JPH04364416A - Method for determining orientation - Google Patents

Abstract

PURPOSE:To enable a program to be simple and a calculation time to be reduced by eliminating a disturbance constituent which is included an angle data thetain nearly a complete form by compensating for the angle data theta with a digital filter for obtaining F theta. CONSTITUTION:An orientation sensor 1 converts an input magnetic vector into voltage signals Vx and Vy in x and y directions and then outputs them. thetaindicates an angle which is formed by a reference axis of the sensor 1 and a direction of an earth magnetism vector. A/D conversion portions 2-1 and 2-2 perform A/D conversion of the signals Vx and Vy. An angle data conversion portion 3 receives A/D conversion results AVx and AVy according to conversion portions 2-1 and 2-2 and then obtains an angle data Q according to an expression (3). A digital filter 4 allows the data theta to be passed through a digital low- pass filter and to be compensated by an expression (4) for obtaining Ftheta, thus preventing scattering of the orientation data. Namely, the filter 4 completely eliminates a disturbance constituent which is contained in the data theta.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a direction determining method suitable for application to a direction meter system mounted on a vehicle.

0002

BACKGROUND OF THE INVENTION Conventionally, as a direction determination method of this type, outputs Vx and Vy of a direction sensor are A/D converted.
An azimuth determining method is employed in which angle data θ is obtained by processing the /D conversion results AVx and AVy, and a detection azimuth is determined based on this angle data θ.

0003

However, in such a conventional orientation determination method, a problem arises in that the orientation data flickers. To explain the flickering of this orientation data, magnetic flux tries to pass through objects with higher magnetic permeability, so when a ferromagnetic material is placed in a parallel magnetic field, the magnetic flux is always attracted to the ferromagnetic material, and the parallel magnetic field It will no longer be. The same goes for geomagnetism; the geomagnetic field is curved around bridges, buildings, large trucks, etc., and the geomagnetic vector does not necessarily point to magnetic north. Therefore, if the angle data θ is displayed as is when the vehicle is passing through this kind of surroundings, the display will flicker.

[0004] As a countermeasure against this, A/D conversion results AVx,
At first glance, it seems that disturbance can be removed by passing AVy through an averaging filter (such as a low-pass filter) and then obtaining angle data θ. However, if we assume a situation where the vehicle 7 passes beside a high permeability object 6 such as a truck (Fig. 7(
a)), a kind of rotation vector is detected as shown in FIG.
The amplitudes of the y sin and cos waves were only reduced, but the angle data θ was not affected, and disturbance removal was incomplete. In other words, if the noise components included in the A/D conversion results AVx, AVy are x'=sinωt, y'=cosωt, filtering results in x'/a, y'/a (a is the attenuation at ωt). rate), θ'=tan-1
[(y'/a)/(x'/a)]=tan-1(y'/
x'), and the angle data θ varies by θ'.

[0005]

[Means for Solving the Problems] The present invention has been made to solve such problems, and the output of the direction sensor is A/D converted, and angle data is obtained by processing the A/D conversion results. Obtain θ, correct this angle data θ, and obtain Fθ
The detection direction is determined based on the angle data F.theta. according to the allocation status of the angle data .theta. in one direction from a predetermined direction.

[0006]

Therefore, according to the present invention, by correcting the angle data θ using, for example, a digital filter to obtain Fθ, it becomes possible to almost completely remove the disturbance component contained in the angle data θ.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The orientation determining method according to the present invention will be explained in detail below.

FIG. 1 is a block diagram showing an outline of signal processing in an azimuth meter system to which this azimuth determining method is applied. In the figure, 1 is a direction sensor, 2-1 and 2-
2 is an A/D conversion section, 3 is an angle data conversion section, 4 is a digital filter, and 5 is an orientation determination section. The direction sensor 1 converts the input magnetic vector into voltage signals Vx (formula (1) below) and Vy (formula (2) below) in the x and y directions, and outputs the voltage signals. However, in equations (1) and (2), θ is the angle formed between the reference axis of the orientation sensor 1 and the direction of the geomagnetic vector.

