical apparatus comprising: an irradiating unit that irradiates a laser beam in response to a first signal; and a light receiving unit that receives the return light reflected from the object by the laser beam. An information processing apparatus for processing an output signal of the light receiving unit, wherein the noise signal caused by the first signal is estimated based on a dark output signal in which the return light is not more than a predetermined value in the output signal A part.

  The information processing apparatus outputs an output signal of a light receiving unit of an optical apparatus, which includes an irradiation unit that emits laser light according to a first signal, and a light receiving unit that receives return light reflected from the object. And includes an estimation unit. The estimation unit estimates a noise signal caused by the first signal based on a dark output signal in which the return light is not more than a predetermined value in the output signal of the light receiving unit. According to this aspect, the information processing apparatus receives the noise signal superimposed on the output signal of the light receiving unit due to the high frequency component of the first signal that defines the timing at which the irradiation unit emits the laser light. Therefore, it can be accurately estimated based on a dark output signal that is an output signal of the light receiving unit when no return light is generated or when the amount of the return light is extremely small.

  In one aspect of the information processing apparatus, the information processing apparatus further includes a first acquisition unit that acquires current position information and map information including position information of a feature, and the estimation unit includes the current position information and the Based on the map information, the dark output signal is identified from the output signal. According to this aspect, the information processing apparatus appropriately recognizes the relative position of the feature based on the map information, and uses the output signal of the light receiving unit obtained when there is no feature reflecting the laser light as the dark output signal. It can specify suitably.

  In another aspect of the information processing apparatus, the estimation unit identifies the output signal at a specific position where the feature does not exist within a predetermined range from the position indicated by the current position information as the dark output signal. According to this aspect, the information processing apparatus can suitably specify the output signal of the light receiving unit obtained at a specific position where there is no feature reflecting the laser light as the dark output signal.

  In another aspect of the information processing apparatus, the estimation unit specifies a specific direction in which the feature does not exist within a predetermined range from the position indicated by the current position information, and the irradiation unit includes the specific direction. The output signal indicating the return light of the laser beam irradiated on is identified as the dark output signal. According to this aspect, the information processing apparatus specifies the output signal of the light receiving unit corresponding to the specific direction where the feature does not exist as a dark output signal even when the feature exists within a predetermined distance, and performs noise estimation Can be suitably performed.

  In another aspect of the information processing apparatus, the information processing apparatus further includes a subtraction unit that subtracts the noise signal estimated by the estimation unit from the output signal of the light receiving unit. According to this aspect, the information processing apparatus can suitably generate the output signal of the light receiving unit from which the noise signal due to the first signal has been removed.

  In another aspect of the information processing apparatus, the estimation unit estimates the noise signal by averaging the dark output signals obtained when the irradiation unit scans a plurality of times. According to this aspect, the information processing apparatus can estimate the noise signal with high accuracy.

  In another aspect of the information processing apparatus, the first signal is a pulse signal, and the irradiation unit irradiates pulse light as the laser light for each different irradiation direction based on the pulse signal. In this aspect, the information processing apparatus can preferably estimate a noise signal due to a high-frequency component included in the rising and falling edges of the pulse signal that defines the irradiation timing of the laser beam of the irradiation unit.

  According to a preferred embodiment of the present invention, an output of the light receiving unit of an optical apparatus comprising: an irradiation unit that irradiates laser light according to a first signal; and a light receiving unit that receives return light of the laser light. An information processing apparatus for processing a signal, wherein the output signal is based on a first acquisition unit that acquires current position information and map information including position information of a feature, and the current position information and the map information. To an estimation unit for estimating a noise signal caused by the first signal. According to this aspect, the information processing apparatus can appropriately grasp the relative positional relationship between the features registered in the map information, and can accurately estimate the noise signal caused by the first signal from the output signal of the light receiving unit. .

  According to another preferred embodiment of the present invention, the optical device includes an irradiation unit that irradiates a laser beam in response to the first signal, and a light receiving unit that receives the return light reflected from the object. And an estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal. According to this aspect, the optical apparatus accurately estimates the noise signal superimposed on the output signal of the light receiving unit due to the high frequency component of the first signal that defines the timing at which the irradiation unit irradiates the laser beam. Can do.

  According to another preferred embodiment of the present invention, an irradiating unit that irradiates a laser beam in response to a first signal, and a light receiving unit that receives the return light reflected from the object by the laser beam. A control method executed by an optical apparatus, comprising: an estimation step of estimating a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal. By executing this control method, the optical apparatus performs a noise signal superimposed on the output signal of the light receiving unit due to the high frequency component of the first signal that defines the timing at which the irradiation unit irradiates the laser beam, It can be estimated accurately.

  According to another preferred embodiment of the present invention, an irradiating unit that irradiates a laser beam in response to a first signal, and a light receiving unit that receives the return light reflected from the object by the laser beam. A noise signal caused by the first signal based on a dark output signal in which the return light is not more than a predetermined value in the output signal, which is a program executed by a computer that processes the output signal of the light receiving unit of the optical device The computer is caused to function as an estimation unit that estimates. By executing this program, the computer accurately outputs a noise signal superimposed on the output signal of the light receiving unit due to the high frequency component of the first signal that defines the timing at which the irradiation unit irradiates the laser beam. Can be estimated. Preferably, the program is stored in a storage medium.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

<Basic explanation>
First, a basic configuration of the rider according to the embodiment will be described.

