JPH0462419A - Azimuth detector - Google Patents

Abstract

PURPOSE:To accurately calculate the traveling direction of a vehicle without using an earth magnetism sensor which is intolerant of magnetic disturbance by combining an azimuth sensor which is hardly affected by disturbance like yaw rate sensor and a wheel speed sensor. CONSTITUTION:A straight traveling detecting means 4 finds a difference in number between pulses from left and right vehicle speed sensors 1 and 2 and decides the straight traveling state of a mobile object from whether or not the difference is larger than a constant value continuously. A zero-point calculating means 5 adds the data from the yaw rate sensor 3 in the section where the straight traveling state is decided and corrects the zero point of the yaw rate sensor 3 from the addition value and the addition time of the yaw rate sensor data when the addition time exceeds a constant time. Then an azimuth calculating means 6 calculates the traveling direction of the mobile object from the zero point of the yaw rate sensor 3 which is found by a calculating means 5.

Description

[Detailed Description of the Invention] Industrial Field of Application The present invention is useful for determining the current position of a moving body such as a car and displaying the current position along with a surrounding road map, or for searching and searching for a route to a destination. This relates to a direction detecting device for calculating the traveling direction of a moving object used in an on-vehicle navigation system for guiding etc. Conventional technology A conventional direction detecting device is disclosed in, for example, Japanese Patent Laid-Open No. 61-2.
However, as shown in this example, the conventional direction detection device is equipped with a wheel speed sensor that measures the number of revolutions of the left and right wheels on the same axis of the vehicle as rotation angle pulses, and uses the output from the wheel speed sensor. The moving distance of the left and right wheels is calculated, the turning angle of the moving object is calculated from the ratio of the difference between the moving distance of the left and right wheels and the distance between the left and right wheels (hereinafter referred to as tread), and the moving direction of the moving object is calculated by integrating the turning angle. Also
It is also equipped with a yaw rate sensor and calculates the turning angle of the moving object by integrating its output, and calculates the moving direction of the moving object by integrating the output. In addition, a geomagnetic direction sensor is used as a conventional direction detection device. In this case, the earth's magnetic field is measured by two coils that are perpendicular to each other in a plane parallel to the earth, and from those values the vehicle's heading is calculated as an absolute heading. Also, in order to compensate for the characteristics of each sensor, Problems to be Solved by the Invention However, the above configuration does not solve the problem of combining sensors that calculate absolute and relative directions, such as a geomagnetic azimuth sensor and a yaw rate sensor, or a geomagnetic azimuth sensor and a wheel speed sensor. Friend: When using wheel speed sensors, the constants that convert rotation angle pulses into travel distance (hereinafter referred to as wheel constants) fluctuate during driving, causing errors in the calculated turning angle of the vehicle.Also, when using yaw rate sensors, On the other hand, when using a geomagnetic azimuth sensor, the azimuth may shift locally because the geomagnetic azimuth sensor is vulnerable to magnetic disturbances. In addition, since the vehicle is made of magnetic material, magnetization phenomenon also occurs, and countermeasures are essential.Therefore, by combining a geomagnetic direction sensor with a wheel speed sensor or a yaw rate sensor, the result is a highly accurate vehicle. The present invention takes into account these points and aims to provide a direction detection device that accurately calculates the direction of travel of a vehicle without using a geomagnetic sensor that is susceptible to magnetic disturbances. Means for Solving the Problems The present invention aims to solve the above-mentioned problems.The present invention is a method for solving the above-mentioned problems. It is an azimuth detection device characterized by calibrating the zero point of the yaw rate sensor in a straight-ahead running state and calculating the traveling direction of the moving object using the output of the yaw rate sensor and the calibrated zero point. is received, the current position of the moving object is measured, the straight running state of the moving object is detected from the history of the measured current position, the zero point of the yaw rate sensor is calibrated in the detected straight running state, and the output of the yaw rate sensor and This method is characterized in that the traveling direction of the moving object is calculated using the calibrated zero point.

