JPH03291516A - Stable value extracting method for earth magnetism azimuth sensor - Google Patents

Abstract

PURPOSE:To accurately process the traveling direction of a vehicle by removing respective heretical data according to a evaluation reference value of evaluation for the running mean value of absolute values of variation angles of several past absolute azimuths of the earth magnetism azimuth sensor. CONSTITUTION:Output data from the earth magnetism azimuth sensor 1, a wheel speed sensor 2, and a travel distance sensor 2 are inputted to an arithmetic part 5 through a sensor interface 4. The arithmetic part 5 performs total evaluation as to variance of data on coordinates, azimuths, and radio according to the XY component output values of the sensor 1 and several data of detected absolute azimuths including current values. The coordinate variance is evaluated as the ratios of several mean coordinate points including the current values, values between individual coordinate points, and the optional set value of the azimuth radius of the sensor 1. Further, azimuth variance is evaluated as the mean value of azimuth variation angles of several absolute azimuths including the current value. Further, radius variance is evaluated as the ratio of the optional set value of the azimuth circle radius of the sensor 1 and the current value.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a method for extracting stable values of a geomagnetic azimuth sensor, in which a geomagnetic azimuth sensor and a wheel speed sensor are mounted on a vehicle, and stable data is extracted by removing irregular data due to the influence of magnetic disturbance.

[Conventional technology]

FIGS. 8 and 9 are block diagrams of a device showing a conventional magnetization correction method for a geomagnetic azimuth sensor disclosed in, for example, Japanese Unexamined Patent Publication No. 63-128222, and an explanation of the principle of output change of the geomagnetic azimuth sensor when the azimuth changes. In the figure, 71
72 is a geomagnetic azimuth sensor, 72 is a vehicle azimuth change detecting means, such as a wheel speed sensor, and 73 is a calculation means. Moreover, FIG. 9 shows the geomagnetic direction sensor 71 when the direction changes.
It is an explanatory diagram of output change. First, when the vehicle changes by A, the azimuth sensor output also changes from A to B. In this case, assuming that the radius of the azimuth circle of the geomagnetic azimuth sensor 71 does not change before and after the azimuth change, as is clear from FIG. It is an equilateral triangle, and the vertex of the triangle has an apex angle of θ. In other words, if the amount of change θ in the vehicle's azimuth, the rotation direction, the output A of the geomagnetic azimuth sensor 71 before the azimuth of the vehicle changes, and the output B of the geomagnetic azimuth sensor 71 after the azimuth has changed, the center of the azimuth, that is, the amount of magnetization can be found. FIG. 6 is an explanatory diagram of output changes and calculations of a magnetic orientation sensor 71. FIG. New magnetization occurs while the vehicle is moving, and the amount of magnetization becomes ΔX-
Suppose that it changes to ΔXI+ Δy−ΔyI. Furthermore, the relatively stable output of the geomagnetic direction sensor 71 before curving is AP(XI, Y+), and the relatively stable output of the geomagnetic direction sensor 71 after curving is BP(XZ,
3'z). First, the output AP(XI, )'I of the geomagnetic direction sensor 71 before the vehicle changes direction
) is stored in the calculation means 73. Next, when the vehicle changes direction, the wheel speed sensors 7 for the left and right wheels
2 etc., the amount of change in orientation θ is determined. Figure (C) shows the vehicle before the azimuth change, and Figure (D) shows the change angle θ after the azimuth has changed due to the counterclockwise rotation of the vehicle. It is assumed that θ determined by the left and right wheel speed sensors and θ are the same angle. After direction change, direction sensor output Bp (Xz, yz)
is found. These APIs (XI +y+) IBP(X
From Z + 3' and θ3, the radius for the azimuth before and after the azimuth change is constant, and the length between AP, B, L = (XI xz)" + (y + V
z) With 2 and Uni, the direction of Hec I Le A,,B is θA
p, llP, θA9+BF− tanri [shiL2V+)(−π<θApBP〈K
x2-x. If the direction of hector SO1 is θsap, π θSOF ′: θAPBP+ 2 Center of magnetization○, (ΔX1+Δy+) is) (8Xl+Δy+)− (SX+fcosθSOP+SV+ρsinθ3o,)
is required.

