JP2000131089A - Navigation system and method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make continuable system operation even when the function of a main navigation system or a location system is lowered or lost by recording the sensor data when a transportation means is operating normally and correcting it automatically. SOLUTION: The system comprises an electric circuit including an anti- magnetic sensor, a microprocessor including a memory, and an I/O unit receiving a data from a speed sensor or an accelerometer. The memory includes a software algorithm which measures the heading direction of a transporting means accurately, calibrates a magnetic sensor automatically, and integrating various sensor data to calculate a flywheel position data. Because of the permanent magnetism of the transporting means, readings of two orthogonal magnetic sensors describes a circular pattern which is distorted elliptically by magnetism induced in an easily magnetizable material, e.g. soft iron. When the transporting means rotates in complete circle, eccentric ellipse is corrected by means of two horizontal axis flux gates or the anti-magnetic sensor.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001] The cross-reference application of the prior application was filed in January 1997
Claims of U.S. Provisional Application No. 60 / 037,025, filed on March 31, the contents of which are incorporated herein by reference.

[0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic sensor and a non-magnetic sensor for determining the traveling direction of a vehicle and measuring the movement, acceleration / deceleration and speed of the vehicle. In particular,
The invention can be used to indicate the direction of travel to the driver, or the invention can be used as an aid to other navigation equipment, such as, but not limited to, a global positioning system (GPS). be able to. Further, the invention relates to a multi-sensor integrated vehicle navigation / location system.

REFERENCE TO APPENDIX The following appendix is attached as part of and / or incorporated by reference herein.

Appendix A: The following list is an example of the software implementation described above.

[0005]

BACKGROUND OF THE INVENTION In all types of vehicles, there is an increasing number of electronic devices that can supply data used by the driver or data about remote locations remote from the vehicle, for example, to determine the direction of travel of the vehicle. is needed. The present application has a stand-alone compass, in which the heading of the vehicle is
For example, eight compass directions, north (N), south (S),
It can be displayed on a miniature electronic display as one of east (E), west (W), northeast (NE), northwest (NW), southeast (SE) and southwest (SW). A more sophisticated version of such a stand-alone compass is to increase the heading of the compass by one or two degrees,
Alternatively, some compass, such as a compass on a compass diagram, have sufficient performance and accuracy to display in an analog manner.

[0006]

A vehicle-mounted compass must be calibrated in consideration of the magnetic characteristics of the vehicle. If the magnetic properties of the vehicle change, a new calibration is needed. However, the driver is often unaware of the change, which leads to incorrect readings.

[0007] Also, when a transient magnetic disturbance occurs, the conventional system retains the heading just before the disturbance started.
At the end of the obstruction, a new and accurate heading will be indicated. Bridges,
Steel structures or adjacent roads built of iron can cause transient magnetic anomalies.

One problem lies in the detection of such abnormalities. This is because such an anomaly may not be detected as an anomaly, in which case the conventional compass will respond with an incorrect reading. In addition, the abnormalities or the abnormalities that occur repeatedly may be present for a long time on modern arterial roads such as elevated intersections, ramps, and other roads constructed of iron materials.

[0009] In another example, various navigation systems have been developed to assist a vehicle driver or passenger in navigating to a particular location or navigating away from a particular location. In general, navigation systems employ some form of triangulation from three or more wireless signal sources to determine the location or location of the vehicle. Such a system is called LORA
It has been created using N, DECCA, GPS, FM and AM radio signals and other RF radio system technologies. Such a navigation system is created by plotting changes in the location of the vehicle on an electronic map over time. Such a navigation system operates with high reliability as long as it can receive radio signals without interference. Such systems have serious limitations because the loss of one or more radio signals due to signal interference can severely reduce the accuracy of navigation systems and often cause them to fail.

[0010] Navigation and location determination systems, such as those employing Global Positioning System (GPS) technology, including differential GPS (DGPS) technology, are often received and used by the system to determine the location of a vehicle. There is a limitation that occurs because the signal being used cannot be used temporarily. In the GPS and DGPS examples, there may not be enough satellite signals available to keep the position of the vehicle continuously measured, for example, in the area around tall buildings and other obstacles.

Another example, transportation distinct from vehicle navigation, in which the location of a particular vehicle can be determined within a vehicle or at a remote location in connection with a map or other vehicle location determination scheme. Vehicle locations have very similar requirements and similar limitations. Such an application provides vehicle location data to a vehicle or other ground, sea, air, or space vehicle occupant, thereby providing vehicle location determination for a vehicle or other ground, sea, air, or space vehicle. Ranging to the system, the latter system generates coordinates in various coordinate systems that allow the vehicle to be located on some form of electronic map display, or on a non-electronic map or on some main grid. By doing so, the location of the vehicle can be pinpointed.

In such a case, there is a need for a means that can continue to calculate the location of the vehicle during these times. That is, it assists the navigation or location system by outputting a continuous series of vehicle locations and maintains the ability of the system to continue functioning in the event of a signal stop or signal interruption. This can be done using the integration of sensor data from various sources and processing those data to calculate vehicle change locations over time. Such techniques include data input such as speed and travel from a speed sensor or accelerometer, along with the heading of the vehicle measured from the last known vehicle position before the signal stop or signal jamming occurs. Process. Such a system is sometimes referred to as a flywheel, and in the case of a GPS system, the system continues to operate even when the main navigation system or location system is degraded or lost. Sometimes called.

[0013]

SUMMARY OF THE INVENTION In one aspect, the present invention is a specific application compass system and method for a vehicle that performs automatic calibration. In particular, by recording sensor data when the vehicle is operated in normal use, until the vehicle is operated in a perfect circle,
Perform automatic calibration. This is detected and recorded by the system by dividing the system-defined circle into sections or bins and ensuring that at least one data point exists for each bin. In fact, the means of ensuring that data points are placed in each bin when the vehicle is driven and visited each bin is by adjusting the calculated offset parameter that describes the center of the ellipse. An important feature of this and other aspects of the invention is that it utilizes high speed readings from magnetic sensors.

Another aspect of the present invention is to detect the presence of a transient magnetic anomaly that moves relative to the vehicle, and then reject abnormal sensor readings in processing and provide heading indication. Make those readings unusable as output.
Transient magnetic bodies, such as bridges, steel, reinforced bridges, elevated crossings, etc., have been found to have high frequency properties called jitter, which can be detected and discarded because they can be detected and identified. As a result, the heading direction before the occurrence of jitter can be maintained as the output of the compass system in which no change occurs as long as the transient abnormality exists. Then, when the jitter is over, the sensor data processing starts giving the current heading again. As a result, for high-speed data reading, good reading can be performed even with a short interruption of jitter, so that a good heading direction can be output. Thus, for example, a good reading is obtained while passing through a relatively long anomaly source, providing an updated heading. Only a short period of good data when transient anomalies are present is sufficient to update and provide good read output.