Furthermore, it is equipped with a yaw rate sensor and a wheel speed sensor, and when it is determined that the vehicle is running straight, it weights the turning angle calculated from the travel distance of the left and right wheels determined by the wheel speed sensor.
The present invention is also characterized by comprising means for estimating the turning angle of the moving object by weighting the turning angle obtained from the yaw rate sensor and calculating the traveling direction of the moving object when it is determined that the moving object is not in a straight running state. be.

Effect of the Invention With the above-described configuration, the present invention uses the output of the wheel speed sensor to determine the straight running state of the moving body even when using a yaw rate sensor with a large zero point fluctuation, and determines the straight running state of the moving body by calibrating the zero point of the yaw rate sensor in the straight running section It is possible to calculate the heading with high accuracy without being affected by magnetic disturbances.

MADOYU Determines the moving state of the moving object from the history of the current position of the moving object calculated by receiving radio waves from the satellite.While calibrating the zero point of the yaw rate sensor in the straight moving section, the moving object's output from the yaw rate sensor is determined. Insertion for calculating the heading It is possible to calculate the heading with high accuracy without receiving magnetic disturbances4 Furthermore, when it is determined that the vehicle is traveling straight, the wheel speed sensor, which has a highly stable heading, is used to calculate the heading. A weight is given to the obtained turning angle, and when it is determined that the moving object is not in a straight running state, the turning angle of the moving object is estimated by weighting the turning angle obtained from a yaw rate sensor with high turning angle calculation accuracy, and the moving object is moved forward. A table that calculates the heading It becomes possible to calculate the heading with high accuracy without being affected by magnetic disturbances.

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a block diagram of a direction detecting device in a first embodiment of the present invention. In FIG. 1, 1 is a left wheel speed sensor, and 2 is a right wheel speed sensor. This is a coil that detects the rotation of a gear-shaped rotor attached to the axle hub of the driven wheel of the vehicle, and a sinusoidal output is obtained in proportion to the rotation of the gear. 3 is the angular velocity. Yaw rate sensor consisting of a sensor, 4
5 is a straight running detection means, 5 is a zero point calculation means, and 6 is an azimuth calculation means.The operation of the azimuth detection device according to the first embodiment of the present invention constructed as described above will be explained below. 1st
The purpose of this embodiment is to automatically correct the zero point of the yaw rate sensor and calculate the moving direction of a moving object with high accuracy even while the vehicle is moving by using only a relative direction sensor without using a geomagnetic sensor. Note that this embodiment can also be configured using hardware (here, we will describe a case in which the wheel speed sensor and angular velocity sensor are realized using a microcomputer or software).

FIG. 2 is a flowchart showing the processing procedure. First, in step 201, initial values are set for each constant and variable. Constants used here (sampling time Ts of sensor data, yaw rate sensitivity As, time threshold Tr for determining whether to perform zero point correction of the angular velocity sensor)
It is min. Ts and Trmin are set at appropriate time intervals (for example, Ts = 0.1 sec, Trmin = 1
0sec), but As is the sensitivity of the initially calibrated angular velocity sensor (for example, As = 50mV/deg/
5ee). On the other hand, the variables for convenience are: the number of rotation angle pulses PL of the left rear wheel detected at a predetermined time interval, the number Pr of rotation angle pulses of the right rear wheel, and the difference numerator 11 of the number of rotation angle pulses, nZ.

Evaluation value E used for straight-ahead determination, yaw rate data Rg that is the output of the yaw rate sensor 3, and this yaw rate data R
g integrated value SRg, zero point (offset) Rgz of yaw rate sensor 3, angular velocity data integration time Tt.

