JPH06288776A - Azimuth detector - Google Patents

Abstract

(57) [Summary] [Structure] The usage ratio of the output data of the turning angular velocity sensor and the geomagnetic sensor is determined by individually analyzing and evaluating the error factors included in the output data of the turning angular velocity sensor and the geomagnetic sensor. , When the current azimuth of the moving body is detected, the error of the gyro output is properly evaluated by considering the error of the gyro scale factor. [Effect] The Kalman gain, which is the ratio between the usage rate of the data of the geomagnetic sensor and the usage rate of the data of the gyro output, can be set to an appropriate value, and as a result, the direction of the vehicle can be detected more accurately. You will be able to.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an azimuth detecting device, and more particularly, it uses a azimuth of a vehicle obtained from a GPS receiver and detects a turning angular velocity of the vehicle (for example, an optical fiber gyro, Mechanical gyro, vibrating gyro, gas rate gyro)
The present invention relates to an improvement in an azimuth detecting device that detects an azimuth in a yaw direction of a vehicle.

[0002]

2. Description of the Related Art Conventionally, a distance sensor, a direction sensor (any one of a GPS receiver and a gyro) has been used as a method for detecting the position of a vehicle traveling in an arbitrary portion of a road traffic network. A processing device that performs necessary processing on output signals from both sensors, and a distance change amount ΔL and a heading θ (in the case of a GPS receiver) that accompany the traveling of the vehicle.
Or using the direction change amount Δθ (for gyro),
Dead Reckoning for obtaining current vehicle position data
g) is proposed. This method is based on ΔL and θ, for example, the east-west direction component Δx of ΔL (= ΔL × cos θ)
And the north-south direction component Δy (= ΔL × sin θ) are calculated, and each of the components Δx and Δy is added to the previous position output data (Px ′, Py ′) to obtain the current position output data (Px, Although this is a method of obtaining Py), it has a drawback that the error included in the obtained current position data is accumulated due to the error that the azimuth sensor necessarily has.

That is, if the azimuth sensor measures the Doppler shift generated by receiving radio waves from a GPS (Global Positioning System) satellite, the GPS azimuth error increases as the vehicle speed decreases. There is a tendency. On the other hand, when using a gyro, for example, when the direction change exceeds a predetermined value, when the power (ignition) is turned on, when traveling at extremely low speed, or when traveling on a bad road such as a mountain road is detected. Is known to have many errors in the output data of the sensor.

Therefore, if any one of the data is used as it is, the azimuth detection accuracy is lowered. Therefore, as the azimuth sensor, two azimuth sensors, a gyro and a GPS receiver, are used together, and if the reliability of either the angular velocity data of the gyro or the azimuth data of the GPS receiver decreases, other data is used. It is possible to make up for it.

That is, the Kalman filter gain is calculated in consideration of the peculiar error components of the azimuth data of the gyro and the GPS receiver, and the azimuth data obtained from the azimuth data of the GPS receiver and the angular velocity data of the gyro is obtained. It is conceivable to obtain the current estimated heading of the vehicle by performing weighting processing based on the Kalman filter gain.

However, when this method is used, the problem is how to evaluate the peculiar error component of the output data of the gyro and the GPS receiver.

[0007]

The process is the simplest if each error component is evaluated by some method and set to a constant value, but as described above, the error component included in the GPS direction data during running is Since it changes depending on the vehicle speed, it is not sufficient to set it at a constant value, and it is desirable to properly estimate the error amount in real time by some means. Further, since the offset included in the angular velocity data of the gyro also varies with time due to changes in temperature and humidity, it is necessary to take into consideration the error in the angular velocity data based on the variation. Further, the scale factor (output gain) of the gyro may be far from the standard value, so it must be taken into consideration.

The applicant of the present invention actually measures the azimuth data of the gyro and the GPS receiver and processes them in real time,
A azimuth detecting device has been proposed that can accurately estimate the current azimuth of a vehicle by giving more weight to more reliable data (Japanese Patent Laid-Open No. 3-188316). The azimuth detecting device measures the dispersion value contained in the final output data of the gyro and the GPS receiver exclusively, and performs processing based on this, and is included in the azimuth data of the gyro and the GPS receiver. It does not focus on individual error factors.

