JP6008509B2 - Measuring device and measuring method - Google Patents

Description

  The present invention relates to a measuring apparatus and a measuring method for measuring the speed and the like of a moving body.

  2. Description of the Related Art Conventionally, there is a technique for detecting a lateral speed and a side slip angle of a moving body by attaching two optical sensors in a forward direction and a lateral direction of the moving body. Further, a technique for calculating a skid angle from a difference between an integral value obtained by adding a skid angular velocity and an yaw angular velocity obtained from an inertial sensor and a relative azimuth angle by a GPS receiver has been proposed (see, for example, Patent Document 1). ). Furthermore, a technique for measuring a skid angle using two GPS antennas, a uniaxial acceleration sensor, and a uniaxial angular velocity sensor has also been proposed (see, for example, Patent Document 2).

JP 2005-17191 A JP 2007-3461 A

  By the way, since the measurement method by the optical sensor detects the uneven pattern on the road surface, for example, there is a problem that the measurement cannot be performed in a low mu (μ) road environment in which water is sprayed, or the accuracy deteriorates. There is a problem that measurement is not possible in the environment where measurement is most desired.

  Further, the measuring method using the optical sensor has a problem in operation that the mounting is complicated due to its large size. Furthermore, the measurement method using an optical sensor consumes a large amount of power because it uses a light source, and is not suitable for the measurement of a mobile object such as an electric vehicle powered by a battery and a motor, which are currently being researched and developed. .

  Further, in Patent Document 1, the difference between the integrated value of the value obtained by adding the side slip angular velocity and the yaw angular velocity obtained from the inertial sensor and the relative azimuth angle by the GPS receiver is set as the side slip angle. Errors will accumulate if operated for a long time. In addition, since the relative azimuth angle obtained from the GPS receiver is used, the accuracy deteriorates immediately in a GPS reception environment that is affected by multipath.

  In Patent Document 2, since a uniaxial acceleration sensor is used, it is necessary to dispose the acceleration sensor at the center of gravity of the moving body to eliminate the influence of gravity and centrifugal force. It is also necessary to avoid the influence of gravity.

  Further, since Patent Documents 1 and 2 use the speed obtained from GPS as it is, the accuracy deteriorates immediately in a GPS reception environment that is affected by multipath.

  One object of the present invention is to provide a measuring apparatus and a measuring method that can solve the above-described problems, improve the accuracy of the lateral velocity and the skid angle, and can measure in real time. .

The present invention provides the following solutions.
(1) In the measuring apparatus, a first antenna and a second antenna that receive a signal output from a GPS satellite are arranged at a predetermined distance on an axis in the traveling direction of the moving body. Relative position calculation that measures the carrier phase at each antenna point based on the signals received by the antenna and the second antenna, and calculates the relative position of the first antenna and the second antenna based on the measured carrier phase Based on the relative position calculated by the relative position calculation unit, the posture angle measurement unit that measures the absolute posture angle of the moving body, the absolute posture angle measured by the posture angle measurement unit, and the traveling direction of the moving body A calibration unit that calculates the amount of deviation and adjusts the absolute attitude angle based on the amount of deviation, and the frequency of the carrier wave output from the GPS satellite based on the signal received by the first antenna Acceleration that measures acceleration in three axes and angular velocity by a moving body speed measurement unit that calculates the speed of the moving body on the earth coordinates from the Doppler shift amount and a motion sensor that is arranged at a predetermined distance from the first antenna. -Using the absolute attitude angle adjusted by the angular velocity measuring unit, the calibration unit, the velocity of the moving body calculated by the moving body velocity measuring unit, and the acceleration and angular velocity measured by the acceleration / angular velocity measuring unit, Based on the real-time posture angle and speed of the moving object calculated by the speed and posture angle calculation unit, and the speed and posture angle calculation unit that calculates the posture angle and speed of the moving object in real time by performing strapdown calculation The specified position coordinate speed calculation unit that calculates the speed at the specified position coordinates in the moving body and the speed / attitude angle calculation unit And a side velocity / side slip angle calculation unit that calculates in real time the side velocity and the side slip angle at the specified position coordinates in the moving body based on the posture angle and the velocity calculated by the specified position coordinate velocity calculation unit. It is.

  In this way, the measurement device uses the absolute posture angle adjusted by the calibration unit, the velocity of the moving object calculated by the moving object velocity measuring unit, and the acceleration and angular velocity measured by the acceleration / angular velocity measuring unit. Then, a speed / posture angle calculation unit that calculates a real-time posture angle and speed of the moving body by performing strapdown calculation, and a real-time posture angle and speed of the mobile body calculated by the speed / posture angle calculation unit Based on the specified position coordinate speed calculation unit for calculating the speed at the specified position coordinates in the moving body, the posture angle calculated by the speed / attitude angle calculation unit, and the speed calculated by the specified position coordinate speed calculation unit. Based on the side velocity and sideslip angle calculation unit for calculating the side velocity and the side slip angle at the position coordinates specified in the inside of the mobile body. It is possible to accurately calculate the lateral speed and slip angle in the location specified in Oite optionally in real time.

  (2) In the measurement device, the first antenna may be arranged behind the second antenna on the axis of the moving body in the traveling direction.

  In this way, in the measurement apparatus, the direction when the position where the first antenna is arranged and the position where the second antenna is arranged is connected with a straight line matches the traveling direction of the moving object. Even if not, since the amount of deviation is corrected by the calibration unit, it is possible to accurately calculate the lateral velocity and the slip angle at a position arbitrarily designated in the moving body in real time.