(1) Overall Configuration FIG. 1 shows the overall configuration of a rider according to an embodiment. The lidar 1 scans the surrounding space by appropriately controlling the emission direction of the light pulse that is repeatedly emitted (hereinafter referred to as “scanning direction”), and observes the return light to thereby detect an object existing in the vicinity. Information about the distance (for example, distance, existence probability or reflectance). Specifically, the lidar 1 emits a light pulse (hereinafter referred to as “emitted light”) Lo and is reflected by an external object (target) (hereinafter referred to as “returned light”) Lr. Is received to generate information about the object. The lidar 1 is an example of the “optical device” in the present invention.

  As shown in FIG. 1, the lidar 1 roughly includes a system CPU 5, an ASIC 10, a transmitter 30, a receiver 40, and a scanning optical unit 50. The transmitter 30 repeatedly outputs a laser light pulse having a width of about 5 nsec in accordance with the pulse trigger signal PT supplied from the ASIC 10. The light pulse output from the transmitter 30 is guided to the scanning optical unit 50.

  The scanning optical unit 50 emits the light pulse output from the transmitter 30 in an appropriate direction, and collects the return light Lr that is returned when the emitted light encounters an object in space and is reflected or scattered. To the receiver 40. The scanning optical unit 50 is an example of the “irradiation unit” in the present invention. The receiver 40 outputs a signal proportional to the intensity of the return light Lr to the ASIC 10. The receiver 40 is an example of the “light receiving unit” in the present invention.

  The ASIC 10 analyzes the output signal of the receiver 40 to estimate and output a parameter related to the object in the scanning space, for example, the distance. Further, the ASIC 10 controls the scanning optical unit 50 so that appropriate scanning is performed. Further, the ASIC 10 supplies the transmitter 30 and the receiver 40 with the high voltages required by each.

  The system CPU 5 performs initial setting, monitoring, and control of the ASIC 10 through at least a communication interface. Other functions differ depending on the application. In the case of the simplest lidar, the system CPU 5 only converts the target information TI output from the ASIC 10 into an appropriate format and outputs it. For example, the system CPU 5 converts the target information TI into a highly versatile point cloud format, and then outputs it through the USB interface.

(2) Transmitter The transmitter 30 outputs an optical pulse having a width of about 5 nsec in response to the pulse trigger signal PT supplied from the ASIC 10. The structure of the transmitter 30 is shown in FIG. The transmitter 30 includes a charging resistor 31, a driver circuit 32, a capacitor 33, a charging diode 34, a laser diode (LD) 35, and a CMOS switch 36.

The pulse trigger signal PT input from the ASIC 10 drives a switch 36 such as a CMOS via the driver circuit 32. The driver circuit 32 is inserted to drive the switch 36 at high speed. During the non-assertion period of the pulse trigger signal PT, the switch 36 is open, and the capacitor 33 in the transmitter 30 is charged with the high voltage V _TX supplied from the ASIC 10. On the other hand, in the assertion period of the pulse trigger signal PT, the switch 36 is closed, and the charge charged in the capacitor 33 is discharged through the LD 35. As a result, an optical pulse is output from the LD 35.

(3) Receiver The receiver 40 outputs a voltage signal proportional to the intensity of the return light Lr from the object. In general, a photodetection element such as a PD or APD has a current output, so the receiver 40 converts this current into a voltage (I / V conversion) and outputs it. The structure of the receiver 40 is shown in FIG. The receiver 40 includes an APD (Avalanche Photodiode) 41, an I / V conversion unit 42, a resistor 45, a capacitor 46, and a low-pass filter (LPF) 47. The I / V conversion unit 42 includes a feedback resistor 43 and an operational amplifier 44.

In the present embodiment, APD 41 is used as the light detection element. A high voltage V _RX supplied from the ASIC 10 is applied to the APD 41 as a reverse bias, and a detection current proportional to the return light Lr from the object flows. By applying a reverse bias close to the breakdown voltage of the APD 41, a high avalanche gain can be obtained and a weak return light can be detected. The LPF 47 at the final stage is installed for the purpose of limiting the signal bandwidth prior to sampling by the ADC 20 in the ASIC 10. In this embodiment, the sampling frequency of the ADC 20 is 512 MHz, and the cutoff frequency of the LPF 47 is about 250 MHz.

(4) Scanning optical unit The scanning optical unit 50 emits the light pulse input from the transmitter 30 as an emitted light Lo in an appropriate direction, and the emitted light Lo encounters an object in the space and is reflected or scattered. The return light Lr returned by this is guided to the receiver 40. A configuration example of the scanning optical unit 50 is shown in FIG. The scanning optical unit 50 includes a rotary mirror 61, a collimator lens 62, a condenser lens 64, an optical filter 65, a coaxial mirror 66, and a rotary encoder 67.

  The light pulse output from the LD 35 of the transmitter 30 enters the collimator lens 62. The collimator lens 62 collimates the laser light at an appropriate divergence angle (generally, about 0 to 1 °). Light emitted from the collimator lens 62 is reflected vertically downward by a small coaxial mirror 66 and enters the rotation axis (center) of the rotary mirror 61. The rotating mirror 61 reflects the laser beam incident from vertically above in the horizontal direction and emits it to the scanning space. The rotating mirror 61 is attached to the rotating portion of the motor 54, and the laser beam reflected by the rotating mirror 61 scans the horizontal plane as the emitted light Lo as the motor 54 rotates.

  The return light Lr returned to the lidar 1 by being reflected or scattered by an object existing in the scanning space is reflected vertically upward by the rotating mirror 61 and enters the optical filter 65. In addition to the return light Lr, background light generated when the object is illuminated by the sun or the like also enters the optical filter 65. The optical filter 65 is installed to selectively exclude such background light. Specifically, the optical filter 65 selectively allows only a component of about ± 10 nm before and after the wavelength of the emitted light Lo (905 nm in this embodiment). When the pass band of the optical filter 65 is wide, a lot of background light enters the receiver 40 at the subsequent stage. As a result, a large DC current component appears in the output of the APD 41 in the receiver 40, and the SN deteriorates due to the influence of shot noise (background light shot noise) caused by this DC component, which is not preferable. However, when the pass band is excessively narrow, the emitted light itself is suppressed, which is not preferable. The condensing lens 64 condenses the light that has passed through the optical filter 65 and guides it to the APD 41 of the receiver 40.