[Problem to be solved by the invention]

The conventional magnetization correction method for a geomagnetic azimuth sensor is performed as described above, so the idea is to average the magnetic information that includes disturbance magnetism other than geomagnetism among the magnetic information detected using the geomagnetic azimuth sensor. is effective against isolated magnetic disturbances in places where the geomagnetic environment is stable, but it is difficult to suppress detection errors when the geomagnetism itself is distorted, such as on an elevated road, and the absolute orientation is biased. The problem was that it couldn't be done. In addition, the idea that the radius of the azimuth circle used for magnetization correction of the geomagnetic azimuth sensor is constant before and after the azimuth change is based on the idea that the radius of the azimuth circle used for magnetization correction of the geomagnetic azimuth sensor is fixed when the geomagnetic environment If is stable, then
Since the radius of the azimuth circle can be stably obtained as the radius changes, it is possible to detect changes in the vehicle body magnetization state and prevent erroneous corrections. There were problems, such as when the magnetization state of the vehicle body could not be changed and there was a risk of erroneous correction being executed. This invention was made to solve the above problems, and is based on several past data including the XY component output values of the geomagnetic azimuth sensor and the current values of the detected absolute azimuth. A method for extracting stable values of a geomagnetic azimuth sensor that comprehensively evaluates the variations in each data of coordinates and azimuth radius, and removes heretical data by assigning a rank (grade) of detection accuracy to the data of the geomagnetic azimuth sensor. The purpose is to obtain.

[Means to solve the problem]

The stable value extraction method of the geomagnetic direction sensor according to the present invention is as follows:
Coordinates are calculated based on several pieces of data including the χY component output value of the geomagnetic direction sensor and the current value of each detected absolute direction.
The arithmetic unit comprehensively evaluates the variation in each data of direction and radius as an evaluation target, and assigns a detection accuracy grade to the data of the geomagnetic direction sensor. It is evaluated as the ratio of the average coordinate point obtained by moving the XY coordinate points of The variation is evaluated as the magnitude of the average value obtained by moving the absolute value of the azimuth change angle of the past several absolute azimuths including the current value, and the radial variation is the magnitude of the average value of the azimuth circle radius of the geomagnetic azimuth sensor. It is designed to be evaluated as a ratio between an arbitrary setting value and the current size.

[Effect]

The stable value extraction method of the geomagnetic azimuth sensor in this invention is the same as the evaluation of the ratio of the current value to an arbitrary setting value of the azimuth circle radius of the geomagnetic azimuth sensor (the current value for an arbitrary setting value of the azimuth circle radius). Past several X, Y including
Evaluate the size ratio between the coordinate points of the χY coordinate point and the moving averaged average coordinate point using the component output value, and move the absolute value of the change angle of the absolute value azimuth for several past values including the current value. Stable data is extracted by removing heretical data based on each evaluation standard value of the evaluation of the magnitude of the average value, and removing all heretical data that cannot be removed by individual evaluation.

