阅读说明：本技术 低成本惯性导航系统 (Low cost inertial navigation system ) 是由 P·里德 J·马洛斯 M·度恩 A·皮特 于 2018-03-28 设计创作，主要内容包括：本公开涉及一种地下采矿车辆,该地下采矿车辆包括：三轴MEMS陀螺仪,其可绕旋转轴线旋转；以及陀螺仪接口,其计算：相对于与所述旋转轴不同的第一轴的第一旋转速率偏差；相对于与所述第一轴和所述旋转轴不同的第二轴的第二旋转速率偏差；通过使用所述第一旋转速率偏差和所述第二旋转速率偏差校正所述旋转测量数据基于地球旋转速率矢量计算绕所述旋转轴的旋转速率；并且基于所计算的绕所述旋转轴的旋转速率计算相对于所述旋转轴的第三旋转速率偏差。导航单元接收所述第一旋转速率偏差、所述第二旋转速率偏差和所述第三旋转速率偏差,并计算所述车辆的姿势。(The present disclosure relates to an underground mining vehicle comprising: a three-axis MEMS gyroscope rotatable about an axis of rotation; and a gyroscope interface that calculates: a first rotational rate offset relative to a first axis different from the rotational axis; a second rotational rate offset relative to a second axis different from the first axis and the rotational axis; calculating a rotation rate about the rotation axis based on an earth rotation rate vector by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias; and calculating a third rotation rate offset relative to the rotational axis based on the calculated rotation rate about the rotational axis. A navigation unit receives the first, second, and third rotation rate deviations and calculates a pose of the vehicle.)

1. An underground mining vehicle, comprising:

a tri-axial MEMS gyroscope mounted on a rotator configured to rotate the tri-axial MEMS gyroscope into a plurality of different orientations about an axis of rotation;

a gyroscope interface connected to the tri-axial MEMS gyroscope, the gyroscope interface configured to receive rotation measurement data from the tri-axial MEMS gyroscope with respect to the plurality of different orientations and further configured to:

calculating a first rotation rate bias of the tri-axis MEMS gyroscope relative to a first axis different from the axis of rotation based on the rotation measurement data;

calculating a second rotation rate bias of the tri-axial MEMS gyroscope relative to a second axis different from the first axis and different from the rotational axis based on the rotation measurement data;

calculating a rotation rate about the rotation axis based on an earth rotation rate vector by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias; and is

Calculating a third rotation rate bias of the tri-axis MEMS gyroscope relative to the rotational axis based on the calculated rotation rate about the rotational axis; and

a navigation unit connected to the gyroscope interface and configured to:

receiving the first, second, and third rotation rate biases from the gyroscope interface; and

calculating a pose of the underground mining vehicle based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

2. The underground mining vehicle of claim 1, further comprising a vehicle controller connected to the vehicle and configured to:

stopping the vehicle;

calibrating an inertial navigation system while the vehicle is stopped;

rotating the three-axis gyroscope by the rotator while the vehicle is stopped;

causing the gyroscope interface to calculate the first, second, and third rotation rate biases; and is

Restoring movement of the vehicle based on the calculated pose.

3. The underground mining vehicle of claim 1 or 2, further comprising a three-axis accelerometer communicatively coupled to the navigation unit, wherein the navigation unit is configured to calculate the pose based on acceleration data from the three-axis accelerometer.

4. The underground mining vehicle of any of the preceding claims, wherein the navigation unit is an inertial navigation unit and is configured to determine an absolute position of the underground mining vehicle based on the pose.

5. The underground mining vehicle of any of the preceding claims, wherein the gyroscope interface is configured as a function ofCalculating around saidRotation rate omega of the rotating shaft_thirdWherein ω is_EarthIs the magnitude, ω, of the vector of the rate of rotation of the earth_firstAnd ω_secondAre the rotation rate about the first axis and the rotation rate about the second axis corrected by the calculated first rotation rate deviation and second rotation rate deviation, respectively.

6. The underground mining vehicle of any of the preceding claims, further comprising a filter connected to the gyroscope interface and configured to continuously track the third rotation rate bias.

7. The underground mining vehicle of any of the preceding claims, further comprising an additional three-axis MEMS gyroscope mounted in a fixed position and attitude relative to the underground mining vehicle,

wherein the navigation unit is connected to the further three-axis MEMS gyroscope and is configured to receive further rotation measurement data from the further three-axis MEMS gyroscope and to calculate the pose of the underground mining vehicle based on the further rotation measurement data.

8. The underground mining vehicle of any of the preceding claims,

the gyroscope interface being configured in accordance withCalculating a rotation rate ω about the rotation axis_thirdWherein ω is_EarthIs the magnitude, ω, of the vector of the rate of rotation of the earth_firstAnd ω_secondIs a rotation rate about the first axis and a rotation rate about the second axis corrected by the calculated first and second rotation rate deviations, respectively;

the underground mining vehicle further includes a three-axis accelerometer communicatively coupled to the navigation unit, and the navigation unit is configured to calculate the pose based on acceleration data from the three-axis accelerometer; and is

The underground mining vehicle further includes a further three-axis MEMS gyroscope mounted in a fixed position and attitude relative to the underground mining vehicle, and the navigation unit is connected to the further three-axis MEMS gyroscope and is configured to receive further rotation measurement data from the further three-axis MEMS gyroscope and to calculate the attitude of the underground mining vehicle based on the further rotation measurement data.