  A rotary encoder 67 is attached to the motor 54 in order to detect the scanning direction. The rotary encoder 67 includes a turntable 68 attached to the motor rotating unit and a code detector 69 attached to the motor base. A slit indicating the rotation angle of the motor 54 is formed on the outer periphery of the rotating disk 68, and the code detector 69 reads and outputs this. Note that specific specifications of the rotary encoder 67 and motor control based on the output will be described later.

  In the above configuration, the collimator lens 62 constitutes the transmission optical system 51 shown in FIG. 1, the rotating mirror 61 and the motor 54 constitute the scanning unit 55 shown in FIG. 1, and the optical filter 65 and the condenser lens 64 are shown in FIG. 1 and the rotary encoder 67 constitutes the scanning direction detector 53 in FIG.

(5) ASIC
The ASIC 10 performs timing control of the emitted light pulse, AD conversion of the APD output signal, and the like. Further, the ASIC 10 estimates parameters (distance, return light intensity, etc.) related to the object by performing appropriate signal processing on the AD conversion output, and outputs the estimation result to the outside. As shown in FIG. 1, the ASIC 10 includes a register unit 11, a clock generation unit 12, a synchronization control unit 13, a gate extraction unit 14, a reception segment memory 15, a DSP 16, and a transmitter high voltage generation unit (TXHV). ) 17, a receiver high voltage generation unit (RXHV) 18, a preamplifier 19, an AD converter (ADC) 20, and a scanning control unit 21.

  The register unit 11 includes a register for communication with the system CPU 5 which is an external processor. The registers provided in the register unit 11 are roughly classified into R registers that can only be referenced from the outside and W registers that can be set from the outside. The R register mainly holds status values inside the ASIC, and the system CPU 5 can monitor the internal status of the ASIC 10 by reading these values through the communication interface. On the other hand, the W register holds various parameter values referred to inside the ASIC 10. These various parameter values can be set from the system CPU 5 through the communication interface. Note that the communication register may be realized by a flip-flop or a RAM.

  The clock generation unit 12 generates a system clock SCK and supplies it to each block in the ASIC 10. Many blocks of the ASIC 10 operate in synchronization with the system clock SCK. In this embodiment, the frequency of the system clock SCK is 512 MHz. The system clock SCK is generated by a PLL so as to be synchronized with a reference clock RCK input from the outside. Usually, a crystal oscillator is used as a generation source of the reference clock RCK.

  The TXHV 17 generates a DC high voltage (about 100 V) required by the transmitter 30. The high voltage is generated by boosting a low voltage (about 5V to 15V) by a DCDC converter circuit.

  The RXHV 18 generates a DC high voltage (about 100 V) required by the receiver 40. The high voltage is generated by boosting a low voltage (about 5V to 15V) by a DCDC converter circuit.

  The synchronization control unit 13 generates and outputs various control signals. The synchronization control unit 13 in this embodiment outputs two control signals, that is, a pulse trigger signal PT and an AD gate signal GT. An example of setting these control signals is shown in FIG. 4, and their temporal relationship is shown in FIG. As shown in FIG. 5, these control signals are generated in synchronization with time intervals (segment slots) divided at a predetermined interval. The time interval width (segment period) of the segment slot can be set by “nSeg”. In the present embodiment, it is assumed that “nSeg = 8192” is set in a range not specifically mentioned.

  The pulse trigger signal PT is supplied to a transmitter 30 provided outside the ASIC 10. The transmitter 30 outputs an optical pulse according to the pulse trigger signal PT. For the pulse trigger signal PT, the delay “dTrg” and the pulse width “wTrg” with respect to the segment slot start point can be set. Note that the pulse width wTrg is determined in view of the trigger response specification of the transmitter 30 because the transmitter 30 does not respond if it is too narrow.

  The AD gate signal GT is supplied to the gate extraction unit 14. As will be described later, the gate extraction unit 14 extracts only the asserted period of the AD gate signal GT from the ADC output signal input from the ADC 20 and stores it in the reception segment memory 15. For the AD gate signal GT, the delay time “dGate” and the gate width “wGate” with respect to the segment slot start point can be set. Here, as the gate width wGate is longer, the maximum distance measurement distance (range measurement limit distance) of the lidar 1 becomes longer.

  The preamplifier 19 amplifies the analog voltage signal input from the receiver 40 installed outside the ASIC 10 and supplies the amplified voltage to the subsequent ADC 20. The voltage gain of the preamplifier 19 can be set by the W register.

  The ADC 20 performs AD conversion on the output signal of the preamplifier 19 to convert it into a digital series. In this embodiment, the system clock SCK is used as the sampling clock of the ADC 20, and the input signal of the ADC 20 is sampled at 512 MHz.

  The gate extraction unit 14 extracts only the asserted period of the AD gate signal GT from the ADC output signal input from the ADC 20 and stores it in the reception segment memory 15. The section signal extracted by the gate extraction unit 14 is hereinafter referred to as “reception segment signal RS”. That is, the reception segment signal RS is a real vector whose vector length is equal to the gate width wGate.