  (3) In the measurement device, the speed / posture angle calculation unit performs a strapdown calculation based on the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit, and calculates a correction speed, a correction position, a correction posture angle, Calculate the difference between the strapdown navigator that calculates the corrected acceleration and the corrected angular velocity, the acceleration based on the velocity of the moving object calculated by the moving object velocity measuring unit, and the acceleration measured by the acceleration / angular velocity measuring unit, and A pass / fail judgment unit that calculates a coefficient based on the difference, and a synchronization process of the correction speed, correction position, and correction attitude angle obtained by the strapdown calculation by the strapdown navigator part and the Doppler speed, position, and attitude angle of the GPS satellite. Doppler velocity, position and calibration of GPS satellites calculated by the synchronization processing unit and the moving body velocity measurement unit The error amount is calculated based on the delay processing unit that delays the absolute attitude angle adjusted by the control unit, the data that is synchronized by the synchronization processing unit, and the data that is delayed by the delay processing unit. A subtracting unit to be calculated; a multiplying unit that multiplies the coefficient calculated by the pass / fail judgment unit and the error amount of the subtracting unit; And a state estimation unit that estimates and calculates an adjustment amount with respect to the corrected angular velocity, and the strapdown navigator unit performs strapdown calculation on the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit, and the position, velocity, and posture angle. And the calculated value is corrected by the adjustment amount estimated and calculated by the state estimation unit, and the state estimation unit calculates the angle between the first antenna and the motion sensor. Using the state equation to which the offset error variable and the mounting position error variable are added, the adjustment amount for the correction speed, correction position, correction posture angle, correction acceleration, and correction angular velocity is calculated in real time from the error amount multiplied by the coefficient by the multiplier. It is also possible to employ a configuration that accurately estimates and calculates.

  In this way, the measuring device is multiplied by the coefficient by the multiplication unit using the state equation in which the angle offset error variable of the first antenna and the motion sensor and the attachment position error variable are added by the state estimation unit. Since the adjustment amount for the correction speed is estimated and calculated from the error amount, the direction when the position where the first antenna is disposed and the position where the second antenna is disposed is connected by a straight line, and the traveling direction of the moving body Even if the direction of the motion sensor does not match, the lateral velocity and the skid angle at a position arbitrarily designated in the moving body can be accurately calculated in real time.

  (4) In the measurement method, a first antenna and a second antenna that receive a signal output from a GPS satellite are arranged at a predetermined distance on an axis in the traveling direction of the moving body. Relative position calculation that measures the carrier phase at each antenna point based on the signals received by the antenna and the second antenna, and calculates the relative position of the first antenna and the second antenna based on the measured carrier phase A posture angle measurement step of measuring the absolute posture angle of the moving body based on the relative position calculated in the step, the relative position calculation step, the absolute posture angle measured in the posture angle measurement step, and the traveling direction of the mobile body A calibration process that calculates the amount of deviation and adjusts the absolute attitude angle based on the amount of deviation, and a carrier that is output from a GPS satellite based on a signal received by the first antenna A three-axis acceleration and angular velocity are measured by a moving body speed measurement step of calculating the speed of the moving body on the earth coordinates from the Doppler shift amount of the frequency of the above and a motion sensor arranged at a predetermined distance from the first antenna. Using the acceleration / angular velocity measurement process to measure, the absolute posture angle adjusted by the calibration process, the velocity of the moving object calculated by the moving object velocity measurement process, and the acceleration and angular velocity measured by the acceleration / angular velocity measurement process Then, a speed / posture angle calculation step for calculating the posture angle and speed of the moving body in real time by performing strapdown calculation, and a posture angle and speed of the real time mobile body calculated by the speed / posture angle calculation step Based on the specified position coordinate speed calculation step of calculating the speed at the specified position coordinates in the moving body Lateral speed to calculate in real time the lateral speed and sideslip angle at the specified position coordinates in the moving body based on the attitude angle calculated in the speed / attitude angle calculation process and the speed calculated in the specified position coordinate speed calculation process A side slip angle calculating step.

  Thus, the measurement method uses the absolute attitude angle adjusted by the calibration process, the speed of the moving object calculated by the moving object speed measurement process, and the acceleration and angular velocity measured by the acceleration / angular speed measurement process. Then, a speed / posture angle calculation step for calculating the posture angle and speed of the moving body in real time by performing strapdown calculation, and a posture angle and speed of the real time mobile body calculated by the speed / posture angle calculation step Based on the specified position coordinate speed calculation step for calculating the speed at the specified position coordinates in the moving body, the posture angle calculated in the speed / posture angle calculation step, and the speed calculated in the specified position coordinate speed calculation step. Based on the side velocity and sideslip angle calculation process to calculate the side velocity and side slip angle at the specified position coordinates in the moving body. Runode, it is possible to accurately calculate the lateral speed and slip angle in real time at any given position in the mobile unit.

  It is possible to accurately calculate the lateral velocity and the slip angle at an arbitrarily designated position in the moving body in real time.

It is a block diagram which shows the structure of a measuring device. It is a figure where it uses for description at the time of arrange | positioning a 1st antenna, a 2nd antenna, and IMU on the roof of a moving body. It is a figure where it uses for description about the specific process of a calibration part. It is a block diagram which shows the 1st structure of a speed and posture angle calculation part. It is a figure where it uses for description about the positional relationship of a 1st antenna, a 2nd antenna, and IMU. It is a figure which shows the precision evaluation result of a skid angle. It is a block diagram which shows the 2nd structure of a speed and posture angle calculation part. It is a figure which shows the waveform of the signal input from the mobile body speed measurement part and the attitude angle measurement part. It is a figure where it uses for description about a resample process. It is a figure where it uses for description about the measurement of sampling timing and shift | offset | difference time. It is a block diagram which shows the structure of a synchronous deviation correction | amendment part. It is a figure when 20 [Hz] RAW data, 100 [Hz] linear interpolation data, and 100 [Hz] synchronization shift correction data are compared. It is a figure where it uses for description about the presence or absence of real-time synchronous correction | amendment. It is a figure which shows the real-time interpolation speed output by a measuring device, and the Doppler speed by a GPS receiver. It is a figure which shows the rise of the real-time interpolation speed output by a measuring device, and the rise of the Doppler speed output from a GPS receiver. It is a figure which shows the real-time interpolation speed output by a measuring device, the Doppler speed output from a GPS receiver, and the speed output from an optical speed measuring device.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a measuring apparatus 1 according to an embodiment of the present invention. The measurement apparatus 1 includes a relative position calculation unit 12, an attitude angle measurement unit 13, a calibration unit 14, a moving body speed measurement unit 15, an acceleration / angular velocity measurement unit 16, a speed / posture angle calculation unit 17, A designated position coordinate speed calculator 18 and a lateral speed / slip angle calculator 19 are provided.