Another aspect of the present invention is to combine a compass system with a radio navigation system such as GPS,
When the GPS signal is lost, the compass system operates to provide heading data. A fully integrated system is also conceivable which integrates a precision wireless navigation system such as GPS with a combination comprising alternative sensors, for example a magnetic compass, a cross-track accelerometer, a wheel rotation sensor and a speed sensor. An angular velocity sensor may be used instead of, or in addition to, a cross-track accelerometer. This integrated system can provide benefits even when all combinations of GPS satellites are present. The integration allows alternative combinations to be used as needed to supplement or replace the initial navigation system. This can be implemented using additional sensors such as speed sensors, rate sensors and cross-track accelerometers. At least four GPS
Good satellite navigation can be obtained using satellites or reception from three GPS satellites if the altitude is known. The present invention takes into account data loss when the number of GPS satellites is reduced to three first, then to two, then to one, and finally lost.

An important aspect of this aspect of the novel invention is that it adjusts the Kalman filter so that the associated speed and heading sensors can be used optimally. Specifically, the states included in the Kalman filter are 2
In one of these states, the coordinate system is not essential. The third position state is the vertical position. It also includes the clock time state and clock frequency state of the GPS receiver. The speed conditions include vertical speed, heading of the vehicle, and track speed of the vehicle. Depending on the particular choice of these conditions, GP
Whenever an S-satellite is lost, it allows the Kalman filter to be easily reconfigured to the extent that system accuracy only slowly degrades. Therefore, when the number of GPS satellites is reduced to three, the vertical speed state is slowly changed to 0, that is, the mode is changed to the height holding mode. As the number of satellites is reduced to two, absolute readings from GPS code measurements are eliminated and only changes in carrier-phase measurements are used. Furthermore, without using carrier-phase measurements,
Provides heading status. The heading state is driven by a magnetic heading sensor and a cross-track accelerometer. As the number of GPS satellites decreases to one, the GPS measurements are used only to update the clock frequency state of the GPS receiver. Whatever the speed sensor can be used to drive the track direction speed state.
The heading state is driven using a magnetic compass and a cross-track accelerometer. The standard time update process obtains the position status.

A speed sensor can be used, where two or more calibration states are provided, which provide speed scaling and match the speed measurement with the GPS derived speed.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In a preferred embodiment of the present invention, in one aspect, the present invention provides an electrical circuit including a magnetically resistant sensor, a microprocessor having a memory, and other sensors, such as a speed sensor or And an input / output device capable of receiving data from the accelerometer.
The memory includes software algorithms, which execute algorithms to calculate the correct heading of the vehicle, automatically calibrate the magnetic sensors, and calculate flywheel position data from the integration of various sensor data. Here, the term flywheel refers to a system that operates with other sensors when the main navigation system as a whole or a part thereof does not work, for example, when GPS reception is interrupted.

Magnetic heading sensors typically have problems associated with induced and permanent magnetism in vehicles that include the sensor. The magnetic properties of a vehicle change frequently. Changes in magnetic properties are caused by changes in the shape of the metal structure of the vehicle, loading a large amount of iron material such as tool boxes and cargo on the vehicle, and interference from magnetic fields such as the operation of various types of electric motors and electric coils. there is a possibility.

Assuming no distortion from the vehicle carrying the sensor, the two orthogonal magnetic sensors draw a circle about the origin when the vehicle rotates in a perfect circle,
Here, the origin is defined as no magnetic field in any of the sensors. Due to the permanent magnetism of the vehicle, the readings of the two orthogonal magnetic sensors draw a circular pattern, which is off-center. Induced magnetism arising from soft iron or other magnetizable material causes the circle to distort into an ellipse having a major axis in the direction of the strong force of the magnetizable material. Several patents have been issued describing how to calibrate the eccentric ellipse drawn by either a two horizontal axis flux gate or a magnetically shunt sensor when the vehicle carrying the gate or sensor rotates in a full circle. U.S. Patent No. 4,611,29 to Hatch et al.
No. 3 and Hatch U.S. Pat. No. 4,797,841.
Signals are two examples where the compass can be calibrated in the presence of the permanent magnet of the vehicle and the induced magnetism of soft iron in all directions (ie, ellipse resolution is quite common).

[0021] Patents have been issued to obtain a similar effect by using a simplified assumption regarding the symmetry of induced magnetism or its existence. One patent, van Lente et al., No. 4,953,305, allows calibration to be performed assuming that soft iron is longitudinally aligned with the vehicle. This means assuming that the semilong axis of the ellipse coincides with the axis of the vehicle. Generally, this is not a bad assumption if the magnetic sensor is located near the centerline of the vehicle. Another patented technique can ignore the induced soft iron magnetism and force the pattern into circular matching, ie, assume that the ellipticity is zero. In general, this is not a very good assumption, but the accuracy required for car navigation displays only is generally good enough.

The parameters that must be determined to map the sensor readings to a circle centered at 0 are:
(1) center eccentricity in both the x and y axes; (2) ellipticity of the sensor reading when induced magnetism may be present; (3) half of the sensor ellipse when induced magnetism asymmetry is allowed. It is the angular direction of the long axis.

Patents that can automatically update the calibration method using certain simplification assumptions, for example, Wanous
US Patent No. 5,046,031 to U.S. Pat. For example, if it is assumed that a magnet in the same horizontal plane has obstructed the sensor reading, the size of the circle mapping the ellipse does not change, and the center eccentricity can be moved to restore the same sensor reading. However, this process introduces errors in certain situations. First, if the disturbing magnet is not in the same horizontal plane as the sensor, the center position of the circle and the size of the circle may change. Second, if the location of the vehicle shifts sufficiently at magnetic latitudes, the horizontal component of the Earth's magnetic field will increase or decrease significantly, and the size of the circle will change accordingly. For these reasons, a new automatic calibration process has been invented that is capable of automatically alerting a user to the need to perform a new "assisted" calibration when an automatic calibration process indicates a need.

A new precision compass with several advanced features and an improved automatic calibration process is disclosed that is ideal for applications requiring high precision heading information.
This compass can be used as a stand-alone compass or integrated with other on-board sensors. G
In more fully developed navigation and location systems with wireless navigation systems such as the PS, compasses and other on-board sensors form a highly integrated system.

This device forms a flywheel, due in part to obtaining many sensor readings per second. In current implementation, 5Hz-6Hz, 12Hz max
(Reads / sec) updates have been used. Using this device with a speed sensor input would otherwise
It is a great aid to PS-only navigation solutions. This device is accurate and self-adjusting and supplies data at a high update rate, thus forming an excellent low cost sensor.

At the heart of the new compass is a pair of orthogonally arranged magnetically resistant sensors (MR sensors). The output of these sensors is the output of Hatch (U.S. Pat.
7,841) is very similar to the output of a flux gate sensor which provided an algorithm to calibrate the sensor in the presence of local magnetism. The MR sensor uses M in the Earth's magnetic field.
It has sufficient sensitivity to measure less than 1 ° rotation of the R sensor. A high-precision direction sensor can be obtained by applying the Hatch algorithm and smoothing the raw inputs from the two orthogonal sensors by software.

Automatic Calibration Procedure To make this tool practically useful, it is necessary to automatically calibrate the sensor data. This is because a change in the local magnetic field may change the sensing direction. The method used for this automatic calibration will be described below.

Automatic calibration is performed in two different steps.