Therefore, according to the present invention, the error factors included in the azimuth data of the GPS receiver and the angular velocity data of the gyro are individually analyzed and evaluated, and then the Kalman filter gain is calculated to calculate the gyro and the GPS. It is an object of the present invention to provide an azimuth detecting device capable of accurately estimating a current azimuth of a vehicle by determining a usage ratio of azimuth data of a receiver.

[0010]

The azimuth detecting apparatus of the present invention for achieving the above object includes the following means (see FIG. 1). First means A for calculating the error of the offset value included in the angular velocity data of the gyro; of the error of the offset value and the error of the offset value included in the angular velocity data of the gyro obtained by the first means A Multiplying the time change rate by the elapsed time after offset estimation,
Second means B for evaluating the error contained in the current gyro angular velocity data based on the gyro scale factor error multiplied by the gyro angular velocity data; azimuth data of the moving GPS receiver A third directional error included in the evaluation of the azimuth error depending on at least the vehicle speed
Means C; the reliability of each data is calculated from the error of the gyro angular velocity data obtained by the second means B and the error of the azimuth data obtained by the third means C,
Fourth means D for calculating the Kalman filter gain; azimuth data obtained from the azimuth data of the GPS receiver and the angular velocity data of the gyro are subjected to weighting processing based on the Kalman filter gain to estimate the current vehicle. Fifth means E for obtaining the bearing.

[0011]

First, the error of the offset value included in the angular velocity data of the gyro is obtained by the first means A. For example, the error in the offset value can be obtained by statistically processing the data when the vehicle is stopped (the angular velocity data of the stopped gyro includes only the offset value). Next, by the second means B, the offset value error obtained by the first means A, the time change rate of the offset value error multiplied by the elapsed time after the offset estimation, and the gyro scale factor error The error contained in the angular velocity data of the gyro after the lapse of time is obtained based on the value obtained by multiplying the angular velocity data of the gyro by.

Next, the GPS which is moving by the third means C.
An azimuth error depending on at least the speed of the vehicle included in the azimuth data of the receiver is obtained. Then, by the fourth means D,
The Kalman filter gain is calculated by calculating the reliability of the output data of each sensor from the error of the gyro angular velocity data obtained by the second means B and the error of the azimuth data of the GPS receiver obtained by the third means C. Then, the azimuth data obtained from the output of the GPS receiver by the fifth means E,
And the azimuth data obtained from the angular velocity data of the gyro,
The current estimated heading of the vehicle is obtained by performing weighting processing based on the Kalman filter gain.

[0013]

Embodiments will be described in detail below with reference to the accompanying drawings showing embodiments. FIG. 2 is a block diagram showing an embodiment of the azimuth detecting device of the present invention applied to a vehicle position detecting device, and a vehicle speed sensor 41 (this sensor is used as a distance sensor for detecting the number of rotations of the left and right wheels). GPS receiver 4
2. Gyro 43 (selected from an optical fiber gyro that reads the turning angular velocity as a phase change of interference light, a vibration gyro that detects the turning angular velocity using the cantilever vibration technology of a piezo electric element, a mechanical gyro, etc.) .

The estimated heading of the vehicle is calculated based on the output data detected by the road map memory 2, the gyro 43, and the GPS receiver 42 which store the road map data, and the position of the vehicle is combined with the data of the vehicle speed sensor 41. Locator 1 for calculating the locator 1, a functional form r (v) of the azimuth error included in the azimuth data of the GPS receiver 42 connected to the locator 1.
(V is the speed of the vehicle), angular velocity data ω of the gyro 43
The data memory 6 storing the error qo of the gyro offset value included in (t), the gyro noise N, the error α of the gyro scale factor, etc., and the current vehicle position is displayed on the display 7 by overlaying it on the map and the keyboard. A navigation controller 5 for interfacing with 8 is included.

More specifically, the locator 1 obtains the rotation speed of the wheel by counting the number of output pulse signals from the vehicle speed sensor 41 with a counter, for example, and a predetermined value indicating the distance per count by a multiplier. Is calculated by multiplying the mileage data per unit time by calculating the relative change of the vehicle azimuth from the gyro 43, and the absolute azimuth output data of the GPS receiver 42 is used to calculate the vehicle azimuth by a method described later. The azimuth data is calculated.