  In the relative position calculation unit 12, a first antenna 10 and a second antenna 11 that receive a signal output from a GPS satellite are arranged at a predetermined distance on the axis of the moving body 2 in the traveling direction. The carrier phase at each antenna point is measured based on the signals received by the first antenna 10 and the second antenna 11, and the relative relationship between the first antenna 10 and the second antenna 11 is measured based on the measured carrier phase. Calculate the position. In addition, the mobile body 2 is a vehicle, for example.

  Specifically, the relative position calculation unit 12 receives radio waves output from GPS satellites by the first antenna 10 and the second antenna 11, and measures the carrier phase. The relative position calculation unit 12 performs real-time kinematics (RTK) positioning based on each measured carrier phase. Thereby, the relative position calculation unit 12 can measure the relative positional relationship between the first antenna 10 and the second antenna 11.

  In the present embodiment, it is assumed that the first antenna 10 (master-side antenna) is arranged behind the second antenna 11 (slave-side antenna) in the traveling direction of the moving body 2. Further, as shown in FIG. 2, the first antenna 10 and the second antenna 11 are arranged on the roof of the moving body 2 or the like. For example, it is not limited to the outside.

The posture angle measurement unit 13 measures the absolute posture angle of the moving body 2 based on the relative position calculated by the relative position calculation unit 12.
Moreover, since the baseline vector between the first antenna 10 and the second antenna 11 is finally determined, the attitude angle measurement unit 13 can determine the attitude angle φ _nGPS [rad] with respect to the earth coordinate system. The accuracy of the posture angle is improved by increasing the distance between the antennas. As a reference value, an accuracy of 0.15 ° RMS at 1 m and 0.1 ° RMS at 2 m can be obtained. Since the attitude angle measurement unit 13 measures the attitude angle using a carrier wave of a GPS satellite, the attitude angle measurement unit 13 measures at a sampling frequency of about 5 to 100 Hz in synchronization with the moving body speed measurement unit 15.

  The calibration unit 14 calculates a deviation amount between the absolute posture angle measured by the posture angle measurement unit 13 and the traveling direction of the moving body 2, and adjusts the absolute posture angle based on the deviation amount.

Hereinafter, specific processing of the calibration unit 14 will be described with reference to FIG.
In order to measure an accurate posture angle of the moving body 2, it is necessary to align (axis alignment) the first antenna 10 and the second antenna 11 with the vehicle center line of the moving body 2.
Since it is difficult to adjust manually, the calibration unit 14 performs calibration (calibration) before measurement and performs axis alignment by the following procedure.

  In this way, the calibration unit 14 can adjust the attitude angle to the vehicle center line by using the average attachment angle error. If the first antenna 10 and the second antenna 11 are on the vehicle center line of the moving body 2, the processing by the calibration unit 14 is not necessary. In this case, the absolute posture angle measured by the posture angle measurement unit 13 is supplied to the speed / posture angle calculation unit 17 without being adjusted.

  Based on the signal received by the first antenna 10, the moving body speed measurement unit 15 calculates the speed (Doppler speed) of the moving body 2 on the earth coordinates from the Doppler shift amount of the frequency of the carrier wave output from the GPS satellite. To do. Specifically, the moving body speed measurement unit 15 uses the Doppler shift frequency of the carrier wave from the GPS satellite based on the signal received by the first antenna 10, and the speed of the moving body 2 (NED coordinate system speed: North, east, bottom coordinates) are measured with high accuracy. Moreover, the moving body speed measurement part 15 measures the speed of the moving body 2 with the sampling frequency of about 5-100 Hz. In addition, the moving body speed measurement unit 15 also measures the latitude, longitude, and altitude in synchronization with a sampling frequency of about 5 to 100 Hz. When the distance between the first antenna 10 and the GPS satellite increases or decreases, the phase of the carrier wave received by the first antenna 10 continuously changes, and the frequency decreases or increases. The moving body speed measurement unit 15 acquires the speed output by the first antenna 10 from the amount of change in frequency.

  The acceleration / angular velocity measuring unit 16 measures the acceleration and angular velocity in the three-axis directions by the motion sensor 20 disposed at a predetermined distance from the first antenna 10. Further, the acceleration / angular velocity measuring unit 16 measures the acceleration / angular velocity in the triaxial direction of the installation point of the motion sensor 20 with a sampling frequency of about 100 to 1 kHz. In addition, the motion sensor 20 is comprised by IMU (Internal Measurement Unit: Inertial measurement apparatus) which can measure the acceleration and angular velocity of a triaxial direction, for example. Hereinafter, the motion sensor 20 is referred to as IMU 20.

  Further, as shown in FIG. 2, the IMU 20 has a magnet or the like on the roof in the direction in which the second antenna is arranged with respect to the traveling direction of the moving body 2 (X-axis direction in FIG. 2). Attach with. Similarly, the first antenna 10 and the second antenna 11 are mounted on the roof with a magnet or the like. In addition, the accuracy is improved as the distance between the first antenna 10 and the second antenna is increased, but it is used at about 1 m or 2 m from the viewpoint of operation.

  The velocity / posture angle calculation unit 17 includes the absolute posture angle adjusted by the calibration unit 14, the speed of the mobile unit 2 calculated by the mobile unit speed measurement unit 15, and the acceleration measured by the acceleration / angular velocity measurement unit 16. Further, the posture angle and speed of the moving body 2 in real time are calculated by performing strapdown calculation using the angular velocity.

  The designated position coordinate speed calculation unit 18 calculates the speed at the position coordinates designated in the moving body 2 based on the real time posture angle and speed of the moving body 2 calculated by the speed / attitude angle calculation unit 17. Here, the designated position coordinates are, for example, the center of gravity (COG, Center Of Gravity) of the moving body 2 as shown in FIG.