1. Wanoous US Patent No. 5,044
The offset parameter is adjusted periodically as in the method described in US Pat. No. 6,031.

2. When sufficient data has been collected, a full calibration algorithm is applied, as described below.

Periodic Calibration Adjustment In one step of the automatic calibration, the offset parameters are adjusted in a manner similar to that described in US Patent No. 5,046,031 to Wanous. Periodic adjustment of the offset parameter is based on a comparison between the radius calculated from the current data and the known radius obtained from the calibration term (the radius squared can be used instead of the radius to reduce calculation requirements). . However, if the technique described in the Wanous patent is simply applied, an error occurs in the offset adjustment. FIG. 1 helps explain the potential problem. Considering the circle denoted by C0, this is visible after re-circularization by applying the matching parameters,
Shows unperturbed raw data. The circle indicated by C1 is
This is the situation where "after alignment" data appears when the magnet is located near the sensor. Here, it is assumed that the transportation means is facing the SW. This gives an (x, y) pair at point r1. The radius comparison technique compares the radius of r1 with the radius of r2 and recognizes that the data has moved to the lower left (by obtaining the component of r2). This is completely wrong (generally, this error is wrong by 2 * radii shifted data in the wrong direction). In our exact calculations, the data was actually moving to the upper right because the vehicle was pointing to the NE.

The problem is that the data is ambiguous and r1 and r3
Both look like points indicating NE. The algorithm of the present invention may make this mistake,
It is only temporary, as more data is collected to find and correct mistakes. 25% of data
Each time we collect more data, we reconsider the data match by calculating the radius using the current matching parameters for the previously stored data pair and comparing that radius to the radius calculated using the latest data I do. If the difference between these two radii exceeds a predetermined selected threshold, discard the data and resume automatic calibration. It is desirable to examine the data and detect problems as soon as possible. 25% was chosen because it usually defines the general structure of the ellipse enough to be able to decide whether to keep or reject the data. Less data is enough,
More data will do the job but delay the rate at which the calibration parameters are updated. The minimum data is data necessary for defining the general structure of the ellipse. this is,
It solves not only the problem of mixing data from two different local magnetic inputs, but also the problem just described.

Since the problem of ambiguity does not occur until the offset is greater than the original radius (r2), the possibility of this problem occurring is also minimized. Since the measurements are taken at approximately 12 Hz and the offset is recalculated for each measurement, a perturbation with a radius greater than 1 must occur within approximately 1/12 second. In other cases, you will be delayed in moving,
Do not enter an ambiguous state.

The calculation of the automatic perfect calibration is based on the following method. Catalog the current data so that the software knows when the vehicle has rotated 360 °. Perform a new calibration whenever sufficient new data is available. This patent proposes a new data cataloging method that provides a better way of detecting when full rotation and automatic calibration are feasible under normal circumstances.
Whenever you turn on the car ignition,
Data is automatically collected and stored for a new calibration. The automatic adjustment of the offset parameter is monitored to avoid mixing data obtained from different local magnetic environments. The difference in radius of the circle from the current data exceeds a threshold (eg, の of the value obtained from the last full calibration).

The automatic calibration process is designed so that the operator usually does not need to perform any special process at all.

In addition to solving for the elliptic parameters, the rm of the match
Calculate s residual. If the rms residual remains below the selected threshold, the parameters of the calibration may be assumed to be accurate, and these parameters may be used to read the sensor readings relative to the central circle to calculate heading. Can be mapped. If the residual exceeds a selected threshold, a user message may be sent requesting that the vehicle be rotated around a full circle to update the sensor data. Once the rotation is complete, a minimum of 2
The power match is calculated again and the new calibration parameters are adopted if the rms residual is good.

[0037] Two additional techniques are applied to provide more accurate results. First, the moving average is unprocessed (x, y)
By applying to pairs, the raw sensor data is smoothed. In the second approach, as much data as possible is provided to the calibration algorithm. Calibration applies a least squares match to the provided data. Clearly, the more data provided, the better the match. One of the keys to the least-squares algorithm is to calculate (HtH) (reading H permutation H), which is the product of the transposition of the observation matrix H and the observation matrix H. In this application, H is a matrix of “n × 5” terms.
Here, “n” indicates the number of measurements. In the prior art, n * 5 terms had to be stored to provide data for matching. By calculating "Hurry" HtH and Hty, all the available data may be used, since there is no need to store the individual data points, so that a good match can be made.

Introduction of Speed Sensor Algorithm A new Kalman filter method is proposed to introduce speed sensor data into the “GPS flywheel”. Various speed sensors can be included in a vehicle navigation system. Sensors that can assist navigation while the GPS signal is missing for a short period of time are particularly beneficial. Such sensors include, but are not limited to, an integrator speed sensor or other drive shaft sensor, wheel rotational speed sensor, Doppler radar speed sensor, or an accelerometer that can be integrated to indicate vehicle speed. Never.

An important aspect of this aspect of the novel invention is that it tunes the Kalman filter so that the associated speed and heading sensors can be used optimally. In particular, the states included in the Kalman filter include two horizontal position states, in which the coordinate system is not critical. The third position state is the vertical position. It also includes the clock time state and clock frequency state of the GPS receiver. The speed states include vertical speed, heading of the vehicle, and track speed of the vehicle. With a particular choice of these conditions, the Kalman filter can be easily reconfigured whenever a GPS satellite is lost to the extent that system accuracy simply degrades slowly. Therefore, when the number of GPS satellites is reduced to three, the vertical speed state is slowly changed to 0, that is, the mode is changed to the height holding mode. As the number of satellites is reduced to two, the absolute readings from GPS code measurements are reduced, and only changes in carrier-phase measurements are used. Further, the heading state is provided without using the carrier-phase measurement.
The entire heading state is driven by a magnetic heading sensor. When the number of GPS satellites is reduced to one, the GPS receiver only updates the clock frequency state of the GPS receiver.
Use measurements. Whatever the speed sensor can be used to drive the track direction speed state.
Using a magnetic compass and a cross-track accelerometer,
Drive heading state. The standard time update process acquires the position status.

In addition to the standard eight position and speed conditions described above, a calibration condition is added to appropriately calibrate the speed sensor and heading sensor. The magnetic compass is driven by GPS measurements and includes a state that determines the magnetic declination, ie, the calibration value of the magnetic compass. If an integrator or wheel sensor is used, it includes two or more calibration states that determine the speed scaling value and match the speed measurement with the GPS derived speed. Different scaling values may be required as a function of vehicle speed (ie, excessive wheel slip at high speeds).

The Kalman filter having the state designed as described above is particularly useful as a "GPS flywheel" system for determining an estimated position and obtaining a standard GPS position fix in the absence of sufficient GPS satellite measurements. It is believed to represent a unique configuration particularly adapted for use.

With reference to the above, altitude hold is appropriate when GPS reception is reduced to three satellites, for example, if an altimeter is not used instead. If GPS reception is reduced to two satellites, the magnetic compass and cross-track accelerometer will operate to provide heading. Further, an angular velocity sensor can be used in combination with these. Thus, as the accuracy of one sensor decreases, the inputs are gradually integrated and another sensor is used. If GPS reception is reduced to one satellite, only time is available, so other sensors are used for speed, decoding and plot location.