Here, the relationship between the ADC output signal and the reception segment and the setting of the gate position will be described. FIG. 6A shows a segment slot. As shown in FIG. 6B, the pulse trigger signal PT is asserted with a delay of dTrg with respect to the segment slot start point. In the example of FIG. 6, since dTrg = 0, the pulse trigger signal PT is asserted at the segment slot start point. FIG. 6C shows an ADC output signal (received segment signal RS) when an object is placed at the scanning origin of the lidar. That is, FIG. 6C illustrates the received segment signal RS when the target distance (radial radius R) is 0 m. As shown in the figure, even when R = 0 m, the rising edge of the received pulse is observed with a delay of the system delay D _SYS from the rising edge of the pulse trigger signal. The system delay D _SYS is caused by the electrical delay of the LD driver circuit in the transmitter 30, the optical delay in the transmission optical system 51, the optical delay in the reception optical system 52, and the electrical delay in the receiver 40. A delay, a conversion delay in the ADC 20, and the like can be considered.

  FIG. 6D illustrates the received segment signal RS when the object is placed on the moving radius R. In this case, as compared with FIG. 6C, the delay increases by the round-trip time of light from the scanning origin to the object. This increased delay is a so-called “TOF (Time Of Flight) delay”. If this TOF delay is D samples, the radius R can be calculated by the following equation.

FIG. 6F illustrates the AD gate signal GT when dGate = 0. As described above, the gate extraction unit 14 extracts only the assert period of the AD gate signal GT from the ADC output signal. The DSP 16, which will be described later, performs parameter estimation on the object based only on this extraction section. Therefore, when the TOF delay time is long, the return pulse component from the object protrudes from the gate, and a valid parameter cannot be estimated. In order to perform proper parameter estimation, the TOF delay time D needs to satisfy the following equation.

Here, L _IR is the length of the overall impulse response of the system, and D _MAX is defined as the maximum TOF delay time that allows valid parameter estimation. FIG. 6E illustrates the received segment signal RS when the TOF delay time is equal to the maximum TOF delay time.

  Instead of the example of FIG. 6, the gate delay dGate may be set equal to the system delay time. By setting in this way, it is possible to estimate a valid parameter up to an object at a longer distance.

  The scanning control unit 21 monitors the output of the rotary encoder 67 installed outside the ASIC 10, and controls the rotation of the motor 54 based on this. Specifically, the scanning control unit 21 supplies the torque control signal TC to the motor 54 based on the scanning direction information SDI output from the rotary encoder 67 (scanning direction detection unit 53) of the scanning optical unit 50. The rotary encoder 67 in the present embodiment outputs two pulse trains of A phase and Z phase (hereinafter referred to as “encoder pulses”). The time relationship between both pulse trains is shown in FIG. As shown in the figure, for the A phase, one pulse is generated and output every 1 ° of rotation of the motor 54. Therefore, 360 A-phase encoder pulses are generated and output every rotation of the motor 54. On the other hand, for the Z phase, one pulse per rotation of the motor 54 is generated and output corresponding to a predetermined rotation angle.

  The scanning control unit 21 measures the rise time of the encoder pulse as a counter value of the system clock SCK, and controls the torque of the motor 54 so that this becomes a desired value. That is, the scanning control unit 21 performs PLL control of the motor 54 so that the encoder pulse and the segment slot have a desired time relationship.

  The time relationship between the encoder pulse and the segment slot can be set by the W register shown in FIG. In “nPpr”, the number of A-phase encoder pulses for each motor rotation is set. This is a value determined by the specification of the rotary encoder 67. In the present embodiment, the above-described 360 is set. “NRpf” gives the number of rotations for each frame, and “nSpf” gives the number of segments for each frame. “DSmpA” and “dSmpZ” are prepared for adjusting the time relationship between the rising edge of the encoder pulse and the segment slot in units of sample clocks, and can define the delay of the encoder pulse with respect to the segment slot start point. . On the other hand, “dSegZ” is prepared for adjusting the time relationship between the rise of the Z-phase pulse and the frame in units of segments.

  FIG. 8 shows the time relationship between the encoder pulse and the segment slot in the steady state. As shown in the figure, in the default setting, one frame is composed of 1800 segments, and the motor 54 makes one rotation in one frame.

(6) DSP
The DSP 16 sequentially reads the received segments y _{frm and seg} from the received segment memory 15 and processes them. Here, “frm” is a frame index, and “seg” is a segment index. Hereinafter, the description of these indexes is omitted within a range where there is no risk of misunderstanding. The reception segment y is a real vector having a vector length wGate and is represented by the following expression.

The DSP 16 is supplied with the current position information IP and the map information IM from the outside or the system CPU 5. Here, the current position information IP is information indicating the current position such as the latitude and longitude of the rider 1, and may be position information output by a GPS receiver or the like installed in the same vehicle as the rider 1, for example. Alternatively, the position information may be position information estimated by a known self-position estimation process using the output of an external sensor other than the target information TI or the lidar 1.

  The map information IM is map information around the current position of the rider 1, and includes at least feature information around the current position. The feature information corresponds to, for example, position information or shape information of a feature existing around the road (that is, a reflecting object that reflects the emitted light Lo). Here, the DSP 16 may extract the map information IM from the map database stored in the lidar 1, and the map is transmitted from the in-vehicle device existing in the same vehicle as the lidar 1 or the server device existing outside the vehicle via the system CPU 5 or the like. Information IM may be received. The DSP 16 is an example of the “information processing apparatus” in the present invention.

  Here, the reception segment y (that is, the reception segment signal RS) sequentially read out from the reception segment memory 15 by the DSP 16 is disturbed in synchronization with the segment period (electromagnetic jump, influence of current flowing in the ground, etc.) Simply referred to as “synchronization disturbance”). FIG. 9 shows the relationship between the pulse trigger signal PT and the received segment signal RS. The example of FIG. 9 shows the received segment signal RS when the intensity of the return light Lr received by the APD 41 with respect to the emitted light Lo emitted by asserting the pulse trigger signal PT is zero.