This is the traveling direction Ah of the vehicle. Among these, Pr, PL
, nl, nZ, E, RgP, SRg, Rgz
, Tt is assigned zero as an initial value. In addition, the initial direction of the vehicle is substituted for Ah. In the next step 202, the number of rotation angle pulses PL and Pr of the left and right rear wheels detected within a predetermined time is detected. Step 2
In step 03, yaw rate data Rg at the current sampling interval is collected. In step 204, the sampling time Ts is added to the angular velocity sensor integration time Tt.The yaw rate data Rg is added to the integration value SRg.In the next steps 205 to 208, straight-ahead running is detected. First, in step 205, the detected PL,
Difference n1 (=
In step 206, the evaluation value E is determined as n1Xn2 from the difference n2 in the number of pulses during the previous sampling and the current difference n1. Once the evaluation value E has been determined, nl is substituted into n2 in step 207 in preparation for the next sampling process. Then, in step 208, it is determined by the evaluation value E whether the vehicle is in a straight-ahead running state, but if it is in a left turn, the difference in the number of rotation angle pulses continues to be 1 or more.
Also, if turning to the right, the state in which the difference is 11 or less continues. If the evaluation value E is E≧1, the state is turning. Otherwise, it is determined that the state is running straight, but it is not determined that the state is running straight. If it is determined that the vehicle is in a turning state, the process proceeds to step 210, and the process proceeds to step 209, where zero is assigned to Tt and SRg, and then the process proceeds to step 210.

Are the angular velocity data Rg and the sampling time Ts integrated only in the straight running state? In step 210, the integrated value Tt of the sampling time Ts is compared with the time threshold Trmin for zero point correction of the angular velocity sensor, and the zero point is determined. Decide whether to perform correction or not. If Tt< Trm
If it is in, it is determined that the amount of data from the yaw rate sensor 3 is small and sufficient accuracy necessary for correction cannot be obtained, and the process moves to step 213 without performing correction.4 Tt≧Trmi
If n, the process moves to step 211 and zero point correction is performed.

The correction in step 211 is performed as follows.
Ku.

First, yaw rate data Rg added by Tt/Ts
Since the number of angular velocity sensors is calculated, the zero point Rgz of the angular velocity sensor is calculated as Rgz=SRg/(Tt/TS) (1). In step 211, the zero point of the new angular velocity sensor is calculated (f,, and in step 212, SRg, T
Zero is assigned to t and the process moves to step 213.

In step 213, the traveling direction A of the vehicle previously determined is
By adding the turning angle of the vehicle obtained from the yaw rate data Rg of the vehicle detected by the yaw rate sensor 3 to h, the new heading of the vehicle is calculated as follows: Ah 4-Ah+ (Rg-Rgz) x Ts/ As・
...(2) If the traveling direction Ah is calculated, step 202
The process from step 202 to step 213 described above is executed within a predetermined time interval Ts using a timer inside the microcomputer. shall be.

As described above, according to this embodiment, focusing on the difference in the number of pulses between the left and right rear wheels measured at a predetermined time interval, the straight running state of the moving object is determined based on whether the difference continues to be equal to or greater than a certain value. If the yaw rate sensor data is added in a section determined to be a straight-ahead running state and the addition time exceeds a certain time, the zero point of the yaw rate sensor is corrected from the addition value and addition time of the yaw rate sensor data. Tortoise: Without using a geomagnetic azimuth sensor that is susceptible to external magnetic fields, it is possible to accurately calculate the vehicle's heading by following the zero point fluctuations of the yaw rate sensor using a combination of a yaw rate sensor, which is a relative azimuth sensor, and a wheel speed sensor. It becomes possible.

In the first embodiment, the straight running state is determined based on whether the pulse number difference n is n≧1 or n≦-1 (this is the sensitivity (resolution) of the wheel speed sensor).
The value of the number of pulses used as the determination standard may be changed depending on the difference in the number of pulses.Also, it may be determined that the vehicle is not in a straight running state until a certain section before and after the difference in the number of pulses continues in a constant direction of positive and negative. The force that uses the newly calculated zero point of the yaw rate sensor 3 in the straight-ahead section to calculate the subsequent direction of travel of the vehicle (a fixed percentage or a fixed value of the value calculated by taking into account the calculation error of the zero point) In this example, it is possible to correct the zero point by the force used by the wheel speed sensor to start traveling by Mmm. A second embodiment of the present invention will now be described.