Additional Description Magnetically resistant sensors use a bridge circuit that is polarized to have directional sensitivity to the magnetic field to measure the strength of the magnetic field. When mounted on a vehicle, the sensor measures the earth's magnetic field and the deviating magnetic field caused by local conditions from the vehicle or an adjacent area. This gives rise to a compass direction from the anti-magnetic sensor.

The voltage output from the sensor corresponds to the angle of the sensor with respect to the magnetic field applied to the sensor. This gives the direction. To clarify the direction of rotation, two anti-magnetic compasses in the orthogonal direction are required. The system converts the output voltage to a digital value.

However, as long as the required angle can be specified in the system, only one angular position can be provided using a different type of sensor, for example a flux gate compass or a Hall effect sensor, and a magnetically resistant type Sensors are one of the desired types that can generate the desired direction signal. The use of anti-magnetic sensors is known in the prior art.

There is a new way to catalog data in a system. Each sensor reading produces an x, y coordinate that, when the vehicle moves in a circle, forms a circle or, more generally, an ellipse, which is the center of the coordinate system due to local magnetic anomalies. It is out of alignment. As before, the purpose of the system of the invention is to make adjustments to accommodate the magnetic properties of the vehicle and to accommodate any changes in the magnetic properties of the vehicle. The present invention also addresses the problem of transient magnetic anomalies. In particular, in one aspect of the invention, transient magnetic anomalies are detected and eliminated so that the system continues to provide accurate heading even when transient anomalies are present.

In the prior art, a circle or an ellipse is shifted with respect to the center of the coordinate system by applying a current to the coil to change the displacement of the circle, and the circle is arranged around the origin. In the present invention, since the ellipse is mathematically circularized, the reading of the sensor is not adjusted.

The radius of the circle is proportional to both the semi-major axis and the semi-minor axis of the ellipse. Therefore, the larger the ellipse, the larger the circle. Larger circles can use higher resolution. The largest size limiting factor is the dynamic range of the microcontroller ADC. If the radius is very small, heading cannot be determined due to noise. The compromise radius for obtaining high quality heading without approaching the limit of the dynamic range of the ADC is between 100 and 200 counts. As a start-up process, the radius is tested to determine if the radius is within the desired range, and the pulse width modulation is adjusted to obtain the desired range.

In summary, the calibration process described herein has the following general steps.

1. Collect data.

2. The calibration algorithm is executed to form the ellipse, and the calibration term, ie, the parameters of the ellipse, is obtained. This step has two steps. It (a) processes every data point used to update the observation matrix, and (b) finishes the calculation and obtains a least squares matching ellipse.

3. To obtain heading, use an acquired calibration term known to circularize heading data.

4. Calculate heading.

The mathematical process of transforming data points on an ellipse into data points on a circle is described in US Pat. No. 4,797, Hatch.
No. 841, the contents of which are incorporated herein, and are generally referred to as least-squares rounding, or more simply, rounding. In one aspect of the invention, H
The least squares round of the ach is a specific selected procedure, as can be seen from the software list,
This is done by providing the data continuously to avoid having to collect and store all the data. The heading direction can be measured with the circularization.

Data is collected and used for the calibration process. The data is used in a procedure called "binning" or "binning". This process requires driving the vehicle in a circle to collect data around a complete revolution, which is required for calibration. In known systems, compass calibration uses whatever system can be used,
This is done by a simple common procedure of driving the vehicle in a circular shape and collecting data. This is called manual calibration. The present invention describes a new automatic calibration procedure. The calibration completion process described here can be applied to both manual calibration and automatic calibration, but is particularly advantageous when applied to an automatic calibration system to enable it.

The key factors making the procedure feasible are:
Frequent acquisition of readings from sensors. One example of a data rate is 12 Hz, or 12 readings per second. The data rate limit can be very high, as it is limited by the processing power of the associated computer. However, very high data rates do not improve the results. The lower limit affects the precision. The method of the present invention uses a smoothing technique, which adjusts to accommodate noisy readings. It is necessary to collect data at such a high speed that noisy measurements are eliminated by smoothing. The lower limit is also driven by the need to record in a timely manner the movement of the vehicle as affected by the speed of the vehicle. A lower limit of approximately 5 Hz can be used to obtain good results for general purpose ground vehicles, such as cars and trucks on regular roads and highways. To this end, the result is influenced to some extent by the speed of the rotating vehicle, which is apparently related to the bin size.

Recall that in the binning process, any readings (except those excluded as invalid) are rounded so that they are placed on a circle, and thus in one of the segments, the result of the rounding process The resulting theoretical circle is divided into equal-sized segments that define equal parts of the circle.

Software Description In the preferred embodiment, the computer is programmed with the 15 sub-programs shown in Appendix A. The functions of these programs will be described with reference to the flowcharts of FIGS. The software is shown in FIGS.
And the electronic circuit of FIG.

FIG. 2 shows the definitions of the symbols that help interpret the flow diagram.

FIG. 3 is a flowchart of the startup routine 200. Optionally, a communication mode may be performed at 202. For example, to test, monitor or modify a program,
Communication mode is preferred. At 204, the subroutine LoadFr
Calls the omEEROM and supplies the calibration values stored in the EEROM. These are either previous sensor readings or default settings. At 206, the calibration terms are organized into structures and stored at 208. There are five calibration terms: x offset, y offset, flattening, rotation angle sine, and rotation angle cosine. Next, at 210, the smoother is initialized. A smoother is a program that calculates a moving average for x and y data readings.

In 212, the jitter adjustment value for detecting a transient abnormality is a jitter threshold value and a sequence length. The jitter aspect of the present invention relates to the detection and elimination of transient anomalies, as described in further detail below.
Jitter threshold and sequence length are selectable variables for processing. These can be set to default values or selected. At 214, ignore the first 20 (or other selected number) readings to ensure that there is no power-on hysteresis error. Next, the smoother array is filled at 216 to prepare for the calculation of the moving average. The heading is calculated using each moving average, unless it is ignored as an abnormal reading as detected by the jitter detection procedure. In the exemplary system, six readings are used for the moving average. The number of readings used for the moving average can be changed,
Select as a function of the rate at which the measurement is taking place. In the case of illustration, reading is performed at a rate of 12 times / second.
The reading rate depends on the expected speed of the vehicle. The higher the speed, the less samples are taken. At 218,
There is the step of providing sufficient resolution to determine the reading. Step 220 is to ensure that the readings match.

FIG. 4 is a flowchart of a main processing loop for performing calibration. As mentioned above, there are optional communication links at 230. The change setting step 232 allows the selection of magnetic declination for the geographic region of the current operation. At 234, the digital sensor reading is received from the subroutine and the reference voltage is removed. 23
At 6, the readings are loaded into the data structure and the jitter array is loaded. At 238, an automatic calibration is typically performed, unless the user selects a manual calibration.

FIG. 5 continues the main processing loop. To perform an automatic calibration, the vehicle must be operated through a full 360 ° rotation to obtain a reading that defines a complete circle (or ellipse) and provide at least one reading for each bin. . However, the driver does not need to be aware of this requirement, as the data is usually collected for each bin by driving, and automatic calibration is performed when each bin has at least one reading. First, a WildEdit subroutine is performed at 242 to eliminate pseudo data,
This is performed after 232A.