[0009]

[Math 1]

0010

[Math 2]

A/D converters 2-1 and 2-2 A/D convert the voltage signals Vx and Vy. The angle data converter 3 receives the A/D conversion results AVx and AVy from the A/D converters 2-1 and 2-2 as input, and calculates the angle data θ using equation (3) below. Note that the resolution of the angle data θ is 360°/256.

0012

[Math 3]

The digital filter 4 corrects the angle data θ through a digital low-pass filter to obtain Fθ (formula (4) below) to prevent flickering of the azimuth data. That is, in this embodiment, the digital filter 4 completely removes the disturbance components included in the angle data θ.

[0014]

[Math 4]

[0015] Note that the disturbance component included in the angle data θ may be removed by a method such as averaging without using a digital filter.

[0016] In the A/D conversion result, it is necessary to pass the two signals AVx and AVy through a filter.
The angle data θ only needs to be passed through the filter once. Since the filter operation requires several arithmetic operations, the program can be simplified and the calculation time can be shortened. Furthermore, since the cutoff frequency and attenuation rate of the filter are determined by the number of data and the sampling frequency, shortening the calculation time leads to an improvement in the degree of freedom in filter design.

In the above description, the digital filter 4 has a filter type of an FIR digital filter, and the window is a Hamming window. In addition, in the above equation (4), the data strings are θ(-7), θ(-6), ... θ(0), θ(1),
...θ(6), θ(7) (latest)
→ → → →
(oldest), and the Hamming window Wn is obtained from the following equation (5) (see FIG. 4). Note that the digital filter is a type of weighted averaging method. In other words, the filter function is realized by adding the results of multiplying the new and old data by coefficients (called windows) (Reference: Beginner's Digital Filter
Shogo Nakamura (Tokyo Denki University Press).

[0018]

[Math 5]

The azimuth determination unit 5 receives the output Fθ of the digital filter 4 as an input, and according to the allocation situation of the angle data θ shown in FIG. 3, that is, the allocation situation of the angle data θ counterclockwise from the azimuth "N". , determine the detection direction.

[0020] In such a compass system, there is a new feature:
The following problem occurred. In other words, since Fθ is obtained by correcting the angle data θ, as a result of the assignment of angle data θ from the direction “N” to the counterclockwise direction, when the direction to be detected changes clockwise, However, a problem arises in that the direction of change occurs in a mismatch between the orientation to be detected and the determined detection orientation.

That is, in FIG. 5, at time t0, the vehicle is facing the "NW" direction, and by turning clockwise, it is facing the "N" direction at t1, and at t2, it is facing the "NW" direction.
Let's imagine a situation in which the vehicle heads in the direction of "E". In this case, when looking at changes in the angle data θ in FIG. 6, the angle data θ is t0
It starts decreasing from time t1 and becomes 0 at t1. Immediately after that, it jumps to FE[HEX] and then starts to decrease again. This t
When there is a data change from 0 to t2, the output Fθ of the digital filter 4 changes as shown by the broken line in FIG. Therefore, the direction of the vehicle, that is, the direction to be detected is “NW
” → “N” → “NE”, while the determined detection direction changes from “NW” → “N” → “W” → “S”.
” → “E” → “NE” and changes counterclockwise with the direction “N” as the border, that is, a gap between 0H and FE [HEX] occurs with the direction “N” as the border, which should be detected. A mismatch occurs in the direction of change between the orientation (rotation direction of the vehicle) and the determined detection orientation (display orientation).

For this reason, in this embodiment, the discrepancy between the direction of change and the determined detection direction, which occurs when the direction to be detected changes clockwise, is made to match by performing offset processing.

A direction determining method using this offset processing will be explained below with reference to FIGS. 1 and 2.

In this embodiment, the currently determined detection direction is “W”.
-N-E" section and the "W-SE" section, the calculation method of F.theta. given to the direction determining section 5 is different.