FIG. 3 shows a block diagram of a direction detecting device according to a second embodiment of the present invention. In Figure 3, 7 is G
Satellite-based positioning arrangement for calculating the current position of a moving body using radio waves from PS satellites 8 is a yaw rate sensor, 9 is a straight running detection means, 10 is a zero point calculation means, and 11 is an azimuth calculation means.

The operation of the direction detecting device according to the second embodiment of the present invention configured as described above will be explained below.
The purpose of this embodiment is to determine the straight-line running state of a moving object from the history of the current position of the moving object such as a vehicle calculated by receiving radio waves from a satellite, and to calculate the magnetic field while calibrating the zero point of the yaw rate sensor in the straight-going section. The goal is to calculate the heading with high accuracy without being affected by external disturbances. Note that this example is a sumo wrestler that can also be configured with hardware.
A case will be described in which this is realized by software using an S reception angular velocity sensor and a microcomputer.

FIG. 4 is a flowchart showing the processing procedure. First, in step 401, initial values are set for each constant and variable. Constants used here: Sampling time Ts of yaw rate data Rg, Yaw rate sensitivity A\ Time threshold Trmin for determining whether to perform zero point correction of the yaw rate sensor 8, Distance threshold LmirL used for straight running determination, Lma person Angle threshold BB, CC
It is. Ts, Trmin are set at appropriate time intervals (for example, Ts = O, 1 sec, Trmin = 10 sec
c), and As is the sensitivity of the initially calibrated angular velocity sensor (
For example, As=50mV/deg/5ee) is substituted.

f = Lmir5 Lmax, appropriate values for BB, CC (for example, Lmin = 1m, Lmax =
500[n.

Assign BB=2deg, CC=lodeg). On the other hand, the variables to be used for convenience are the current position coordinates of the moving object (X(n), ¥(n)) sequentially calculated by satellite-based positioning means.
, the moving distance of the current position determined by the satellite-based positioning means, and the moving direction A of the current position determined by the satellite-based positioning means.
(n), amount of change in A(n) B, A(n) and A(1)
The difference C1 is the yaw rate data Rg of the output of the yaw rate sensor, the integrated value SRg of this yaw rate data Rg,
Zero point (offset) Rgz of yaw rate sensor 8,
Yaw rate data integration time T - Vehicle traveling direction Ah. As the coordinate system of X(n) and Y(n), a coordinate system that does not normally become 0 is selected. Among these, R
g, SRg, Rgz, Tt, X(n), Y(n),
Zero is assigned to B, C, A(n), and L as initial values. Further, the initial direction of the vehicle is substituted for Ah.

In the next step 402, yaw rate data Rg at the current sampling interval Ts is collected. In step 403, the sampling time T is added to the toe rate data integration time Tt.
s is added and the yaw rate data Rg is added to the integrated value SRg.
Add.

In the next step 404 to step 414, straight-ahead traveling is detected. First, in step 404, it is determined whether or not the current position can be measured using GPS satellites.4 If the number of GPS satellites that can be used for positioning is 2 or less, positioning is not possible f
@, positioning is possible with 3 or more. If positioning is impossible, in step 405, Tt, SRg, X(n
), Y(n) and proceed to step 418. If positioning is possible, in step 406 the current position (X(
n), Y(n)) and memorize the coordinates of the fourth order step 407. It is determined whether Teha (X(0), Y(0)) is set. If X(0)=Y(0)20, it is assumed that the current position was calculated only after positioning became possible, and the process moves to step 418. Otherwise, it is assumed that positioning was possible during the previous sampling, and step 408 Move to. In step 408, the distance traveled is calculated from the previously calculated current position (χ(T1~1), Y(n-1)) and the currently calculated current position (X(n), Y(n)) using the following formula. calculate.