Next, at 244, the matrix and bin counters are cleared. Thereafter, at 246, the smoothed data is placed in a data structure used for heading calculations.
Also, a new term is put into the data structure for bin recording. In the illustrated example, thirty 12 bins are used. The size of the bin is related to the expected speed during rotation of the vehicle. The exemplary system obtains readings at a rate of 12 times / second. This is a good rate for ordinary vehicles on roads and highways. The reading rate may be determined by the circuit, but the rate must be fast enough to recognize that the vehicle is spinning, as it affects the quality of the resulting output to the user. Variables are data acquisition rate, vehicle speed, and number of bins. In this example, at least one data point must be provided for each bin in order to properly set the data for automatic calibration. For this reason, a minimum of 30 points are used, but this is very rare. As a result, if the vehicle rotates so fast that the readings skip the bin, then no automatic calibration is performed until the bin is filled by the vehicle in the heading of the omitted bin during operation. Although the 12 ° bin is a very small angle,
In general, twelve readings per second ensures that the vehicle obtains sufficient readings during rotation. For example, 9 per second
For a 0 ° rotation, twelve readings are performed on seven or eight bins. It is also appropriate to have a maximum reading for each bin, and in the illustrated system 2
5 was selected. Apply the circularization algorithm at 248,
Get the radius of the circle. This is an example system,
Provide the lowest rate bin content of 1:25. 1: 1
A ratio exceeding 00 is likely to distort the ellipse and was judged to be undesirable.

Step 250 determines whether the data does not match the selected amount of 30%.
If not, at 252 the offset is shifted and a new heading is calculated. The general concept of offset is discussed in Wanous Patent 5,046,031. If you can accept Delta,
Processing proceeds from 254 to FIG. If 250 deltas cannot be accepted, steps 254a-254f
Then check to see if the result is better or worse. The restart operation concludes that a major anomaly exists in the vehicle, for example, a large new magnetic event that requires re-reading of the vehicle's magnetic properties. Otherwise, 2
If you conclude “No” at 54e, automatic calibration (256)
followed by.

Referring to FIG. 6, the automatic calibration procedure proceeds.
At 256, each input signal is identified for that bin. 2
At 58, a process of checking the usefulness of the data and creating a matrix is performed. When 25% of the bins, in this case 8 bins, had data, they were chosen to perform a consistency or validity check. A second check is performed on another 25% of the data. A check is also made to determine if the shift is too large. With this check successfully completed and one of the matrices filled, Hatch
Is applied at 260 and the heading is calculated and stored at 262.

Referring to FIGS. 7, 8 and 9, the manual calibration procedure is shown. In this case, no binning process is used. It is assumed that the short time required to drive the vehicle in a circle does not result in a magnetic change. First, the original xy
Save the data pair and perform a circular run until the data pair is reached again. Block 264 sets a debug message. At 266, it is checked from the data history whether the smoothed data and the unprocessed data are close. If these data are not close, 268 and subsequent processing are performed. If the data is good, the process of 270 is performed. Data is used because the data is good. Alternatively, if the data is poor but input for a long time, the data is accepted and a smoothing step is performed using the data. The goal is to discard bad data, but high frequency data reading works favorably. For example, in 2 seconds, the procedure looks at 24 data points. Since this is done continuously, if all the past data points are bad, the process starts with those points.

The initialization process of the calibration collection term is started at 272, sensor data is acquired at 274, and the reference voltage is removed at 276. A check is made at 278 to make sure the last two measurements are close, and if not, the process loops again at 280. Otherwise, the procedure continues at 282.

At 284, it is checked whether the maximum allowable points have been collected (if full rotation has not been performed). 286
Check to see if the vehicle has left the departure point, and if so, notify 288 at that point. Block 290 asks whether the vehicle has returned to the starting point. If the circle is a perfect circle or if the maximum number of points has been collected at 292, then a calibration algorithm is performed at 294. In the case of No at 292, the heading direction is calculated at 296 using the data points.

FIG. 10 shows a Wi-Fi system for abandoning bad data.
3 shows a flowchart of ldEdit. Since the data is found to be bad at 300, the data is not used, but the jitter is still searched. The raw data was too different from the smoothed data to detect jumps in the data (more than half the radius is the selected text). Check for jitter, even if ignored for calibration or heading purposes. Block 302 looks for jitter. At 304 the data is accepted and the jitter is also checked.

FIG. 11 shows a case where the received data is smoothed.

FIG. 12 shows a process of searching for a PC command, hearing the PC command, and responding to it.

FIG. 13 shows a system initialization procedure and E
5 illustrates loading from an EROM.

Description of Hardware-Circuit The hardware and circuit are shown in FIGS.
It is shown in FIG. 17A.

A small, low-cost magnetic compass implemented in an electronic circuit will now be described. The compass is suitable for use in vehicles or other applications suitable for magnetic compasses (eg, boats). The compass operates alternately in set / reset mode. The compass includes means implemented in hardware, which eliminates measurement drift due to changes in operating temperature that affect compass accuracy.
Along with mathematical calculations performed by the software, means are provided for eliminating perturbation effects of external magnetic influences that reduce the accuracy of a standard magnetic compass when used near large metal objects.

The above compass uses a magnetic resistance sensor,
Measure the alignment of the compass with the earth's magnetic field.

The sensor comprises four magnetic sensing elements, which change their resistivity when an external magnetic field is applied, and are arranged in a standard bridge circuit.
The sensing elements are arranged such that there is a "sense" measurement direction (or orientation) for the applied external magnetic field. The bridge is excited by an external voltage, and when there is no component of the external magnetic field applied in the sensing direction, the bridge is balanced and does not output. When an external magnetic field is applied in the sensing direction, the bridge becomes unbalanced and produces an output voltage proportional to the applied magnetic field. As the sensor (and the sensing direction of the sensor) rotates in the magnetic field, the output voltage of the bridge changes sinusoidally with a change in the orientation of the sensor angle with respect to the external magnetic field.

The sensor comprises means for flipping the polarity of the measurement output, so that the sensor can be used in a "set / reset" mode. That is, the sensing direction of the sensor bridge can be reversed. This polarity reversal provides a means of removing bias at the output of the sensor bridge caused by manufacturing tolerances and other error sources. Only the polarity of the output signal is flipped, and the error voltage is not flipped. The flip operation has the effect of converting the actual measured output to an alternating current (AC) waveform while the error voltage remains direct current (DC). The two signals can then be separated using a standard filtering circuit. After that, the DC of the sensor is
The error signal can be removed from the output.

The flip operation is generated by pulsing a “flip coil” provided in the sensor package. Very short time (about 10uS)
Of positive current pulses, the sensing direction of the sensor coincides with the "normal" orientation. The negative current pulse causes the output of the sensor to reverse polarity very quickly. By alternately applying current pulses in each direction at a certain normal rate (typically 1 kHz to 1 MHz), the measured output of the sensor changes to an alternating signal whose peak-to-peak amplitude is a function of the applied magnetic field. .