[Embodiment of the Invention] An embodiment of the present invention will be briefly described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a mobile navigation device according to an embodiment of the present invention, in which numeral 1 denotes a geomagnetic azimuth sensor as a first azimuth sensor that detects absolute azimuth based on geomagnetism; 3 is a wheel speed sensor serving as a second azimuth sensor that detects changes in the orientation of the vehicle due to differences in left and right wheel speeds of the vehicle; 3 is a mileage sensor that detects the distance traveled by the vehicle as a moving object; 1~
3 is input to the sensor interface 4. 5 extracts the stable value of geomagnetic sensor data based on the output data of each sensor 1 to 3 sent sequentially, the distance traveled by the vehicle,
An arithmetic unit calculates the heading direction and the position of the own army, and 6 is a display unit that draws the position of the own army. Next, as evaluation targets for extracting stable values of the output data of the geomagnetic direction sensor 1, coordinate variations, direction variations,
and radius variations, etc., and the evaluation methods for each are shown below. As shown in Figure 2 (b), the coordinate variation is used to determine magnetic disturbances including disturbance magnetism other than geomagnetism, and is determined by the self-learning value (arbitrary set value) of the radius of the azimuth circle of the geomagnetic azimuth sensor 1.
For RAi, divide by the ratio ΔL/RA of the average coordinate point obtained by moving the moving average of several past XY coordinate points including the current value and the size ΔL between the coordinate points of the individual coordinate points used for the average. This will be evaluated in three stages: large, medium, and small. As shown in Figure 2 (c), the direction variation is used to determine the straightness of the vehicle running condition, and is a moving average of the absolute values of the change angles of the past several absolute directions θ, including the current value. By classifying the size of the average value, it is evaluated in three stages: large, medium, and small. In addition, the radius variation is used to determine the magnitude of disturbance magnetism other than geomagnetism, as shown in Figure 2 (A), and is the magnitude of the current value with respect to the self-learning value RA of the radius of the azimuth circle of the geomagnetic azimuth sensor 1. It is evaluated in three stages: large, medium, and small by classifying it by the ratio of R/RA. An example of the evaluation fi[[i standard value used in evaluating these variations is shown in Part 3. In the 31st ki, a is the coordinate variation, b is the force position variation, and C is the evaluation reference value of the radius variation. Figure 4 is a diagram showing the grade classification of stable value extraction, and is based on the evaluation results of the coordinate range and orientation variation obtained in the third round.From this, there are five grades from 1 to 5, with 1 being the most stable. I'm spelling it out in the same way. Next, the operation will be explained using the flowcharts shown in FIGS. 5 and 6. First, in step ST41 in FIG. 5Q, initial processing of the device, initial display, and rotation correction of the geomagnetic azimuth sensor 1 are executed. Also, the self-learning value of the diameter of the azimuth circle f of the geomagnetic azimuth sensor 1 is initialized here. Ste, bu ST4
In step 2, the system waits until the vehicle has moved a certain distance or more. In step 5T43, the geomagnetic direction sensor 1 and the wheel speed sensor 2
The output data is inputted to the calculation section 5 via the sensor interface 4. In step ST44, the absolute heading and the change in the heading of the vehicle are calculated. In step 5T45, variations in coordinates and orientation are evaluated and detection accuracy is graded for extracting stable values of geomagnetic orientation sensor data. Step 5
At T46, the traveling direction of the vehicle is calculated using equation (1). In equation (1), θ and θi-1 are the current and previous values of the vehicle's heading, JK is the absolute heading, φ is the change in the vehicle's heading, and k is a coefficient that changes the value depending on the class.The weighting of the absolute heading is a coefficient. are shown respectively. θ,=θ=−++(JK−θ1−1)×+ψX (1−k) (1) In step ST47, the magnetization of the geomagnetic direction sensor 1 is corrected. In step ST48, each data is stored in the memory and the vehicle position is updated. In FIG. 6, in step ST51, the turning angle and turning distance of the vehicle are updated, and in step 5T52, it is determined whether the detection accuracy class is class 1 (moving straight with small magnetic disturbance), and if it is not class 1, it is determined. It exits the magnetization correction process and executes step 5T53 if it is class 1. In step ST53, it is determined whether the settings of the X and Y component output values before the azimuth change and the turning angle and turning distance of the vehicle are valid, and whether the X and Y component output values before the azimuth change are not set or the turning angle is less than 60° or more than 120°, or if the turning distance exceeds 200m, execute step ST54; otherwise, execute step ST54.
Execute 55. Here, the reason why the turning angle and turning distance are included in the processing conditions is to remove data before and after changes in the magnetic environment and to enable correction based on left and right turning movements of the vehicle. In step 5T54, X before the direction change. The Y component output value is set and the turning angle and turning distance of the vehicle are cleared to zero, and then the process exits from the magnetization correction process. In step ST55, variation evaluation of the radius of the azimuth circle of the geomagnetic azimuth sensor 1 is performed (determination of the magnitude of disturbance magnetism other than the terrestrial magnetism). In step 5T56, if the evaluation result in step 5T55 is large, the magnetization correction process is skipped, and if not, step ST57 is executed. In step ST57, the self-learning value of the azimuth circle radius of the geomagnetic azimuth sensor 1 is updated. In step ST58, if the evaluation result in step ST55 is within the range, the magnetization correction process is exited, otherwise step ST59 is executed. In step 5T59, the amount of magnetization is calculated and corrected based on equations (2) and (3) based on the relationship between the X and Y component output values before and after the azimuth change and the turning angle of the vehicle shown in FIG. In equations (2) and (3), χ1. Y1 and X2. Y2 is the X and Y component output values before and after the azimuth change, ψ is the turning angle of the vehicle, X CO + YC6 and X,
Y is the fixed center X and Y component output value, δX, δ, which is determined by the circuit as the estimated center of the azimuth circle of the geomagnetic orientation sensor
Y is the XY component output value of the amount of magnetization. δX=Xc, -X. δY=Yc, -Y. In addition, in the above embodiment, coordinate variations, azimuth variations,
Although the individual evaluation methods, usage, and grading of radial variation and radial variation have been described, grading by other evaluation methods may also be used and the same effects as in the above embodiments can be achieved. Further, in the magnetization correction of the geomagnetic sensor according to this embodiment, the X and Y component output values before and after the azimuth change are used as raw coordinate points, but the average coordinate point of the averaged X and Y component output values may be used. Effects of the Invention As described above, according to the present invention, it is possible to evaluate the ratio of the current value to an arbitrary set value of the radius of the azimuth circle of a geomagnetic azimuth sensor, and also to Evaluation of the size ratio between X and Y coordinate points and the moving averaged average coordinate point using several past XY component output values including the current value, and several past XY component output values including the current value. Since each heretical data is removed based on the evaluation standard value of the magnitude of the average value obtained by moving the moving average of the absolute value of the change angle of the absolute azimuth, the magnitude of magnetic disturbance including disturbance magnetism is removed. It is possible to remove outlier data from the viewpoint of the magnitude of magnetic disturbance, the magnitude of azimuth disturbance, and accurate processing such as correction of vehicle heading and magnetization of the geomagnetic azimuth sensor. Moreover, there is an effect that the geomagnetic direction sensor data can be easily used (weighted).