  The lateral speed / slip angle calculation unit 19 is based on the position coordinates specified in the moving body 2 based on the posture angle calculated by the speed / posture angle calculation unit 17 and the speed calculated by the specified position coordinate speed calculation unit 18. The lateral speed and side slip angle at are calculated in real time.

  In this way, the measuring device 1 can accurately calculate the lateral velocity and the skid angle at a position arbitrarily designated in the moving body 2 in real time.

<First embodiment>
Next, the configuration of the speed / attitude angle calculator 17 will be described with reference to FIG.
As shown in FIG. 4, the speed / attitude angle calculation unit 17 includes a strapdown navigator unit 100, a pass / fail determination unit 101, a synchronization processing unit 102, a multiplication unit 103, a state estimation unit 104, and a delay processing unit. 105, and calculates real-time high-precision speed and attitude angle.

  The strapdown navigator unit 100 (autonomous navigation algorithm) performs a strapdown calculation based on the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit 16, and calculates a corrected velocity (NED direction) and a corrected position (latitude, Longitude, altitude), corrected attitude angle, corrected triaxial acceleration, and corrected triaxial angular velocity are calculated. Specifically, the strapdown navigator unit 100 performs a strapdown operation on the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit 16 to calculate a position, velocity, posture angle, and the like, and further estimates by the state estimation unit 104. The position, speed, posture angle, etc. obtained by the strapdown calculation are corrected with the calculated adjustment amount and output.

  The pass / fail judgment unit 101 calculates a difference between the acceleration based on the speed of the mobile body 2 calculated by the mobile body speed measurement unit 15 and the acceleration measured by the acceleration / angular velocity measurement unit 16, and calculates a coefficient based on the difference. calculate.

  Here, details of the pass / fail judgment unit 101 will be described. The pass / fail judgment unit 101 calculates the difference between the acceleration calculated from the Doppler velocity measured by the moving body velocity measurement unit 15 and the acceleration measured by the acceleration / angular velocity measurement unit 16, and based on the calculated difference, the coefficient α Is calculated.

  For example, the pass / fail determination unit 101 calculates the difference between the acceleration obtained by differentiating the Doppler velocity and the acceleration measured by the IMU 20, squares the calculated difference, and performs envelope processing on the squared difference. And the quality determination part 101 calculates | requires the inverse function of the function which shows the difference after performing an envelope process, and calculates coefficient (alpha) based on the calculated | required inverse function.

  That is, the pass / fail determination unit 101 performs a pass / fail determination when it is determined that the Doppler velocity does not include noise based on the obtained inverse function, and performs a determination when it is determined that the Doppler speed includes noise. A coefficient α obtained by quantifying each determination is calculated.

  Also, the pass / fail determination unit 101 calculates the difference between the (absolute) posture angle measured by the posture angle measurement unit 13 and the corrected posture angle calculated by the strapdown navigator unit 100, and applies the pass / fail determination method described above. As a result, the quality of the posture angle measured by the posture angle measurement unit 13 can be determined in real time.

  The synchronization processing unit 102 corrects the correction speed (real-time interpolation speed), the correction position, the correction posture angle obtained by the strapdown calculation by the strapdown navigator unit 100, and the speed in the NED direction obtained from the moving body speed measurement unit 15. (Doppler speed), position (latitude, longitude, altitude), and the posture angle obtained from the calibration unit 14 are synchronized by a predetermined time to be synchronized.

The delay processing unit 105 delays the GPS satellite Doppler velocity and position calculated by the moving body velocity measuring unit 15 and the absolute attitude angle adjusted by the calibration unit 14.
Specifically, the delay processing unit 105 performs the NED direction speed (Doppler speed) and position (latitude, latitude) obtained from the moving body speed measurement unit 15 for the time required for the coefficient α to be calculated by the pass / fail judgment unit 101. (Longitude, altitude) and the absolute attitude angle obtained from the calibration unit 14 are delayed by the time used for the pass / fail judgment calculation.

The subtracting unit 108 subtracts the speed, posture angle, and position of the moving body 2 delayed by the delay processing unit 105 from the real-time interpolation speed, correction position, and corrected posture angle delayed by the synchronization processing unit 102 to obtain an error. The quantity δx is determined.
The multiplication unit 103 multiplies the coefficient α calculated by the pass / fail determination unit 101 and the error amount δx obtained by subtracting the delay processing unit 105 and the synchronization processing unit 102 by the subtraction unit 108.

  The state estimation unit 104 estimates and calculates an adjustment amount for the correction speed, the correction position, the correction posture angle, the correction triaxial acceleration, and the correction triaxial angular velocity from the error amount δx multiplied by the coefficient α by the multiplication unit 103. Specifically, the state estimation unit 104 uses the state equation in which the angle offset error variable of the first antenna 10 and the IMU 20 and the attachment position error variable are added, and the error amount multiplied by the coefficient α by the multiplication unit 103. From δx, a correction speed, a correction position, a correction posture angle, a correction triaxial acceleration, and an adjustment amount for the correction triaxial angular velocity are accurately estimated and calculated.

なお、Δｔは、サンプリング時間であり、δｘは、状態変数であり、（８）式で表わされる１９×１の状態変数行列である。

Hereinafter, specific processing of the state estimation unit 104 will be described.
The state estimation unit 104 includes a so-called extended Kalman filter.
The extended Kalman filter can be derived from equations (5) and (6) obtained by discretizing the state equation (equation (3)) and the observation equation (equation (4)).

Note that Δt is a sampling time, δx is a state variable, and is a 19 × 1 state variable matrix expressed by equation (8).

  Thereby, the state estimation part 104 can calculate the speed and the attitude | position angle of the position point of IMU20 with high precision. The attachment position relationship can be corrected by estimating, for example, 0.50 [m], and the attachment angle error can be estimated up to, for example, about 5 °.