L= [(X(n)-X(n-1)) 2+(Y(n)
-Y(n-1)) ' ] ” ...(3) If the calculated travel distance is sufficiently small, the vehicle is considered not to be moving. If it is much larger, the current position is calculated using GPS satellites during that time. Assume that the accuracy has changed, i.e.
In steps 409 and 410, L and Lmin, Lmax
, and only when Ln+in<L<Lmax, proceed to step 411 and calculate the moving direction A(n) L
When Lmin≧L, the amount of movement is determined to be sufficiently small, and the process proceeds to step 418. When L≧Lmax, it is assumed that the current position calculation accuracy has changed, and the process proceeds to step 417.

In step 411, the moving direction A(n) is calculated using the following equation.

A(n)=tan-' ((Y(n)-Y(n-
1))/(X(n)-X(n-1))) ・ ・ (
4) In step 412, the amount of change in the traveling direction, which is the standard for determining whether to proceed straight, is calculated using the following formula.

B=A(n)-A(n-1)...(5)C=
A(n)-8(1)...(6) B is the amount of change in the moving direction when the previous moving direction is taken as a reference, and C is the initial moving direction when straight-ahead judgment started. represents the amount of change in the direction of movement when .

In steps 413 and 414, these azimuth changes B1C
and the threshold value B CC, ll]<BB,
Only when the conditions IC<CC are satisfied at the same time, it is assumed that the vehicle is traveling straight and the process proceeds to step 415, and when either condition is not satisfied, it is assumed that the vehicle is not traveling straight and the process is transferred to step 417.

Figure 5 shows an example of the relationship between the current position calculated using a GPS satellite and the traveling direction A(n) at that time. If ko(7), from (X(0), Y(0)) to (X(5)
, Y(5)) are determined to be in a straight-ahead running state (X
It is determined that the vehicle is no longer traveling straight between (5), Y(5)) and (X(6), Y(6)).

Are the yaw rate data Rg and the sampling time Ts integrated only when driving straight? In step 415, the integrated value Tt of the sampling time Ts is compared with the minimum time Trmin for zero point correction of the yaw rate sensor 8, and zero point correction is performed. However, if Tt<
If it is Trmin, it is determined that the amount of data from the yaw rate sensor 8 is small and sufficient accuracy required for correction cannot be obtained, so the process moves to step 418 without performing correction, and Tt≧1'
If r min , the process moves to step 416 and zero point correction is performed. Since the number of yaw rate data Rg being added is found by Tt/Ts, zero point Rgz L of yaw rate sensor 8 Rgz=SRg/(Tt/Ts) (7)
4 In step 416, a new zero point of the yaw rate sensor is calculated.
gSTt to turtle X(0) to X(n), Y(0) to Y(n)
is substituted and the process moves to step 418. Step 418
Then, the turning angle of the vehicle obtained from the yaw rate sensor 8 is added to the previously determined traveling direction Ah of the vehicle, and a new traveling direction of the vehicle is calculated as follows.

Ah4-Ah10 (Rg-Rgz)xTs/As ・
・ ・ ・(8) If the traveling direction Ah is calculated, the process moves to step 402, and steps 402 to 418
Repeat the above steps.

It is assumed that the processing from step 402 to step 418 described above is performed within a predetermined time interval Ts using a timer inside the microcomputer.

As described above, according to this embodiment, we focus on the history of the vehicle's position calculated using radio waves from the 1iGPs satellite, and if the movement direction calculated from there is within a certain range, it is determined that the vehicle is traveling straight. Then, yaw rate data is added in the section determined to be a straight-ahead running state, and when the addition time exceeds a certain time, the zero point of the yaw rate sensor is corrected from the addition value and addition time of the yaw rate sensor data.