  In the example of FIG. 9, high frequency components included in the rising and falling edges of the pulse trigger signal PT are superimposed on the reception segment signal RS. Here, the TTL level of the pulse trigger signal PT is usually 3.3 V or 5 V, and the current level output from the APD 41 is on the order of nA or pA. Therefore, in this case, since the influence of the synchronization interference on the output of the APD 41 is relatively large, the synchronization interference may cause a decrease in the object detection performance and the distance measurement performance of the lidar 1. Considering the above, in the first and second embodiments described below, the DSP 16 accurately extracts the received segment y that is not affected by the return light Lr from the feature based on the current position information IP and the map information IM. Therefore, the synchronization disturbance is preferably estimated.

  The pulse trigger signal PT is an example of the “first signal” in the present invention, and the synchronization disturbance is an example of the “noise signal” in the present invention.

<First embodiment>
First, the first embodiment will be described. Schematically, the DSP 16 averages the received segments y corresponding to the scanning direction determined that there is no detectable feature among the 360-degree scanning directions based on the current position information IP and the map information IM. Thus, the synchronization disturbance is estimated. Then, the DSP 16 subtracts the estimated synchronization disturbance from the reception segment y. Accordingly, the DSP 16 suitably reduces the influence of synchronization interference while enabling object detection in all directions.

  FIG. 10 shows an overhead view of the periphery of the vehicle when the feature 80 exists around the vehicle. Here, in FIG. 10, a specific scanning direction used for estimation of synchronization interference (also simply referred to as “specific scanning direction Dtag”) and an area (“specific” within the maximum distance from the rider 1 in the specific scanning direction Dtag). The scanning direction (also referred to as “scanning area Tag”) and the scanning direction in which the feature exists within the maximum distance measurement distance (also simply referred to as “exclusion scanning direction Dout”) are clearly shown.

  In this case, the DSP 16 first, based on the current position information IP and the position information of the feature 80 included in the map information IM, the excluded scanning direction Dout that is the scanning direction in which the feature 80 exists within the maximum distance. Recognize At this time, the DSP 16 may recognize the excluded scanning direction Dout by acquiring the received segments y for all directions instead of the current position information IP and the map information IM and analyzing the received segments y.

  Then, the DSP 16 sets the specific scanning direction Dtag other than the excluded scanning direction Dout. In other words, the specific scanning direction Dtag indicates a direction in which a feature is predicted not to exist within the maximum distance measurement distance of the lidar 1. In the example of FIG. 10, the DSP 16 sets the direction away from the excluded scanning direction Dout by a predetermined angle as the specific scanning direction Dtag. The specific scanning direction Dtag is not limited to the example of FIG. 10 as long as it is other than the excluded scanning direction Dout. The specific scanning direction Dtag set by the DSP 16 in the first embodiment is an example of the “specific direction” in the present invention.

  As described above, in the first embodiment, the lidar 1 determines the direction other than the direction overlapping the feature as the specific scanning direction Dtag even when the feature exists in the vicinity of the rider 1, and performs the specific scanning. Based on the received segment signal corresponding to the direction Dtag, synchronization interference estimation processing described later is performed. Thereby, the lidar 1 can estimate synchronization interference at an arbitrary timing, and can suitably suppress deterioration in object detection performance and distance measurement performance due to synchronization interference.

  The DSP 1 recognizes the excluded scanning direction Dout, which is the scanning direction in which the feature 80 is present within the maximum distance, and then changes the other direction to the specific scanning direction Dtag to obtain the current position information. Based on the IP and the position information of the feature 80 included in the map information IM, the specific scanning direction Dtag, which is the direction in which the feature does not exist within the maximum distance, is recognized first, and then the other direction is determined. You may make it recognize with the exclusion direction Dout.

  FIG. 11 shows a block diagram of signal processing executed by the DSP 16 in the first embodiment. As shown in FIG. 11, the DSP 16 according to the first embodiment includes a synchronization interference estimation unit 71, a subtracter 72, a reception filter 73, a peak detection unit 74, a determination unit 75, and a formatter 76. . The DSP 16 sequentially reads the received segment y from the received segment memory 15 and performs processing on it.

The synchronization disturbance estimation unit 71 averages the received segments y when the feature does not exist within the specific scanning area Atag, and the averaged received segments y are also referred to as estimated synchronization disturbances (also referred to as “estimated synchronization disturbance w”). .) Is supplied to the subtracter 72. Details of the synchronization disturbance estimation unit 71 will be described later. The subtracter 72 subtracts the estimated synchronization interference w supplied from the synchronization interference estimation unit 71 from the reception segment y, and the reception segment y (also referred to as “corrected reception segment y _dash ”) from which the estimated synchronization interference w is subtracted. Is supplied to the reception filter 73. The synchronization interference estimation unit 71 is an example of the “first acquisition unit” and the “estimation unit” in the present invention, and the subtractor 72 is an example of the “subtraction unit” in the present invention.

The reception filter 73 _convolves a predetermined impulse response “h” with the corrected reception segment y _dash (cyclic convolution) to calculate a filtered segment “z”. The peak detection unit 74 detects a point where the amplitude is maximum in the filtered segment z, that is, a peak point, and outputs a delay “D” and an amplitude “A” of the peak point. The determination unit 75 selectively sends only the points where the amplitude A is greater than the predetermined threshold value “tDet” to the formatter 76. The formatter 76 converts the delay D and the amplitude A, the frame index frm and the segment index seg of the segment into appropriate formats, and outputs them to the system CPU 5 as target information TI.