Next, calculation of the speed at the specified position by the specified position coordinate speed calculation unit 18 will be described.
When the distance from the place where the IMU 20 is installed to the specified position (for example, the center of gravity of the moving body 2) is ζ _b [rad], the specified position coordinate speed calculation unit 18 calculates the NED coordinate system speed vector of the center of gravity point. Is calculated by equation (9).

The calculation by the lateral speed / slip angle calculation unit 19 will be described.
The side velocity / slip angle calculation unit 19 calculates the side slip angle β [rad] according to the equation (10).

Further, the lateral velocity / slip angle calculating unit 19 calculates the lateral velocity V _L [m / s] by the equation (11).

  Here, FIG. 6 shows the accuracy evaluation result of the skid angle. The horizontal axis indicates the straight traveling speed, and the vertical axis indicates the side slip angle accuracy. FIG. 6 shows the results of measuring the accuracy of the skid angle when the straight traveling speed is 10, 20, 40, 60, 80, 100 [km / h] five times.

  In addition, the accuracy of the optical detector as a comparison target was verified in the same test. The measuring device 1 achieves a high accuracy of within 0.15 ° RMS at a side slip angle of 30 [km / h] or more in a straight traveling speed. Also, it has been confirmed that the output delay time is within 5 [ms] and has high responsiveness. From these results, it can be seen that the measurement apparatus 1 according to the present embodiment is more accurate than the optical detector.

  Therefore, the measuring device 1 has an advantage that it can be used as a measuring instrument for ESC (side slip prevention device) evaluation. Moreover, since the measuring device 1 can perform real-time output (5 ms or less), there is an advantage that it can be used without forcing the user to perform post-processing.

<Second embodiment>
Next, a case where the sampling frequency used in the acceleration / angular velocity measuring unit 16 is different from the sampling frequency used in the moving body velocity measuring unit 15 and the posture angle measuring unit 13 will be described. Specifically, the sampling frequency used by the acceleration / angular velocity measuring unit 16 is 100 [Hz], and the sampling frequency used by the moving body velocity measuring unit 15 and the posture angle measuring unit 13 is 20 [Hz].

Here, if the acceleration / angular velocity measuring unit 16, the moving body velocity measuring unit 15, and the posture angle measuring unit 13 are not time synchronized, there is a problem that accuracy is not improved and an accurate response time is not obtained.
The moving body speed measurement unit 15 and the posture angle measurement unit 13 sample data at 20 [Hz] based on real time. This period of 20 [Hz] is accurate with respect to absolute time. However, since the output I / F of the GPS receiver composed of the first antenna 10 and the second antenna 11 adopts UART, 50 [ms] ± 2 to 3 [ms] due to the difference in message length. ] Jitter will occur.

  On the other hand, the acceleration / angular velocity measuring unit 16 samples at 100 [Hz] based on the accuracy of the crystal oscillator. This period of 100 [Hz] depends on the crystal oscillator (about 60 [ppm]). Thus, the period is not accurate with respect to absolute time. Specifically, a cumulative error of about several [ms] occurs in several hundred seconds.

Therefore, it is difficult to synchronize the acceleration / angular velocity measurement unit 16, the moving body velocity measurement unit 15, and the posture angle measurement unit 13 even if synchronization is simply performed using the conventional technology.
One object of the measuring apparatus 1 according to the present embodiment is to solve with a simple algorithm without using special hardware for synchronization.

Specifically, the speed / attitude angle calculation unit 17 of the measurement apparatus 1 includes a strapdown navigator unit 100, a pass / fail determination unit 101, a synchronization processing unit 102, a multiplication unit 103, as shown in FIG. By providing the state estimation unit 104, the interpolation processing unit 106, and the synchronization deviation correction unit 107, the speed measurement values of the acceleration / angular velocity measurement unit 16 and the posture angle measurement unit 13 having different sampling rates are time-synchronized, and the speed accuracy is obtained. In addition, the accumulated distance accuracy and real-time responsiveness are improved.
Note that the basic operations of the strapdown navigator unit 100, the pass / fail determination unit 101, the synchronization processing unit 102, the multiplication unit 103, and the state estimation unit 104 are the same as in the first embodiment. Hereinafter, a receiver including the first antenna 10 and the second antenna 11 is referred to as a GPS receiver.

  The strapdown navigator unit 100 (autonomous navigation algorithm) performs a strapdown calculation based on the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit 16, and calculates a corrected velocity (NED direction) and a corrected position (latitude, Longitude, altitude), corrected attitude angle, corrected triaxial acceleration, and corrected triaxial angular velocity are calculated. In detail, the strapdown navigator unit 100 performs the strapdown calculation by merging the acceleration and angular velocity measured by the acceleration / angular velocity measurement unit 16 with the adjustment amount estimated by the state estimation unit 104.

  The interpolation processing unit 106 performs linear interpolation processing on the waveform of the moving body speed measured by the moving body speed measuring unit 15 so that the sampling frequency on the GPS receiver side matches the sampling frequency on the IMU 20 side. Specifically, the interpolation processing unit 106 matches the sampling frequency of the acceleration / angular velocity measuring unit 16 with the sampling frequency of the moving body velocity measuring unit 15 and the attitude angle measuring unit 13, Linear interpolation processing is performed from the current measurement value of the measurement unit 13 and past sampling, and processing to obtain the same sampling frequency as that of the acceleration / angular velocity measurement unit 16 is performed.

  In the present embodiment, the sampling frequency of the moving body speed measuring unit 15 and the posture angle measuring unit 13 is described as performing so-called up-sampling that matches the high sampling frequency of the acceleration / angular velocity measuring unit 16. Not limited. For example, so-called down-sampling may be performed in which the sampling frequency of the acceleration / angular velocity measuring unit 16 is matched with the low sampling frequency of the moving body velocity measuring unit 15 and the posture angle measuring unit 13. In the case of this configuration, the interpolation processing unit 106 is disposed in front of the strapdown navigator unit 100.