8. Calculate the accurate heading of the vehicle while following the zero point fluctuation of the yaw rate sensor using a combination of a yaw rate sensor, which is a relative azimuth sensor, and a GPS receiver, without using a geomagnetic azimuth sensor that is easily affected by external magnetic fields. becomes possible.

In the second embodiment, the traveling direction is calculated from the relationship between the current position of the vehicle calculated using radio waves from a GPS satellite, the immediately previous position, and the current position and the initial position when the vehicle starts to go straight. The sumo wrestler who determined whether the vehicle was traveling straight by In the example, the force used to calculate the subsequent direction of travel of the vehicle by directly using the zero point of the yaw rate sensor newly calculated in the straight-ahead driving section (considering the calculation error of the zero point and applying a constant percentage or constant value to the calculated value) The zero point may be corrected by the value (--Next, a third embodiment of the present invention will be described. FIG. 6 shows a block diagram of a direction detecting device in a third embodiment of the present invention. In Fig. 6, 12 is a left wheel speed sensor, and 13 is a right wheel speed sensor. 14 is a yaw rate sensor consisting of an angular velocity sensor, and 15 is a wheel speed sensor that determines the turning angle of the vehicle from the output. The first turning angle estimating means 16 calculates the turning angle of the vehicle from the output of the yaw rate sensor.
Reference numeral 18 denotes a straight running detection means, and numeral 19 denotes an azimuth calculating means.The operation of the azimuth detecting device according to the third embodiment of the present invention constructed as described above will be described below. The purpose of this third embodiment is to calculate the moving direction of a moving object with high precision using only a relative direction sensor without using a geomagnetic direction sensor. Here, we will discuss the case where it is realized by software using wheel speed sensors, angular velocity sensors, and a microcomputer.

FIG. 7 is a flowchart showing the processing procedure. First, in step 701, initial values are set for each constant and variable. The constants used here are: sensor data sampling time T & yaw rate sensitivity As; wheel constants for the left and right rear wheels: 1, lcr, and tread TR. T.S.

Set TR to an appropriate value (for example, Ts=O, 1sec,
TR = 1.5m), As is the sensitivity of the initially calibrated angular velocity sensor (for example, As = 50mV/deg/5ee
). In addition, 1 is the wheel constant, and Kr is the distance traveled by each rotation angle pulse of the left and right rear wheels.The initial calibration is based on the ratio of the travel distance measured on a long-distance straight course with a fixed distance and the number of rotation angle pulses. and substitute the value found there (for example, 2C[D/pulse)]. On the other hand, the variables to be used for convenience are the number of rotation angle pulses P1 of the left rear wheel detected at predetermined time intervals, the number of rotation angle pulses Pr of the right rear wheel, and the difference n1, n2 between the rotation angle pulse numbers. Evaluation value E1 used for straight-ahead determination
Output R of the yaw rate sensor 14 & Zero point offset of the yaw rate sensor 14) Rgz, Turning angle D of the vehicle
L is the traveling direction Ah of the vehicle. Among these, PL, Pr
, nl, n2. E, Rg, RgzXD
Zero is assigned to h as an initial value. Further, the initial direction of the vehicle is substituted for Ah. In the next step 702, the number of rotation angle pulses P1, Pr of the left and right rear wheels detected within a predetermined time is
In the detection step 703, yaw rate data Rg at the current sampling interval is collected.

In steps 704 to 707, straight-ahead travel is detected. First, in step 704, the detected Pi, Pr
Difference n1 (=Pr-P) in the number of pulses at a predetermined time interval from
L) is calculated, and in step 705, the evaluation value E is calculated as nl x n2 from the difference η2 in the number of pulses during the previous sampling and the current difference numerator 11. When the evaluation value is calculated, nl is substituted into n2 in step 706. Prepare for processing during sampling. Then, in step 707, it is determined whether the vehicle is in a straight running state using the evaluation value E. If the vehicle is turning left, the difference in the number of rotation angle pulses continues to be 1 or more, and if it is turning right, the difference is 11 or less. Since the state of is continuous,
When the evaluation value E is E≧1, the turning condition is determined. Otherwise, it is determined that the running condition is straight.a If it is determined that the traveling condition is straight, the process moves to step 708. If it is determined that the turning condition is determined, the process moves to step 709. The turning angle of the vehicle is then calculated.