Although the sensor also has a "compensation coil", its use may be omitted. In the preferred embodiment, no compensation coil is used, since the compensation for the effects of external perturbations is performed by software. The coil may be provided in a sensor package proximate to the sensor bridge and include a magnetic field that can be measured by the sensor bridge. One use of this coil (and its common name) is for compensating for external perturbation effects. Current flows through the compensation coil to counteract the perturbation effect, effectively removing the perturbation effect from the measured output.

Another application of the compensation coil is to prevent changes in sensor sensitivity over a range of operating temperatures. Typically, the measurement sensitivity of a sensor decreases with increasing temperature. In certain applications, particularly in the case of automobiles, the temperature range is wide in the environment, which can lead to significant measurement errors. For example, due to solar load, winter driving and other factors, compasses used in automobiles may be between -40C and + 120C.
Exposure to operating temperatures of ° C. Changes in sensor output over this temperature range can significantly affect compass accuracy. For this reason, it is desirable to provide some means to compensate for changes in sensor output over temperature.

The sensor has the property of not sensing temperature at zero output (ie, when the bridge is in equilibrium). For this reason, a magnetic field having the same polarity as the applied external magnetic field but of the opposite polarity can be induced by using the compensation coil. The sum of these two fields, the external field and the compensating field, balances the bridge and operates the sensor at a point where the sensor output does not feel temperature. By measuring the amount of current required by the compensation coil to generate a magnetic field of equal and opposite polarity, the strength of the external (ie, earth) magnetic field can be measured indirectly.

However, using a compensation coil in this way eliminates the use for compensating for perturbation effects. However, these perturbation effects still exist and means must be provided to eliminate them. In the above compass, a mathematical algorithm is used instead of electromagnetic feedback through the compensation coil to eliminate measurement errors caused by perturbed external magnetic fields.

For use as a compass, two magnetic resistance sensors are required. The two sensors are orthogonal to each other (ie, at a 90 ° angle). Their sensing directions are such that one sensor measures the angle of the sensor bridge with respect to the Earth's magnetic north pole.
The other sensor is rotated 90 degrees, so when the compass points east, the sensor actually adjusts its sensing direction to the Earth's magnetic field. The use of two sensors gives enough information to be able to calculate the angular orientation of the sensor with respect to the earth's magnetic field using trigonometric functions.

This description makes reference to circuit diagrams, one set of which is shown in the provisional application, and another set of which is shown in FIGS.
16 to 16B and FIGS. 17 to 17A. Although the particular embodiment described suggests using an application specific integrated circuit (ASIC), this is not the only feasible approach. Depending on the application, some ancillary circuits that are not required or not suitable for all possible compass designs are also described. For example, there are some external components such as a power supply, which are not shown.

[0086] In general, the compass is implemented with two anti-magnetic sensors, several adjustment circuits and a microcomputer arranged as described above.

When a compensation coil is used, when a current flows through the compensation coil, the sensor output from the sensor changes.
How and in what direction the sensor output changes depends on the magnitude and direction of the current in the compensation coil, and it is the PWM calculated by the operating software program.
Directly controlled by value.

The Intel 89C51 microcomputer core is used to perform calculations, and various support circuits are required to implement a working processor. Random access memory (RAM) is provided for temporary read / write storage. A program for operating software is stored using a mask ROM (read only memory). EEROM (Electrically-Erasable-Re
ad-Only-Memory) is used for nonvolatile storage of data (for example, calibration information, variation parameters, and the like). This particular embodiment also includes circuitry for performing serial communication with other computing devices.

The sensor output is flipped via the current provided by switches Q1, Q1. The microcontroller controls software to generate switching speed signals and control signals for these switches.

As described above, the orientation of the compass with respect to the Earth's magnetic field is measured using two magnetically resistant sensors, one facing north and the other facing east. The output of each sensor is a relatively small voltage plus the error voltage caused by the non-ideal operation of the sensor. The set / reset mode of sensor operation separates the error component from the signal component. The AC measurement signal is converted to DC using a low-pass filter.
Isolate from error voltage. The cutoff frequency of the filter is set well below the flip frequency to isolate DC errors.
The resulting DC error is fed back to the amplifier,
This amplifier performs two functions: amplifying the AC measurement signal and removing DC error bias by negative feedback.
After removing the sensor error bias, the second amplifier
Provides greater signal gain.

The same amplification / feedback circuit is duplicated for each of the two sensors. There are also blocks that are used to generate the reference voltage and are required to be used with a unipolar power supply.

The adjustment output of the sensor is measured by an analog-digital converter (ADC) provided in the microcontroller. This converts the analog voltage generated by the sensor into a digital value that can be processed by the operating software. There is one ADC channel used for each of the two sensors, and yet another channel is required to measure the reference voltage (the reference voltage is a result of considering actual operation and requires the use of a reference voltage) In some embodiments, this is not the case.)

As described above, the temperature compensation of the sensor output is achieved by forcibly setting the sensor output to 0 by the compensation coil. The microcomputer performs temperature compensation by digital-to-analog conversion (DAC) of the digital values calculated by the software and used to generate the current through the compensation coil.
The circuit of the provisional application shows how to do this.

The digital value determined by the software is input to a pulse width modulator (PWM), which generates a continuous digital waveform, which has a duty cycle that depends on the digital input value. Digital waveform is 2
There can be only one state, ON and OFF, and the duty cycle is simply the quantitative ratio of the time to turn the waveform on and the time to turn the waveform off. By integrating the PWM output waveform with a suitable filter, an analog DC output voltage proportional to the duty cycle of the digital input signal can be generated.

Next, using the analog output voltage generated by the PWM and the filter, a current is generated in the compensation coil by the sensor of the illustrated circuit. The PWM value is used as the raw data input to the mathematical algorithm used by the program software to calculate the compass heading.

The software program can check the sensor output by ASDC measurement after the current is applied to the compensation coil and force the sensor output to zero by a process known as continuous approximation. The value of PWM required to invalidate the sensor output is an indirect measurement of the external magnetic field applied to the sensor. The PWM value is used as the raw data input to the mathematical algorithm used by the program software to calculate the compass heading.

FIG. 14 to FIG. 14B, FIG. 15, FIG.
6B and FIGS. 17-17A, an improved embodiment is shown. In general, the above description can be applied to this embodiment, but there are differences that can be understood by those skilled in the art.

14 to 14B and FIG. 15 are block diagrams of the circuit, and FIG. 15 is a detailed view of a portion surrounded by a dotted line 302.

Referring to FIGS. 16 to 16B, the member KM
Z51 is two magnetic resistance sensors, and their outputs V
O- and VO + are differential pressure outputs, which correspond to the angular position of the sensor in the magnetic field. 16 to 16B, outputs NOUT and EOUT are a north component and an east component of the heading direction vector, respectively. The VREF output is the output of the reference voltage for the amplifier circuit.

The input NPWM is a digital waveform input from the microcontroller having a duty cycle that can be set by the microcontroller controlling software. The circuit following NPWM is a low-pass filter, which converts the digital waveform to an analog DC voltage, which is proportional to the duty cycle of the input waveform. Further, the analog DC voltage is bisected and the resulting voltage is used as a reference voltage.