  Here, the waveforms of signals (digital signals) input from the moving body speed measurement unit 15 and the posture angle measurement unit 13 are shown in FIG. Since the sampling frequency is 20 [Hz], it is 50 [ms] per sample.

  The interpolation processing unit 106 performs a resampling process on the signals (20 [Hz] RAW data) input from the moving body speed measurement unit 15 and the posture angle measurement unit 13 at 100 [Hz] (see FIG. 9). When linear interpolation processing is performed in real time, a time delay (25 [ms]) due to processing always occurs due to causality (A shown in FIG. 9). Since this time delay is a constant value, the synchronization processing is performed. The unit 102 performs correction.

  The sampling time of the acceleration / angular velocity measuring unit 16 is 10 [ms]. The accuracy of this sampling time depends on the crystal oscillator. The sampling time of the moving body speed measuring unit 15 is 50 [ms] ± 2 to 3 [ms]. This sampling time is accurate with respect to absolute time.

In this way, a deviation of 0 to 10 [ms] occurs between the measurement time of the acceleration / angular velocity measurement unit 16 and the measurement time of the moving body velocity measurement unit 15 (and the posture angle measurement unit 13). End up. Further, the deviation amount varies depending on the accuracy of the measurement timing of the acceleration / angular velocity measuring unit 16 and the jitter of the moving body velocity measuring unit 15 (and the posture angle measuring unit 13) (see FIG. 10).
Note that the interpolation processing unit 106 may use n-order function interpolation, spline interpolation, Lagrange interpolation, or a multi-rate filter in order to further improve the interpolation accuracy.

  The pass / fail judgment unit 101 calculates an acceleration based on the speed of the moving body 2 subjected to the linear interpolation processing by the interpolation processing unit 106, and calculates a difference between the calculated acceleration and the acceleration measured by the acceleration / angular velocity measurement unit 16. And a coefficient α based on the difference is calculated.

  The synchronization deviation correction unit 107 performs correction for advancing the phase of the speed of the moving body 2 subjected to the linear interpolation processing by the interpolation processing unit 106 by the time difference between the measurement timings of the GPS receiver side and the IMU 20 side. Specifically, the synchronization deviation correction unit 107 performs linear interpolation on the synchronization deviation (several [ms]) between the output of the strapdown navigator unit 100 and the speed and attitude angle interpolated by the interpolation processing unit 106. By performing the processing for predicting the processed speed and posture angle, the level is synchronized to a level within 1 [ms]. Accordingly, the synchronization error correction unit 107 can implement the error amount δx input to the state estimation unit 104 (extended Kalman filter) with the measurement value at the same timing.

Further, as shown in FIG. 11, the synchronization shift correction unit 107 includes a differentiation processing unit 107a, an LPF 107b, a shift time adjustment unit 107c, and an adder 107d.
The differential processing unit 107a calculates the acceleration a by differentiating the input GPS speed v (speed in the NED direction).
The LPF 107b performs a filtering process on the acceleration a calculated by the differentiation processing unit 107a.
The deviation time adjustment unit 107c adds the deviation time to the acceleration filtered by the LPF 107b.
The adder 107d adds the value adjusted by the shift time adjustment unit 107c to the GPS speed (see equation (12)).

In this way, the synchronization deviation correction unit 107 performs a process of advancing the speed phase of the GPS receiver by the deviation time Δt.
Note that, similarly to the delay processing unit 105, the synchronization deviation correction unit 107 uses the Doppler velocity measured by the moving body velocity measurement unit 15 for the calculation only for the time for which the coefficient α is calculated by the pass / fail judgment unit 101. The process which delays is also performed. That is, the synchronization deviation correction unit 107 delays the calculation for calculating the error amount due to the Doppler speed by the time for which the coefficient α is calculated by the pass / fail determination unit 101.

Here, FIG. 12 shows a signal (20 [Hz] RAW data) input from the moving body velocity measuring unit 15 and the posture angle measuring unit 13 and a signal (100 [Hz] linear interpolation) after processing by the interpolation processing unit 106. Data) and a signal (100 [Hz] synchronization error correction data) after processing by the synchronization error correction unit 107 are shown in comparison.
As shown in FIG. 12, the synchronization deviation correction unit 107 measures the measurement time of the moving body velocity measurement unit 15 (and the posture angle measurement unit 13) and the measurement time of the acceleration / angular velocity measurement unit 16 in which the deviation amount varies for each sample. It is possible to reliably correct the difference.

The synchronization processing unit 102 corrects the correction speed (real-time interpolation speed), the correction position, the correction posture angle obtained by the strapdown calculation by the strapdown navigator unit 100, and the speed in the NED direction obtained from the moving body speed measurement unit 15. (Doppler speed), position (latitude, longitude, altitude), and the posture angle obtained from the calibration unit 14 are synchronized by a predetermined time to be synchronized.
The subtracting unit 108 includes the speed, posture angle, and position of the moving body 2 that have been synchronously corrected and delayed by the synchronization deviation correcting unit 107, and the real-time interpolation speed, correction position, and corrected posture angle that have been synchronously delayed by the synchronization processing unit 102. Is subtracted to calculate the error amount δx.
The multiplication unit 103 multiplies the coefficient α calculated by the pass / fail determination unit 101 by the error amount δx obtained by the subtraction process between the synchronization deviation correction unit 107 and the synchronization processing unit 102 by the subtraction unit 108.

  The state estimation unit 104 estimates and calculates an adjustment amount for the correction speed, the correction position, the correction posture angle, the correction triaxial acceleration, and the correction triaxial angular velocity from the error amount multiplied by the coefficient by the multiplication unit 103. Specifically, the state estimation unit 104 uses the state equation to which the angular offset error variable of the GPS receiver and the IMU 20 and the attachment position error variable are added, and an error amount δx multiplied by the coefficient α by the multiplication unit 103. Are used to estimate and calculate the correction speed (−kαδx) for the correction speed, the correction position, the correction posture angle, the correction triaxial acceleration, and the correction triaxial angular velocity.