  Hereinafter, each block will be described in detail.

  Synchronization interference estimation unit 71 includes a switch 77, a switch control unit 78, and an averaging processing unit 79. The switch 77 is a switch controlled by a switch control unit 78 to be described later, and supplies the reception segment y generated within the period turned on by the switch control unit 78 to the averaging processing unit 79. The reception segment y supplied to the averaging processing unit 79 is an example of the “dark output signal” in the present invention.

  The switch control unit 78 specifies the scanning direction in which the feature is present within the maximum distance as the excluded scanning direction Dout, and specifies the direction corresponding to a predetermined angle other than the excluded scanning direction Dout as the specific scanning direction Dtag. Then, the switch control unit 79 sets the switch 77 to ON during a period in which the reception segment y related to the return light Lr of the emitted light Lo emitted in the specified specific scanning direction Dtag is supplied to the switch 77, and in other periods Switch 77 is set to OFF. Note that the switch control unit 78 does not have to turn on the switch 77 in the entire period corresponding to the specific scanning direction Dtag, and may be set to turn on the switch 77 in a part of the period.

  The averaging processing unit 79 averages the received segments y supplied during the period when the switch 77 is on, and supplies the averaged received segments y to the subtracter 72 as the estimated synchronization disturbance w. The estimated synchronization disturbance w is a real vector having a vector length wGate. In this case, for example, the averaging processing unit 79 integrates the reception segments y sequentially supplied from the switch 77 within one frame period, and divides the integrated reception segments y by the number of reception segments y integrated. The averaged received segment y is calculated as the estimated synchronization disturbance w. In another example, the averaging processing unit 79 further estimates the average of the received segment y calculated within one frame period in the frame direction (that is, between different frame indexes) by an IIR filter or the like, Calculate as interference w.

The reception filter 73 calculates the filtered segment z by convolving the impulse response h with the corrected reception segment _{y_dash} . The impulse response h of the reception filter unit 73 can be set by the W register, and is set in advance by the system CPU 5 so that, for example, the SNR at the filter output is increased. For example, the impulse response h is set so as to satisfy the following equation. With this setting, when the noise is white and the system total impulse response is significantly shorter than wGate, optimal performance (high SNR) can be realized.

In the above equation, the reference received pulse “g” is a received segment waveform observed when an object is placed at the scanning origin (R = 0 m), and represents the total impulse response of the entire system including the transmitter 30 and the receiver 40. ing. When it is difficult to actually place an object at the scanning origin, for example, the received segment waveform at “R = 1 m” is observed, and this is mathematically shifted in time to measure the reference received pulse equivalently. Just do it.

FIG. 12A shows waveforms of the reference reception pulse g and the impulse response h within one segment period. As shown in FIG. 12A, the reference reception pulse g and the impulse response h are in a time-reversed relationship within the segment period as defined by the above equation. FIG. 12B shows the waveform of the corrected reception segment y _dash and the filtered segment z observed when an object is placed at the scanning origin (R = 0 m). FIG. 12C shows “R = 10 m ” _Shows the waveform of the corrected received segment y _dash and the filtered segment z observed in the case of“. The reception filter 73 performs noise removal filtering by convolving the impulse response h with the corrected reception segment y _dash and, as shown in FIGS. 12B and 12C, the system delay D _SYS (FIG. 6). The phase is adjusted by the amount corresponding to (C).

  Note that the cyclic convolution operation of the reception filter 71 may be realized in the frequency domain using DFT. By doing so, the amount of calculation can be greatly reduced. In this case, instead of enabling the impulse response h to be set by the W register, the frequency response H may be set by obtaining the frequency response H by performing a DFT operation on the impulse response h in advance.

The peak detector 74 detects the point where the amplitude is maximum in the filtered segment, that is, the peak point with sub-sample accuracy, and outputs the delay D and the amplitude A of the peak point. FIG. 13 illustrates a filtered segment when “R = 10 m”. The curve in the figure represents a continuous time waveform before sampling, and a round dot indicates a sampling point. The peak detector 74 calculates the peak position in the continuous time system based on the sampling series. In the example of FIG. 13, the peak position in sample units on the filtered segment {z _k : k = 0, 1,..., WGate−1} is “k = 34”. On the other hand, the peak position in the continuous time waveform, that is, the sub-sample accuracy is
D = R · Fsmp / (c / 2) = 34.157
It is. The peak detector 74 estimates and calculates the delay D and the amplitude A of the peak point with this subsample accuracy.

Various algorithms can be applied to the peak point detection process with sub-sample accuracy. An example is shown below.
(Procedure 1) A sample point (point P in FIG. 13) having the maximum amplitude is obtained.
(Procedure 2) A quadratic curve passing through these three points is obtained for the point (P point) obtained in Procedure 1 and the points before and after it (Point A and Point B).
(Procedure 3) The delay D and the amplitude A are obtained as the maximum points of the quadratic curve obtained in the procedure 2.

  Based on the peak point information D and A (delay D, amplitude A) output from the peak detection unit 74, the determination unit 75 determines whether or not an object exists at the detection point. This determination is performed by comparing the amplitude A of the peak point with the determination threshold “tDec”. Specifically, the determination unit 75 determines that “the object exists” when A> tDec, and outputs the peak point information. On the other hand, the determination unit 75 determines that “the object does not exist” when A ≦ tDec, and does not output the peak point information.

The formatter 76 converts the peak point information D, A output from the determination unit 75 and the scanning information (frame index frm, segment index seg) corresponding to the peak point into a format that is easy for the user (higher system) to use. The formatter 76 in the present embodiment performs the following format conversion.
(1) The frame index frm is output as it is.
(2) The segment index seg is converted into a horizontal scanning angle θ and output.
(3) The delay D is converted into a radius vector (distance) R and output (R = D (c / 2) / Fsmp).
(4) The amplitude A is output as it is.
(5) The reflectance U is calculated from the amplitude A and the radius vector R and output (U = A / Ψ (R)).