When driving in a straight line, the turning angle calculated from the wheel speed sensor, which is less affected by the zero point (offset) of the sensor, is more accurate, and when the vehicle is turning, the turning angle calculated from the angular velocity sensor, which is less affected by wheel deformation, etc. Since the turning angle has higher accuracy, in step 708 when it is determined that the vehicle is in a straight running state, the output of the angular velocity sensor is calculated as Dh= (Rg-Rgz)xTs/As...
The turning angle of the vehicle is calculated as (9), and if it is determined that the vehicle is turning, the output of the wheel speed sensor is calculated in step 709.
(180/π) x (KrxPr-K]xP1)/TR...(10
) to calculate the turning angle of the vehicle. In the case of Type 10, a left turn is considered a positive direction. Then, in step 710, a new vehicle traveling direction Ah is determined from the previously determined traveling direction Ah and the turning angle obtained using equations 9 or 10.
, Ah←Ah+ Dh ・・・・・・・・・(11
). Once the new heading Ah has been calculated, the process moves to step 702, and the processes from step 702 to step 710 are repeated.

The processing from step 702 to step 710 described above is performed within a predetermined time interval Ts using a timer inside the microcomputer.As described above, according to this embodiment, measurements are performed at predetermined time intervals. Focusing on the difference in the number of pulses between the left and right rear wheels, the straight-ahead driving state of the moving object is determined based on whether the difference continues to be above a certain value. The turning angle is calculated from the output, and in the section determined to be in a turning state, the turning angle is calculated from the output of the yaw rate sensor to calculate the moving direction of the moving object. The number
Without using a geomagnetic azimuth sensor that is easily affected by external magnetic fields, it is possible to accurately calculate the heading of the vehicle using a combination of a yaw rate sensor and a wheel speed sensor, which are relative azimuth sensors.

In the third embodiment, the straight running state is determined based on whether the difference n in the number of pulses is continuously n≧1 or n≦-1.This is the sensitivity (resolution) of the wheel speed sensor. The value of the number of pulses used as the determination standard may be changed depending on the situation.Also, it may be determined that the vehicle is not in a straight running state until a certain section before and after the difference in the number of pulses continues in a constant direction of positive and negative. Nomo α For example, it is possible to calculate the position of the vehicle using radio waves from a GPS satellite and determine whether it is traveling straight from the position history.Furthermore, in this embodiment, when the vehicle is traveling straight, the output of the wheel speed sensor is In addition, in a turning state, the turning angle of the vehicle is calculated from the output of the yaw rate sensor (in a straight-ahead state, the specific gravity of the wheel speed sensor is increased (de), and in a turning state, the specific gravity of the yaw rate sensor is increased to calculate the turning angle). For example, the output of the yaw rate sensor may be multiplied by a certain coefficient k, and the output of one wheel speed sensor may be multiplied by (1-k). Add each of them and divide by k. Take the so-called arithmetic mean (°C. Or you can use the geometric mean (t?) This weighting coefficient may be a constant set in advance, or the weighting coefficient may be a constant set in advance. It may be a function of the speed of the turning curve, or it may be a function of the radius of curvature of the turning curve, or the time it takes for the vehicle to turn (℃).The zero point of the yaw rate sensor may be corrected when the vehicle is traveling straight, or it may be a function of the curvature radius of the turning curve or the time it takes for the vehicle to turn. It is also possible to calibrate the wheel constants under running conditions. The left and right wheels can be either driving wheels or driven wheels (front wheels or rear wheels), and the turning angle calculation model uses Ackermann's geometric relationship, which places the turning center on the axis of the rear wheels, which holds true at extremely low speeds. As described in detail, according to the present invention, it is possible to accurately calculate the traveling direction of the vehicle by combining the yaw rate sensor and the wheel speed sensor, which are direction sensors that are less affected by disturbances. The practical effect is great.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a direction detection device according to a first embodiment of the present invention. FIG. 2 is a flow chart of the same embodiment, and FIG. 3 is a block diagram of a direction detection device of a second embodiment of the present invention.
FIG. 4 is a flowchart in the same embodiment, FIG. 5 shows the relationship between the history of the current position and the direction of travel in the same embodiment, and FIG. 6 is a block diagram of the direction detection device of the third embodiment. In the flowchart in the same example,
・Left wheel speed sensor, 2... Right wheel speed sensor, 3.
... Yaw rate sensor, 4... Straight running detection means
5... Zero point calculation half sunk 6... Direction calculation means, 7.
... Satellite-based positioning means, 8... Yaw rate sensor,
9... Straight running detection means 10... Zero point calculation hand! ,
11... Orientation detection device 12... Left wheel speed sensor, 1
3... Right wheel speed sensor, 14... Yaw rate sensor, 15... First turning angle calculating means, 16... Second turning angle calculating means, 17... Turning angle calculating means 18.・・・
Straight running detection means, 19...Name of direction calculation hand dragon agent Patent attorney Shigetaka Awano