  In addition, when the moving body 2 is accelerated from 10 [km / h] to 40 [km / h] and decelerated from 40 [km / h] to 10 [km / h] over a 100 m calibrated accurately (acceleration) FIG. 13 shows the accuracy results of comparing the conventional technique (without real-time synchronization correction) and the measuring apparatus 1 according to the present embodiment (with real-time synchronization correction) in the deceleration test). It can be seen that the measuring device 1 according to the present example is clearly more accurate.

  FIG. 14 is a diagram illustrating a real-time interpolation speed output by the measurement apparatus 1 and a Doppler speed by a GPS receiver configured by the first antenna 10 and the second antenna 11. FIG. 14A is a graph showing the Doppler speed measured by the GPS receiver in a graph in which the vertical axis represents speed and the horizontal axis represents time. Similarly, FIG. 14B is a graph showing the real-time interpolation speed output by the measuring apparatus 1 on a graph with the vertical axis representing speed and the horizontal axis representing time.

  As shown in FIG. 14A, when a signal from a GPS satellite is reflected by trees, buildings, etc. and a multipath is generated, the Doppler speed by the GPS receiver is output including the speed due to the noise due to the multipath. Is done. On the other hand, as shown in FIG. 14B, even if the Doppler speed by the GPS receiver includes noise due to multipath, the measuring apparatus 1 reduces the influence of noise and moves the moving object 2 with a very smooth waveform. Outputs the speed.

  FIG. 15 is a diagram illustrating the rise of the real-time interpolation speed output by the measuring apparatus 1 and the rise of the Doppler speed output from the GPS receiver. FIG. 15A is a diagram showing a comparison between the speed output by the measuring device 1, the Doppler speed by the GPS receiver, and the speed obtained by performing a low-pass process of 1 Hz on the Doppler speed. FIG. 15B is an enlarged view of a part of FIG. 15A (a part surrounded by a broken line circle).

  As shown in FIG. 15B, the speed output by the measuring device 1 is compared to the Doppler speed including noise due to multipath and the speed obtained by removing noise from the Doppler speed including noise due to multipath by low-pass processing. Realizes real-time performance by reducing output delay due to computation for noise removal.

  FIG. 16 is a diagram illustrating a real-time interpolation speed output from the measurement apparatus 1, a Doppler speed output from the GPS receiver, and a speed output from the optical speed measurement apparatus. Here, the optical speed measuring device extracts only the irregular reflection at a specific interval from the irregular pattern on the road surface by a special light receiving element having a comb-like structure at a specific interval (for example, 2.3 [mm]). The high-accuracy speed calculated by multiplying the extracted reflection unevenness count value by a specific interval is output.

  As shown in FIG. 16, the measuring apparatus 1 reduces the influence of noise on the Doppler speed including noise due to multipath, outputs the speed of the moving body 2 in real time, and the output speed is determined by the optical speed measuring apparatus. Similar to high-precision speed.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

DESCRIPTION OF SYMBOLS 1 Measurement apparatus 10 1st antenna 11 2nd antenna 12 Relative position calculation part 13 Attitude angle measurement part 14 Calibration part 15 Mobile body speed measurement part 16 Acceleration / angular speed measurement part 17 Speed / attitude angle calculation part 18 Designated position coordinate Speed calculation unit 19 Side velocity / slip angle calculation unit 100 Strapdown navigator unit 101 Pass / fail judgment unit 102 Synchronization processing unit 103 Multiplication unit 104 State estimation unit 105 Delay processing unit 106 Interpolation processing unit 107 Synchronization deviation correction unit 108 Subtraction unit

Claims (4)