That is, the current determined detection direction is "W-N
-E” interval, a certain constant (in this example, 8
0[HEX]) is added to each angle data θ(n), and each added angle data θ(n)' is provided to the digital filter 4. Then, processing is executed in the digital filter 4, and the result obtained by subtracting the above constant from the output Fθ′ obtained thereby is set as Fθ, and this Fθ is provided to the orientation determining section 5. In other words, if the current determined detection direction is in the "W-N-E" section, the data string is θ(-7)'=θ(-7)+
80H, θ(-6)'=θ(-6)+80H,...θ
(6)'=θ(6)+80H, θ(7)'=θ(7)+
80H, and a carryover (increase in 8-bit data) may occur at this time, but the 9th bit is ignored and only the lower 8 bits are treated as data. Find Fθ' as the output of the digital filter 4 using the following formula (6),
Subtract 80H from this Fθ' to obtain Fθ (Fθ=F
θ'-80H), a borrow may occur at this time, but the borrow is ignored. This Fθ is given to the orientation determining section 5. The broken line shown in Fig. 2 shows the change in Fθ when there is a data change from t0 to t2, and the direction of the vehicle, that is, the direction to be detected changes clockwise from "NW" → "N" → "NE". However, the determined detection direction is also “NW
"→"N"→"NE" in a clockwise direction, and the direction to be detected (rotation direction of the vehicle) and the determined detection direction (display direction) match.

[0026]

[Math 6]

[0027] When the current determined detection direction is in the "W-S-E" section, there is no need for the above-mentioned offset processing applied when it is in the "W-N-E" section, and each angle data θ(n) is given to the digital filter 4 as it is. Then, Fθ is obtained as the output of the digital filter 4 using the above equation (4), and this Fθ is provided to the orientation determining section 5.

[0028] In the above description, the prerequisite for sampling the azimuth data is that the speed is sufficiently faster than the turning speed of the vehicle.

[0029]

As is clear from the above explanation, according to the present invention, according to the first invention (the invention according to claim 1), the angle data θ is corrected by, for example, a digital filter to become Fθ. As a result, it is possible to almost completely remove the disturbance component included in the angle data θ, and A
Compared to the method of filtering the /D conversion results, the program can be simplified and the calculation time can be shortened.
It becomes possible to improve the degree of freedom in filter design. Further, according to the second invention (invention according to claim 2), the direction of change between the direction to be detected and the determined detection direction can be made the same for any direction change, and is applicable to a direction meter system mounted on a vehicle. In this case, it is possible to avoid problems such as the direction of rotation of the vehicle and the direction of change in the display orientation not matching.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an outline of signal processing in a direction meter system to which a direction determining method according to the present invention is applied.

FIG. 2 is a diagram illustrating a situation where there is no mismatch in the direction of change between the direction to be detected and the determined direction of detection.

FIG. 3 is a diagram showing the allocation status of angle data θ.

FIG. 4 is a diagram illustrating a humming wind.

FIG. 5 is a diagram showing a clockwise turning situation of the vehicle from time t0 to t2.

FIG. 6 is a diagram illustrating a situation where a mismatch occurs in the direction of change between the orientation to be detected and the determined detection orientation.

FIG. 7 is a diagram illustrating a detection situation of a rotation vector when a vehicle passes along the side of a high magnetic susceptibility object.

[Explanation of symbols]

1 Direction sensor 2-1, 2-2 A/D conversion section 3　　　　　　Angle data conversion section 4 Digital filter 5 Direction determining section

Claims (2)

[Claims] Claim 1: A/D converting the output of the azimuth sensor, processing the A/D conversion result to obtain angle data θ, correcting this angular data θ to obtain Fθ, and converting the output from a predetermined direction in one direction. An azimuth determining method for determining a detection azimuth based on the angular data Fθ according to the assignment status of the angular data θ to the angular data Fθ. [Claim 2] Claim 1, characterized in that the discrepancy in the direction of change with the determined detection direction that occurs when the direction to be detected changes in another direction is made to match by performing offset processing. Direction determination method.