The road map memory 2 stores road map data covering a predetermined range in advance, and a semiconductor memory, a cassette tape, a CD-ROM, an IC memory, a DAT or the like can be used. The display 7 uses a CRT, a liquid crystal display or the like to display a road map and a vehicle position while the vehicle is traveling.

The navigation controller 5 is
Consists of a graphics processor, image processing memory, etc.,
Searching the map on the display 7, switching the scale,
Scrolling, display control of the current position of the vehicle, etc. are performed.
The data memory 6 has a gyro offset error qo.
, The estimated change rate ε of the gyro offset error, the noise component N included in the gyro output, the error of the gyro scale factor (the ratio between the turning angle measured from the gyro output and the actual turning angle) α, and the GPS direction Error r (v)
Where r (v) = av ⁻¹ + b is stored, the values of constants a and b are stored.

Angular velocity data ω (t) of the gyro 43, GP
The azimuth data φ (t) of the S receiver 42 is sampled at regular time intervals, and if the time from the previous processing to the current processing is Δt, the number of samplings is proportional to the time Δt. The angular velocity data ω (t) of the gyro when the vehicle is stopped is originally 0, but an output appears if the gyro is offset. In estimating the gyro offset value, the value used for traveling before the vehicle is stopped is used as it is (of course, the angular velocity data ω (t) of the stopped gyro may be accumulated and averaged. ).

The error qo of the gyro offset value represents how much the gyro offset value varies. The error qo can be known by collecting a large number of angular velocity data ω (t) of the stopped gyro and obtaining the variance contained therein. The estimated change rate ε of the gyro offset error is a value empirically obtained as a function of temperature or the like.

A procedure for detecting the vehicle direction by the device having the above-mentioned configuration will be described. While the vehicle is traveling, the position of the vehicle is displayed on the display 7 together with the map. Even during the display, the output data of each sensor is fetched by interruption at regular intervals to update the vehicle direction. FIG. 3 shows the flow of vehicle direction detection at the time of this interruption. Note that this interruption may be performed for each fixed traveling distance obtained based on the traveling distance output data of the vehicle. The constant time or constant mileage depends on the type of gyro used and G
It is set as appropriate according to the performance of the PS receiver.

First, in step (1), the vehicle speed v,
Angular velocity data ω (t) of the gyro 43 and the GPS receiver 42
Take in the orientation data φ (t) of. Next, the data memory 6
The gyro offset value Ω, the gyro offset value error qo, the gyro offset error change rate ε, the noise component N, and the gyro scale factor error α, which are stored in the memory, are read (step (2)).

Next, the G stored in the data memory 6
The values of constants a and b representing the PS orientation error r (v) are read (step (3)). When the direction of the vehicle is calculated from the direction data of the GPS receiver 42, the smaller the speed of the vehicle is, the more the direction of the vehicle obtained includes the error. Therefore, the direction error of the vehicle is a monotone decreasing function of the speed. Therefore, the function r (v) = av ⁻¹ + b as described above is used for approximation. However, the function form is not limited to this, and a function form such as r (v) = a arctan (b / v) + c may be adopted. Further, not only the speed of the vehicle but also the GPS positioning error may be taken into consideration.

Then, the current estimated heading is obtained based on the read data. For that purpose, first, gyro 4
The variance q ² (t) of the angular velocity data ω (t) of 3 is calculated by the following equation (1) (step (5)). q ² (t) = (q o + εT) ² Δt ² + N ² Δt + (αΔθ) ² (1) where T is the elapsed time from the previous stop of the vehicle.
As described above, qo is a gyro offset correction error (including a quantization error) and is a constant. εT is the change rate of the gyro offset error multiplied by the elapsed time T, and is an error caused by the change (drift) of the gyro offset. The reason that the errors qo and εT are simply added is that these errors are not considered as independent events (if they are considered as independent events, the squares of the respective errors may be added together (Japanese Patent Laid-Open No. HEI 3). -279809 gazette))). N ² represents the variance due to noise. Δt of 2
No power is added because the noise error N is proportional to the 1/2 power of the number of times of integration.