A first antenna and a second antenna of a GPS receiver that receives a signal output from a GPS satellite are arranged at a predetermined distance on an axis in the traveling direction of the moving body, and the first antenna and Relative position for measuring the carrier phase at each antenna point based on the signal received by the second antenna, and calculating the relative position of the first antenna and the second antenna based on the measured carrier phase A calculation unit;
A posture angle measuring unit that measures an absolute posture angle of the moving body based on the relative position calculated by the relative position calculating unit;
A calibration unit that calculates a deviation amount between the absolute posture angle measured by the posture angle measurement unit and the traveling direction of the moving body, and adjusts the absolute posture angle based on the deviation amount;
Based on the signal received by the first antenna, the velocity of the moving body on the earth coordinates is calculated from the Doppler shift amount of the frequency of the carrier wave output from the GPS satellite, and the signal received by the first antenna is calculated. Based on the moving body speed measurement unit for calculating the position (latitude, longitude, altitude) of the moving body,
An acceleration / angular velocity measuring unit that measures acceleration and angular velocity in three axes by a motion sensor arranged at a predetermined distance from the first antenna;
The absolute attitude angle adjusted by the calibration unit, the speed of the moving body calculated by the moving body speed measuring unit , the position (latitude, longitude, altitude) of the moving body, and the acceleration / angular velocity measurement A strapdown calculation is performed using the acceleration and angular velocity measured by the unit , and an attachment angle offset error variable that is a deviation of the axis of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor ; By using a state equation to which an attachment position error variable between the motion sensor and the first antenna of the GPS receiver is added , the axes of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor installation point of the motion sensor and the attachment angle error which is a deviation, real-time the mobile Definitive, a correction NED speed correction position, correcting attitude angle, correction 3-axis angular velocity, and the velocity and attitude angle calculation unit for calculating a correction three-axis acceleration,
A corrected NED speed, a corrected position, a corrected attitude angle, a corrected triaxial angular velocity, a corrected triaxial acceleration , and the motion sensor at the installation point of the motion sensor of the moving body calculated in real time by the speed / posture angle calculating unit. based of the position coordinates of the designated position in the coordinate system, and specifies the position coordinate velocity calculation unit for calculating a NED coordinate system velocity vector at your Keru the specified position on the moving body,
The corrected attitude angle calculated by the speed / attitude angle calculation unit and the mounting angle error, which is a deviation of the axis of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor, are calculated by the calibration unit. wherein a shift amount between the traveling direction of the moving body, and NED coordinate system velocity vector at the specified position calculated by the specified position coordinate velocity calculation unit, based on, Keru before Symbol fingers on the moving body has calculates the sideslip angle in place in real time, based on the slip angle in the NED coordinate velocity vector and the specified position in front Symbol specified position calculated by the specified position coordinate velocity calculation unit, wherein measuring device and a lateral speed and lateral slip angle calculating unit for calculating a lateral speed in your Keru before Symbol specified position in the moving body.   The measuring apparatus according to claim 1, wherein the first antenna is arranged behind the second antenna on an axis in a traveling direction of the moving body. The speed / attitude angle calculation unit further includes:
Strapdown calculation is performed based on the acceleration and angular velocity measured by the acceleration / angular velocity measuring unit, and the corrected velocity, corrected position, corrected attitude angle, corrected triaxial acceleration, and corrected triaxial angular velocity are calculated as calculated values. Navigator section,
A pass / fail judgment unit that calculates a difference between an acceleration based on the speed of the mobile body calculated by the mobile body speed measurement unit and an acceleration measured by the acceleration / angular velocity measurement unit, and calculates a coefficient based on the difference; ,
Correction speed, correction position, correction posture angle obtained by strapdown calculation by the strapdown navigator section, speed of the mobile body on the earth coordinates calculated by the mobile body speed measurement section, and position of the mobile body (Latitude, longitude, altitude), a synchronization processing unit that performs synchronization processing of the absolute posture angle obtained from the calibration unit,
Delay processing for delaying the speed of the moving body on the earth coordinates calculated by the moving body speed measuring unit, the position (latitude, longitude, altitude) of the moving body and the absolute attitude angle adjusted by the calibration unit And
A subtraction unit that calculates an error amount based on data that has been subjected to synchronization processing by the synchronization processing unit and data that has been subjected to delay processing by the delay processing unit;
A multiplication unit that multiplies the coefficient calculated by the pass / fail judgment unit by the error amount of the subtraction unit;
A state estimation unit that estimates and calculates an adjustment amount for the correction speed, correction position, correction posture angle, correction triaxial acceleration, and correction triaxial angular velocity from the error amount multiplied by the coefficient by the multiplication unit,
The strapdown navigator unit further performs a strapdown calculation on the acceleration and angular velocity measured by the acceleration / angular velocity measuring unit to calculate a correction speed, a correction position, a correction posture angle, a correction triaxial acceleration, and a correction triaxial angular velocity. Further, the calculated value is corrected by the adjustment amount estimated by the state estimation unit,
The state estimation unit further includes a mounting angle offset error variable, which is a shift of an axis of the first antenna and the second antenna with respect to a coordinate axis of the motion sensor, and a first of the motion sensor and the GPS receiver. Using the state equation to which the attachment position error variable with the antenna is added, the attachment angle error and the attachment position error are estimated to correct the attachment position relationship, and the correction is performed from the error amount multiplied by the coefficient by the multiplication unit. The measuring apparatus according to claim 1, wherein the adjustment amount for the speed, the corrected position, the corrected attitude angle, the corrected triaxial acceleration, and the corrected triaxial angular velocity is accurately estimated and calculated in real time. A first antenna and a second antenna of a GPS receiver that receives a signal output from a GPS satellite are arranged at a predetermined distance on an axis in the traveling direction of the moving body, and the first antenna and Relative position for measuring the carrier phase at each antenna point based on the signal received by the second antenna, and calculating the relative position of the first antenna and the second antenna based on the measured carrier phase A calculation process;
A posture angle measuring step of measuring an absolute posture angle of the moving body based on the relative position calculated by the relative position calculating step;
A calibration step of calculating a deviation amount between the absolute posture angle measured in the posture angle measurement step and the traveling direction of the moving body, and adjusting the absolute posture angle based on the deviation amount;
Based on the signal received by the first antenna, the Doppler shift amount of the frequency of the carrier wave output from the GPS satellite, the speed of the moving object on the earth coordinates, and the position (latitude, longitude, altitude) of the moving object A moving body speed measurement step for calculating
An acceleration / angular velocity measuring step of measuring acceleration and angular velocity in three axes by a motion sensor arranged at a predetermined distance from the first antenna;
The absolute attitude angle adjusted by the calibration process, the speed of the mobile object calculated by the mobile object speed measurement process , the position (latitude, longitude, altitude) of the mobile object, and the acceleration / angular velocity measurement A strapdown calculation is performed using the acceleration and angular velocity measured in the process , and an attachment angle offset error variable which is a deviation of the alignment axis of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor ; By using a state equation to which an attachment position error variable between the motion sensor and the first antenna of the GPS receiver is added , the axes of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor and attachment angle error which is a deviation, the real-time the movement of the motion sensor In 置点 a correction NED speed correction position, correcting attitude angle, correction 3-axis angular velocity, and the velocity and attitude angle calculation step of calculating a corrected three-axis acceleration,
A corrected NED speed, a corrected position, a corrected posture angle, a corrected triaxial angular velocity, a corrected triaxial acceleration , and the motion sensor at the installation point of the motion sensor of the moving body calculated in the speed / posture angle calculating step. based on the position coordinates of a specified position in a coordinate system, and specifies the position coordinate velocity calculating step of calculating the NED coordinate system velocity vector at your Keru the specified position on the moving body,
The corrected attitude angle calculated in the speed / orientation angle calculation step and the mounting angle error which is the deviation of the axis of the first antenna and the second antenna with respect to the coordinate axis of the motion sensor, and the calibration step are calculated. wherein a shift amount between the traveling direction of the moving body, and NED coordinate system velocity vector at the specified position calculated by the specified position coordinate velocity calculation step, Keru before Symbol specified in the mobile body based on the calculates the sideslip angle at the position, based on the slip angle in the NED coordinate velocity vector and the specified position in front Symbol specified position calculated by the specified position coordinate velocity calculation process, the mobile unit measurement method and a lateral speed and lateral slip angle calculating step of calculating a lateral speed of your Keru before Symbol specified position.

Images