  Here, the function “Ψ (R)” for calculating the reflectance U is a reflectance conversion table, and can be set from the system CPU 5. By setting this table to the expected value of the peak amplitude obtained from a Lambertian diffuser with 100% reflectivity installed at the radius R, reflectivity estimation with less error can be performed.

  As described above, the lidar 1 according to the first embodiment includes the scanning unit 55 that emits the emitted light Lo according to the pulse trigger signal PT, the APD 41 that receives the return light Lr of the emitted light Lo, and the DSP 16. Prepare. The DSP 16 extracts a reception segment y where the amount of the return light Lr is estimated to be equal to or less than a predetermined value based on the current position information IP and the like, and estimates the noise signal caused by the pulse trigger signal PT and the like from the reception segment y. A synchronization disturbance estimation unit 71 that calculates the synchronization disturbance w is provided. Thereby, the lidar 1 can suitably suppress the deterioration of the object detection performance and the distance measurement performance due to the synchronization interference while executing the object detection in all directions.

<Second embodiment>
Next, a second embodiment will be described. In the second embodiment, the DSP 16 predetermines a specific scanning direction Dtag used for estimation of synchronization disturbance in a predetermined direction (in other words, predetermined exclusion scanning direction Dout that is not used for estimation of synchronization disturbance. The difference from the first embodiment. Also in the second embodiment, the DSP 16 suitably reduces the influence of synchronization disturbance while enabling object detection in all directions. Hereinafter, the same elements as those in the first embodiment are appropriately given the same elements, and the description thereof is omitted.

  14A and 14B are overhead views of the vicinity of the vehicle on which the rider 1 is mounted. Here, in FIGS. 14A and 14B, a specific scanning direction Dtag, which is a specific scanning direction used for estimation of synchronization disturbance, and a specific scanning that is within the maximum distance from the lidar 1 in the specific scanning direction Dtag. An area Tag is clearly shown.

  In the example of FIGS. 14A and 14B, the specific scanning direction Dtag is set obliquely rearward so as not to overlap with the following vehicle. In this case, information on the specific scanning direction Dtag and the like is stored in advance in a W register or the like so that the DSP 16 can refer to the information. Note that the reception segment y corresponding to the specific scanning direction Dtag is also preferably used for object detection at times other than when estimating synchronization interference.

  Further, in the examples of FIGS. 14A and 14B, a direction other than the specific scanning direction Dtag (diagonal rearward that does not overlap the following vehicle) is set in advance as an excluded scanning direction Dout that is not used for estimation of synchronization interference. You may do it.

  The DSP 16 specifies the specific scanning area Tag from the maximum distance determined based on the gate width wGate and the specific scanning direction Dtag set in advance, and exists around the lidar 1 based on the current position information IP and the map information IM. The relative position of the feature to be performed with respect to the rider 1 is specified. At this time, the DSP 16 acquires the reception segment y for all directions in addition to or instead of the current position information IP and the map information IM, and analyzes the reception segment y so that The relative position of the existing feature with respect to the rider 1 may be specified.

  In the example of FIG. 14A, the DSP 16 recognizes that the feature 80 exists in the specific scanning area Tag and determines that the synchronization disturbance cannot be estimated. As described above, when there is a feature in the specific scanning area Tag, the DSP 16 affects the reception segment y because the return light Lr reflected by the feature affects the synchronization disturbance based on the reception segment y. Do not estimate

  On the other hand, in the example of FIG. 14B in which no feature exists in the specific scanning area Tag, the DSP 16 determines that all the features specified based on the current position information IP and the map information IM are in the specific scanning area Tag. It is determined that it does not exist, and it is determined that synchronization interference can be estimated. As described above, when there is no feature in the specific scanning area Tag, the DSP 16 is hardly affected by the return light Lr reflected by the feature (that is, affected by only the synchronization disturbance). Since the received segment y can be extracted, synchronization interference is estimated based on the received segment y. In this case, the DSP 16 executes synchronization interference estimation processing described later. The position of the rider 1 shown in FIG. 14B is an example of the “specific position” in the present invention.

  Next, a block configuration of signal processing executed by the DSP 16 in the second embodiment will be described. The block configuration of signal processing executed by the DSP 16 in the second embodiment has the same configuration as shown in FIG. 11 as in the first embodiment. Here, the control of the switch control unit 79 in the second embodiment will be described.

  The switch control unit 78 determines whether or not the feature exists in the specific scanning area Tag based on the current position information IP and the map information IM. If the feature does not exist in the specific scanning area Tag, the switch control unit 78 specifies The switch 77 is set to ON during a period corresponding to the scanning direction Dtag. Specifically, in this case, the switch control unit 78 sets the switch 77 to ON during a period in which the reception segment y related to the return light Lr of the emitted light Lo emitted in the specific scanning direction Dtag is supplied to the switch 77, In other periods, the switch 77 is set to OFF. Note that the switch control unit 78 does not need to turn on the switch 77 in all periods corresponding to the specific scanning direction Dtag determined that the feature does not exist, and turns on the switch 77 in some periods. It may be set.

  Thus, as in the second embodiment, the specific scanning direction Dtag used for estimation of synchronization disturbance is set in a predetermined direction in advance, or the excluded scanning direction Dout that is not used for estimation of synchronization disturbance is set in advance. In other words, it is possible to estimate the synchronization interference more efficiently than the configuration in which the specific scanning direction Dtag is dynamically specified as in the first embodiment. Become.