Claims (5)

[Claims] (1) A wheel speed sensor that measures the number of rotations of left and right wheels independently, a straight-line running detection means that detects a straight-line running state of a moving object from the moving distance of the left and right wheels determined by the wheel speed sensor, and a moving object. a yaw rate sensor that detects the yaw rate of the moving object; a zero point calculation means that calibrates the zero point of the yaw rate sensor when the moving object is determined to be in a straight running state by the straight running detection means; An azimuth detection device comprising azimuth calculation means for calculating the traveling azimuth of a moving object from the zero point of a yaw rate sensor determined by the yaw rate sensor. (2) In place of the wheel speed sensor, a satellite-based positioning means is provided that measures the current position of a moving object by receiving radio waves from a satellite, and the straight-ahead running detection means detects the moving object determined by the satellite-based positioning means. 2. The direction detection device according to claim 1, wherein the straight running state of the moving object is detected from the history of the current position of the moving object. (3) A wheel speed sensor that measures the rotation speed of the left and right wheels independently, a yaw rate sensor that detects the yaw rate of the moving object, and a turning angle of the moving object from the travel distance of the left and right wheels determined by the wheel speed sensor. a first turning angle calculation means for calculating a turning angle of the moving object from the yaw rate of the moving object determined by the yaw rate sensor; Weighting the outputs of the travel detection means, the first turning angle calculation means, and the second turning angle calculation means to determine the turning angle of the moving object;
and when it is determined by the straight-ahead running detection means that the moving object is in a straight-ahead running state, the weight of the output of the first turning angle calculating means is increased by the output of the second turning angle calculating means, and the mobile body is brought into a straight-going running state. a turning angle estimating means for estimating the turning angle of the moving body by increasing the weight of the output of the second turning angle calculating means than the output of the first turning angle calculating means when it is determined that the turning angle is not present, and the turning angle estimating means. A direction detection device comprising a direction calculation means for calculating a traveling direction of a moving object from a turning angle of the moving object calculated by the means. (4) The azimuth detecting device according to claim 3, wherein the straight-ahead running detecting means detects the straight-ahead running state of the moving body from the output of a wheel speed sensor. (5) A satellite-based positioning means that receives radio waves from a satellite and measures the current position of the moving body, and the straight-ahead running detection means:
4. The azimuth detecting device according to claim 3, wherein the straight running state of the moving object is detected from the history of the current position of the moving object determined by the satellite-based positioning means.