The ADREF output goes to the microcontroller and is used as a reference voltage for the microcontroller's ADC. It also generates VREF, where VREF is the bisected ADREF and is used as the reference voltage for all analog amplifiers in the circuit.

HIDRIVE and LODRIVE are:
Digital signal sent from the microcontroller to the complementary MOSFET switch. These flip the polarity of the anti-magnetic sensor. The flip increases the sensitivity of the sensor and removes the DC bias.

FIGS. 17 to 17A show a microcontroller U1. In this example, the microcontroller U1 is an INTEL 8051, in which a program is stored. Also, the microcontroller U1 has a PWM,
It incorporates a serial port circuit, a parallel input / output circuit, a random access data memory, and a processor state clock circuit. The microcontroller also has a HIDRI used to control the flip circuit for the circuit.
Provide VE and LODRIVE signals.

P1 is a power input and communication input / output connector. The circuit is a 5V voltage regulator, supplying all voltage to the circuit via VCC and VDD. U
Reference numeral 2 denotes an RS-485 transceiver, which can perform serial communication with an external device. U3 is an EEROM which is used for nonvolatile storage of calibration constants and operation parameters.

The operator sets various modes,
Switch used by the user to perform other control (S1)
There is also.

Integrated System FIGS. 18 and 19 show a variant of the system as an integrated inferred position sensor system based on a number of sensors including GPS, magnetic compass, cross-track accelerometer, angular velocity sensor and velocity sensor.

Due to the ambiguity due to the building or any physical structure, the ability to always track more than three GPS satellite signals cannot be guaranteed. For this reason, there is a strong desire to develop a system that can provide navigation during brief GPS signal loss. GPS
Flywheel integrated concepts and systems solve this problem by providing an integrated system of sensors. Successful selection of low cost sensor arrays provides high reliability at low cost. To this end, the preferred embodiment selects three sensors and one optional sensor to help GPS provide a constant and accurate navigation solution.

Heading information may be provided using the described low cost MR sensor compass and cross track accelerometer. A speed sensor is used to provide speed information in the heading direction. Angular velocity sensors are expensive, but are more accurate devices that can be used in place of cross-track accelerometers in applications requiring improved accuracy. These sensors complement each other well. The compass provides heading information but cannot provide movement information. Speed sensors provide speed information,
Cannot provide direction information or heading information. The cross-track accelerometer provides a second source of heading information and checks the compass heading, but must integrate the accelerometer data and apply it to the known heading. These sensor inputs can be sent to a Kalman filter. To optimally use the sensors, the variances of the various sensors may be calculated and provided to a Kalman filter. Further, a heuristic step may be performed before inputting the sensor input to the Kalman filter. Specifically, for example, an unreasonable change in heading supplied by a compass when a magnetic abnormality is present can be detected using a cross-track accelerometer. If the comparison between the cross-track accelerometer data and the compass heading information does not match due to the selected predetermined threshold,
This reading does not affect the output or calibration of the Kalman filter because the compass reading can be disabled and rejected.

The system operates in calibration mode when GPS quality is good. In the event of a fault, the calibrated output from other sensors provides navigation means.

The calibration of various sensors can be performed as follows.

Calibration of cross-track accelerometer: integral G
Means for calculating accurate velocity information from the carrier phase cycle of the PS are well known. By calculating the change in the rate of the speed, it is possible to perform a good estimation of the acceleration. The acceleration can be shown in a locally horizontal-north coordinate system. From the acceleration component, the acceleration component that does not follow the traveling direction can be calculated and directly distinguished from the acceleration supplied by the cross-track accelerometer. This calibration term can be stored and used to assist navigation in the event of a GPS failure. Also, a calibrated accelerometer may be used to reject erroneous compass readings.

Calibration of speed sensor: The speed sensor input may be calibrated again using reliable GPS speed information. In this case, the magnitude of the horizontal component of the speed derived by the GPS is calculated and compared with the speed sensor input. The speed sensor calibration term can be easily calculated as the difference between the above two speeds. A more accurate method stores the bias at various speeds and calculates the slope of the bias as a function of speed.

Calibration of the compass: GPS when moving again
The heading direction of the compass can be calibrated using the speed information. Some heuristic algorithms and filtering techniques need to be applied to ensure that the calibration term is generated only when the compass is providing normal heading information. That is, compass calibration should be performed only when there is no temporary magnetic abnormality. To get proofreading,
The compass heading should be compared directly with the north and east components of GPS speed, using a local horizontal north coordinate system. Further, it is natural that the calibration should be performed only when the magnitude of the GPS speed exceeds the specified threshold value.

Calibration of rotational speed sensor: Again, accurate G
The rotational speed sensor can be calibrated using the PS speed. In this case, the GPS speed is converted to a local horizontal coordinate system adjusted to the north. The output of the rotational speed sensor can be integrated to provide an angular displacement. The same displacement at the same time may be calculated by the difference in GPS speed. The difference between the two angular displacements is a calibration term for the speed sensor.

The magnetic compass, one of the sensors, has been described above. Referring to FIG.
2, 404 and 406 are a cross-track accelerometer, an angular velocity sensor and a velocity sensor interface, respectively. The cross-track accelerometer is oriented in a direction perpendicular to the vehicle truck and is sensitive to vehicle perpendicular acceleration. Using a cross-track accelerometer, measure the acceleration due to centrifugal force when the vehicle rotates. In that capacity, a cross-track accelerometer is used as a rotation sensor to assist the jitter algorithm. This algorithm assists in distinguishing the actual jitter from transient magnetic anomalies from a similar effect (measured vector shift) due to the rotation of the vehicle. In other cases, the vector change due to rotation may appear as jitter, so it is ignored by the input of the accelerometer as jitter, thereby processing only the jitter due to anomalies. Jitter due to magnetic anomalies can be distinguished from measured vector changes due to rotation of the vehicle.

The angular velocity sensor outputs a voltage output proportional to the rotation speed about the sensing axis. In this case, the sensing axis is provided perpendicular to the traveling surface of the vehicle, that is, in the vertical direction. Thus, it measures how fast the vehicle is rotating, ie the angular rotational speed. The output of the accelerometer is amplified by circuit 308,
The output of the microcontroller to the ADC is ACCEL as an analog voltage proportional to the cross-track acceleration.
And then used by the microcontroller software.

The anti-magnetic compass is looking north. In the integrated estimating position method, the output of the angular velocity sensor is used to assist or replace the output of the compass when the compass is affected by transient abnormalities. Thus, for example, if transient anomalies are detected by jitter detection, the system can weight the angular velocity sensor heavily or rely entirely on the angular velocity sensor to adjust heading.

The angular velocity sensor is initialized as usual in its application.

The output of the angular velocity sensor is amplified by the circuit 410 and the output R of the microcontroller to the ADC is obtained.
ATE is an analog voltage proportional to the angular velocity.

[0120] The speed sensor interface obtains speed information from the sensor, including but not limited to speed and direction. The logic circuit 412 decodes the information and outputs the outputs DIRECTION and SPEED to the parallel I / O.
And make the information available to the microcontroller bin, which information is used by the microcontroller.