<Modification>
Below, the modification suitable for the 1st Example mentioned above and the 2nd Example is demonstrated.

(Modification 1)
In the second embodiment, the DSP 16 may determine whether or not there is a feature within a predetermined distance from the rider 1 instead of determining whether or not the feature exists in the specific scanning area Tag. The above-mentioned predetermined distance is set to the maximum distance measurement distance of the rider 1, for example. That is, in the first modification, the estimated synchronization disturbance w is estimated using the received signal when no feature exists within the maximum distance of the lidar 1.

  In this case, the switch control unit 79 calculates the distance between the feature around the current position and the lidar 1 based on the position information of each feature included in the map information IM and the current position information IP, and the distance is predetermined. When there is no feature that is within the distance, the switch 77 is set to ON. Also according to this aspect, the DSP 16 can preferably calculate the estimated synchronization disturbance w based on the reception segment y that is not affected by the return light Lr reflected from the feature.

(Modification 2)
The rider 1 may receive data generated by the map generation device as the current position information IP and the map information IM when the vehicle is present in the same vehicle as the map generation device that executes SLAM (Simultaneous Localization and Mapping). Even in this case, the rider 1 obtains the map information IM including the current position information IP and the information of the features existing around the road, and preferably uses the estimated synchronization disturbance w based on the first or second embodiment. Can be calculated.

(Modification 3)
In the first and second embodiments, the DSP 16 may further determine the presence / absence of a reflected object of the emitted light Lo, such as a feature and a surrounding vehicle, based on the output of an external sensor such as a camera.

  In this case, for example, in the first embodiment, the switch control unit 79 of the DSP 16 is in the scanning direction in which it is determined that the feature does not exist within the maximum distance measurement distance based on the map information IM and the current position information IP, and The scanning direction in which it is determined that no object is present within the maximum distance measurement distance based on the output of the external sensor is set as the specific scanning direction Dtag. Then, the switch control unit 79 turns on the switch 77 during a period corresponding to the set specific scanning direction Dtag. By doing in this way, the lidar 1 can calculate the estimated synchronization disturbance w with high accuracy even when there is an arbitrary reflecting object other than the feature not included in the map information IM.

  Similarly, in the second embodiment, the switch control unit 79 of the DSP 16 determines the presence / absence of the feature in the specific scanning area Tag based on the map information IM and the current position information IP, and based on the output of the external sensor, The presence / absence of an object such as a feature in the specific scanning area Tag is determined. When the switch control unit 79 determines that there is no feature or the like based on the above determination, the switch control unit 79 turns on the switch 77 during a period corresponding to the specific scanning direction Dtag. In this case, the switch control unit 79 may determine the presence / absence of an arbitrary object including a feature in the specific scanning area Tag using the target information TI output from the lidar 1.

1 Rider 5 System CPU
10 ASIC
30 Transmitter 40 Receiver 50 Scan Optics

Claims (12)

An irradiation unit that emits laser light in response to the first signal;
A light receiving unit that receives the return light reflected from the object by the laser beam;
An information processing apparatus for processing an output signal of the light receiving unit of an optical device comprising:
An information processing apparatus comprising: an estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is equal to or less than a predetermined value in the output signal. A first acquisition unit that acquires current position information and map information including the position information of the feature;
The information processing apparatus according to claim 1, wherein the estimation unit specifies the dark output signal from the output signal based on the current position information and the map information.   The information processing apparatus according to claim 2, wherein the estimation unit identifies the output signal at a specific position where the feature does not exist within a predetermined range from the position indicated by the current position information as the dark output signal.   The estimation unit specifies a specific direction that is a direction in which the feature does not exist within a predetermined range from the position indicated by the current position information, and indicates the return light of the laser beam irradiated in the specific direction by the irradiation unit The information processing apparatus according to claim 2, wherein the output signal is specified as the dark output signal.   The information processing apparatus according to any one of claims 1 to 4, further comprising a subtracting unit that subtracts a noise signal estimated by the estimating unit from an output signal of the light receiving unit.   The said estimation part presumes the said noise signal by averaging the said dark output signal each obtained when the said irradiation part scanned in multiple times, The Claim 1 any one of Claims 1-5. Information processing device. The first signal is a pulse signal;
The information processing apparatus according to claim 1, wherein the irradiation unit irradiates pulsed light as the laser light for each different irradiation direction based on the pulse signal. An irradiation unit that emits laser light in response to the first signal;
A light receiving unit for receiving the return light of the laser beam;
An information processing apparatus for processing an output signal of the light receiving unit of an optical device comprising:
A first acquisition unit for acquiring current position information and map information including position information of features;
An information processing apparatus comprising: an estimation unit that estimates a noise signal caused by the first signal from the output signal based on the current position information and the map information. An irradiation unit that emits laser light in response to the first signal;
A light receiving unit that receives the return light reflected from the object by the laser beam;
An estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is equal to or less than a predetermined value in the output signal;
An optical instrument comprising: An irradiation unit that emits laser light in response to the first signal;
A light receiving unit that receives the return light reflected from the object by the laser beam;
A control method executed by an optical apparatus comprising:
A control method including an estimation step of estimating a noise signal caused by the first signal based on a dark output signal in which the return light is equal to or less than a predetermined value in the output signal. An irradiation unit that emits laser light in response to the first signal;
A light receiving unit that receives the return light reflected from the object by the laser beam;
A program executed by a computer that processes an output signal of the light receiving unit of an optical device comprising:
A program that causes the computer to function as an estimation unit that estimates a noise signal caused by the first signal based on a dark output signal in which the return light is less than or equal to a predetermined value in the output signal.   A storage medium storing the program according to claim 11.