Δθ is the turning angle from the previous processing to the present processing, and αΔθ represents the distribution of the turning angle caused by the error of the scale factor. The value of α varies depending on the ambient temperature and the like, and is a value determined as a gyro standard for each temperature range from how many times to how many times. Therefore, the value is used as it is. Next, GPS
The variance r ² (t) of the azimuth data φ (t) of the receiver 42 is obtained by r ² (t) = (av ⁻¹ + b) ² .

In step (6), q ² (t), r
^{Using 2} (t), the estimated bearing θ considering the error is calculated based on the following equation. θ (t) = K (t) φ (t) + (1-K (t)) (θ (t-1) + ω
(t)) where θ (t) is the estimated heading obtained by this interruption, θ (t)
(t-1) is the previously obtained bearing. ω (t) and φ (t) are the output data of the sensor used when calculating the estimated bearing this time. K (t) is a variable such that 0 <K (t) <1 and is Kalman gain. Using K (t) Kalman gain K obtained last time (t-1), K ( t) = (q 2 (t) + σ 2 (t-1)) / (q 2 (t) + r 2 (t ) + Σ
² (t-1)).

The variance of the estimated azimuth is calculated by σ (t) = √ (K (t) r ² (t)). As described above, the average and variance of the gyro offset estimation value, the error of the change rate, the noise component,
When the error of the gyro scale factor and the error included in the output of the GPS receiver are obtained and stored and the estimated heading of the vehicle is obtained, the variance included in the data of both sensors is obtained from the stored data, and each data is weighted. It is possible to obtain the estimated bearing with.

When the GPS azimuth data cannot be obtained for some reason, the estimated azimuth θ is expressed as θ (t) = θ (t-1) + ω (t) with K (t) = 0. To be done. The estimated position of the vehicle can be calculated from this azimuth and the distance data of the vehicle speed sensor 41.

At this time, of course, a map matching method may be employed in which the estimated position of the vehicle is corrected by comparing the degree of correlation with the road map data, the estimated position of the vehicle is corrected, and the current position of the vehicle is set on the road. Good (JP-A-63-148115)
No., Sho 64-53112). Although the azimuth detecting apparatus of the present invention has been described above based on the embodiments, the present invention is not limited to the above embodiments. In carrying out the azimuth detecting apparatus of the present invention, it is needless to say that various errors that are usually considered, such as an error other than those described above, such as a quantization error that occurs during A / D conversion, may be incorporated. . Besides, various design changes can be made without departing from the scope of the present invention.

[0029]

As described above, according to the azimuth detecting apparatus of the present invention, the error factors included in the azimuth data of the gyro and the GPS receiver are individually analyzed and evaluated, and thus the G
The Kalman gain, which is the ratio between the usage rate of the PS azimuth data and the usage rate of the gyro angular velocity data, can be set to an appropriate value, and the current azimuth of the vehicle can be accurately estimated.

[Brief description of drawings]

FIG. 1 is a functional block diagram of an azimuth detecting device of the present invention.

FIG. 2 is a block diagram showing a hardware configuration of a vehicle position detection device including an azimuth detection device.

FIG. 3 is a flowchart showing an azimuth detection procedure.

[Explanation of symbols]

 1 Locator 6 Data Memory 42 GPS Receiver 43 Gyro

Claims (1)

[Claims] 1. A bearing data acquired from a GPS receiver,
And the azimuth data obtained from the angular velocity data of the gyro are taken in, and the azimuth detection device that finds the current estimated azimuth of the vehicle from those values and the past azimuth is used to calculate the error in the offset value included in the gyro angular velocity data. The first means (A) for performing the offset value error and the time change rate of the offset value error included in the angular velocity data of the gyro obtained by the first means (A) The second means (B) for evaluating the error contained in the current gyro angular velocity data based on the product of the gyro scale factor error and the gyro scale factor error multiplied by the gyro angular velocity data. Means for evaluating a bearing error depending on at least the speed of the vehicle included in the bearing data of the GPS receiver inside
(C), the error of the gyro angular velocity data obtained by the second means (B), and the error of the azimuth data obtained by the third means (C) are used to calculate the reliability of each data, A fourth means (D) for calculating the Kalman filter gain, the azimuth data obtained from the azimuth data of the GPS receiver, and the angular velocity data of the gyro are subjected to weighting processing based on the Kalman filter gain to obtain the current state of the vehicle. And a fifth means (E) for obtaining an estimated azimuth of the azimuth.