[Brief explanation of drawings]

Fig. 1 is a configuration diagram of a vehicle position detection device according to an embodiment of the present invention, Figs. 2 (a) to (c) are explanatory diagrams of evaluation of variation targets, and Fig. 3 is an example of evaluation reference values used in variation evaluation. Figure 4 is an explanatory diagram of stable value extraction,
5 and 6 are flowcharts showing the operation of the present invention, FIG. 7 is an explanatory diagram of the X and Y component output values and vehicle turning angles before and after changes in azimuth, and FIG. 8 is a diagram showing magnetization in a conventional vehicle detection device. A configuration diagram of the correction, Fig. 9 is an explanatory diagram of the output change of the geomagnetic azimuth sensor when the direction changes, Fig. 10 (a) is an explanatory diagram of the output change and calculation of the geomagnetic azimuth sensor, and the same figure (b) is an explanatory diagram of the output change of the geomagnetic azimuth sensor and calculation by the left and right wheel speed sensors. Figures (C) and (D) are diagrams illustrating the vehicle's orientation change. In the figure, 1 is a geomagnetic azimuth sensor (first azimuth sensor), 2 is a wheel speed sensor (second azimuth sensor), 3 is a travel distance sensor, and 5 is a calculation unit. In addition, in the figures, the same reference numerals indicate the same or equivalent parts.

Claims (1)

[Claims] Based on several past pieces of data including the X and Y component output values of the geomagnetic azimuth sensor and the current values of the detected absolute azimuth, the arithmetic unit evaluates the variations in the coordinates, azimuth, and radius data. After a comprehensive evaluation, the data of the geomagnetic direction sensor was graded for detection accuracy, and the coordinate variation was determined using the average coordinate point obtained by moving the average of several past XY coordinate points including the current value. It is evaluated as the ratio of the size between the coordinate points of each coordinate point and the arbitrary set value of the azimuth circle radius of the geomagnetic azimuth sensor, and the azimuth variation is calculated based on the absolute azimuth of several past values including the current value. The absolute value of the azimuth change angle is evaluated as a moving average value, and the radius variation is evaluated as a ratio between an arbitrary set value and a current value of the azimuth circle radius of the geomagnetic azimuth sensor. A method for extracting stable values from a geomagnetic orientation sensor.