Although not shown, the GPS receiver is connected to the serial port of the microcontroller.

As shown in FIG. 19, all sensors are integrated and each sensor is used in the most advantageous case.

While the principles of the invention have been described above with particular embodiments, it will be understood that this is by way of example and does not limit the scope of the invention.

[0124]

(Equation 1)

[0125]

(Equation 2)

[0126]

(Equation 3)

[0127]

(Equation 4)

[0128]

(Equation 5)

[0129]

(Equation 6)

[0130]

(Equation 7)

[0131]

(Equation 8)

[0132]

(Equation 9)

[0133]

(Equation 10)

[0134]

[Equation 11]

[0135]

(Equation 12)

[0136]

(Equation 13)

[0137]

[Equation 14]

[0138]

(Equation 15)

[0139]

(Equation 16)

[0140]

[Equation 17]

[0141]

(Equation 18)

[0142]

[Equation 19]

[0143]

(Equation 20)

[0144]

(Equation 21)

[0145]

(Equation 22)

[0146]

(Equation 23)

[0147]

(Equation 24)

[0148]

(Equation 25)

[0149]

(Equation 26)

[0150]

[Equation 27]

[0151]

[Equation 28]

[0152]

(Equation 29)

[0153]

[Equation 30]

[0154]

(Equation 31)

[0155]

(Equation 32)

[0156]

[Equation 33]

[0157]

(Equation 34)

[0158]

(Equation 35)

[0159]

[Equation 36]

[0160]

(37)

[0161]

(38)

[0162]

[Equation 39]

[0163]

(Equation 40)

[Brief description of the drawings]

FIG. 1 illustrates a problem that can occur in the prior art.

FIG. 2 is a guide for symbols used in the software flowcharts shown in FIGS. 3 to 13;

FIG. 3 is a flowchart of a software starting unit.

FIG. 4 is a flowchart of a main processing loop unit of software.

FIG. 5 is a continued flowchart of the main processing loop unit of the software.

FIG. 6 is a continued flowchart of the main processing loop unit of the software.

FIG. 7 is a flowchart of a manual calibration unit of software.

FIG. 8 is a continued flowchart of the software manual calibration section.

FIG. 9 is a continued flowchart of the software manual calibration section.

FIG. 10 is a flowchart of a Wild Edit unit of software.

FIG. 11 is a flowchart that follows the Wild Edit section of the software.

FIG. 12 is a flowchart of a communication input check unit of software.

FIG. 13 is a flowchart of a software loading unit from an EEPROM.

FIG. 14 is a block diagram of a circuit.

FIG. 14A is a block diagram of a circuit.

FIG. 14B is a block diagram of a circuit.

FIG. 15 is a detailed view following the block diagram of the circuit.

FIG. 16 is a schematic diagram of a circuit including a sensor.

FIG. 16A is a schematic diagram of a circuit including a sensor.

FIG. 16B is a schematic diagram of a circuit including a sensor.

FIG. 17 is a schematic diagram of a circuit including a microcontroller.

FIG. 17A is a schematic diagram of a circuit including a microcontroller.

FIG. 18 is a block diagram of a circuit of an integrated system using an on-board sensor.

FIG. 19 is another block diagram of a circuit of an integrated system using on-board sensors.

FIG. 19A is another block diagram of a circuit of an integrated system using on-board sensors.

FIG. 19B is another block diagram of a circuit of an integrated system using on-board sensors.

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification FI FI Theme Court ゛ (Reference) G01S 5/14 G01S 5/14 (71) Applicant 598166870 26 Misty Acres Road, Rolling Hills, CA 90274, United States of America (72) Inventor Kevin S. Judges Via Gavilan 4028, Paros Verdes Estates, California, 90274 California 7272 Inventor Ronald A. Hatch United States, 90744 Larkum Avenue, Wilmington, California 1142 F-term (reference) 2F029 AA02 AA04 AA05 AA08 AB07 AC02 AC04 AC12 5J062 AA05 BB01 BB02 BB03 CC07 EE04

Claims (14)

[Claims] An apparatus for automatically calibrating a mobile compass having a compass for providing heading output, electronic circuitry, and a computer programmed to calibrate the compass, wherein the apparatus is responsive to a magnetic field; An electronic compass having a sensor element for providing an output signal related to the earth's magnetic field and detected distortion; a means for converting the sensor output signal to a digital value; receiving the digital value of the sensor output signal and being defined by the output signal Ellipse matching and circularization of the ellipse
A computer section programmed to define a circle and determine heading readings; define a selected number of angle segments of the circles as bins and heading according to the bin to which the heading reading belongs. A computer programmed to store azimuth readings as data points; a computer programmed to offset the ellipse when data points are captured; one or more of a selected number of all bins A computer section programmed to calibrate the program being executed when the data points appear. 2. The apparatus of claim 1 wherein the computer is programmed to process input data from a compass at a rate of at least about five times per second. 3. The apparatus of claim 2, wherein said rate is approximately twelve times per second. 4. A method of automatically calibrating a mobile compass system installed in a vehicle, the system comprising: a compass providing heading output; an electronic circuit;
Providing an output signal from an electronic compass having a sensor element, converting the output signal into digital data values for use by the computer, and adjusting any path of the vehicle. Operating the vehicle, defining ellipses for each data point by performing elliptic matching and circularization procedures on the output data, and defining bins by selecting a large number of angular segments of the circle. Identifying the data points relative to the bins according to the location of the bins on the circle, offsetting the ellipse when the data points are captured, and defining a compass when each bin has at least one data point. Calibrating. 5. The method of claim 4, wherein the computer is programmed to process input data from a compass at a rate of at least about five times per second. 6. The method of claim 5, wherein the rate is approximately twelve times per second. 7. A system for detecting and rejecting transient magnetic anomalies in a mobile compass system having a programmed computer, wherein the system receives a reading from a compass and characterizes abnormal magnetic properties from the reading. A computer portion programmed to reject the use of readings belonging to the range of the label. 8. The system of claim 7, wherein the computer is programmed to process input data from a compass at a rate of at least about five times per second. 9. The system of claim 8, wherein said rate is approximately 12 times per second. 10. A method for detecting and rejecting transient magnetic anomalies in a mobile electronic compass system, comprising: receiving a reading from a compass and characterizing abnormal high frequency characteristics from the reading in a programmed computer; Rejecting use of compass readings within said characterization. 11. The method of claim 10, wherein the computer is programmed to process input data from a compass at a rate of at least about five times per second. 12. The method of claim 11, wherein said rate is approximately 12 times per second. 13. A mobile integrated location system, comprising: a wireless navigation system, a magnetic compass, and another sensor selected from the group consisting of: a. A cross-track accelerometer, b. A cross-track accelerometer and speed sensor, c. Angular velocity sensor, d. An angular velocity sensor and a velocity sensor; and means for determining heading and position using the selected sensor data. 14. The wireless navigation system according to claim 1, wherein
Means for determining a heading and a position using the selected sensor data, the signal providing means for providing a signal
14. The system of claim 13, wherein when the number of S satellites decreases from at least four to zero, the system selects to use one or more of the compass and other sensors.