9. A method for calculating a pose of an underground mining vehicle, the method comprising:

rotating a three-axis MEMS gyroscope about a rotation axis into a plurality of different orientations;

receiving rotation measurement data from the tri-axial MEMS gyroscope for the plurality of different orientations;

calculating a first rotation rate bias of the tri-axis MEMS gyroscope relative to a first axis different from the axis of rotation based on the rotation measurement data;

calculating a second rotation rate bias of the tri-axial MEMS gyroscope relative to a second axis different from the first axis and different from the rotational axis based on the rotation measurement data; and

calculating a rotation rate about the rotation axis based on the earth rotation rate vector by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias; and

calculating a third rotation rate bias of the tri-axis MEMS gyroscope relative to the rotational axis based on the calculated rotation rate about the rotational axis; and

calculating a pose of the underground mining vehicle based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

10. The method of claim 9, further comprising:

stopping the vehicle;

calibrating an inertial navigation system while the vehicle is stopped;

rotating the three-axis gyroscope while the vehicle is stopped;

calculating the first, second, and third rotation rate deviations; and

restoring movement of the vehicle based on the calculated pose.

11. The method of claim 10, wherein stopping the vehicle comprises stopping the vehicle for 10 seconds or more and 300 seconds or less before resuming movement of the vehicle.

12. The method of claim 10 or 11, wherein restoring movement of the vehicle comprises restoring movement greater than or equal to zero meters and less than 100 meters.

13. The method of any of claims 9 to 12, wherein the method further comprises calculating a heading-independent gesture component based on acceleration data from a three-axis accelerometer.

14. The method of any one of claims 9 to 13, wherein the method further comprises determining an absolute position of the underground mining vehicle based on the pose.

15. The method of any of claims 9 to 14, wherein the method further comprises calculating the third rotation rate bias by calculating a rotation rate about the rotation axis and calculating the third bias based on the calculated rotation rate about the rotation axis and the earth rotation rate vector.

16. The method of claim 15 wherein calculating the rotation rate comprises correcting the rotation measurement data using the first and second rotation rate biases.

17. The method of claim 16 wherein calculating a rate of rotation ω about the axis of rotation_thirdIncluding calculation ofWherein ω is_EarthIs the magnitude, ω, of the vector of the rate of rotation of the earth_firstAnd ω_secondAre the rotation rate about the first axis and the rotation rate about the second axis corrected by the calculated first rotation rate deviation and second rotation rate deviation, respectively.

18. The method of any of claims 9 to 17, wherein the method further comprises continuously tracking the first, second and third rotation rate deviations.

19. The method of any of claims 9 to 18, wherein the method further comprises:

calculating two candidate solutions for the third rotation rate;

correcting the rotation measurement data based on the initial estimate of the third rotation rate bias to determine an estimated third rotation rate; and

selecting one of the two candidate solutions that is closest to the estimated third rotation rate.

Technical Field

The present disclosure relates to inertial navigation systems and methods.

Background

Navigation is a basic process that is performed in a moving vehicle and broadly involves determining the current position and pose of the vehicle with reference to a global reference frame so that a direction or path can be determined toward a target position or along a target path. Some navigation methods rely on reference information, such as stars, landmarks, or satellites (GNSS). However, in certain applications, such as underground mining, such reference information is not always available. Thus, the navigation unit of the vehicle typically navigates based on dead reckoning. This involves determining the speed and direction of travel of the vehicle and then adding a corresponding vector to the current position to predict the new position.

While the current velocity or acceleration can be determined relatively accurately, it is often difficult to determine the north direction. The accelerometer is sufficiently accurate to determine the local vector direction of acceleration due to earth's vertical gravity (or simply "gravity vector"), which is a combination of acceleration due to earth's gravity and centripetal rotation, and is equivalent to the surface normal of a local horizontal surface (or simply "local horizontal plane"). FIG. 1 shows the Earth 100 rotating about a north axis 101, as indicated by arrow 102. The vehicles 103 are located on the earth 100 (not to scale). FIG. 1 shows a vehicle reference frame 105 including an x-axis 106, a y-axis 107, and a z-axis 108. Assuming that the vehicle 103 is horizontal, the z-axis represents the surface normal, which means that the only unknown in the vehicle 103 pose is the rotation of the reference frame 105 about the z-axis. The process of determining the rotation or, in other words, the direction of the y-axis 107 (or x-axis 106) is referred to as north seeking. If the gyroscope is accurate enough to detect the rate of rotation of the earth, then the gyroscope can be used to find north. Based on the vector of the rotation rate of the earth (or simply "rotation rate vector") and the locally measured gravity vector, the orientation of the local reference frame 105 relative to the earth reference frame can be completely determined. This allows the vehicle to be steered to a desired direction (e.g., north or any other direction).

Compact North-seeking System in IEEE Sensors Journal (IEEE sensor Journal) by Bojja et al, 2016, 6, 8, discloses a North-seeking System that includes a 2-axis MEMS accelerometer and a 1-axis Fiber Optic Gyroscope (FOG) rate sensor. FOG has the advantage that they have good accuracy, but they have the disadvantage that their fiber components are difficult to manufacture, which makes these devices expensive.

Disclosure of Invention

The present disclosure provides a north-seeking system using a three-axis MEMS gyroscope. The system allows a correction process to be performed that estimates the rotation rate bias in all three axes and uses the rotation rate bias to correct inaccuracies inherent in low cost MEMS gyroscopes.

MEMS represent micro-electromechanical systems, typically systems with dimensions between 20 microns and 1 mm, and include electronic and mechanical components. In the case of MEMS gyroscopes, there are mechanical components (e.g. the vibrating mass) as well as electronic components that detect the out-of-plane vibration of the vibrating mass. The mechanical and electronic components are typically manufactured as a unitary structure. The weight (weight) is manufactured, for example, on the same silicon substrate by grinding or etching, and the electronic component is manufactured, for example, on the silicon substrate by photolithography.

Underground mining vehicles may include longwall mining machines (including associated rails, roof supports, drives, conveyors, platform loaders and crushers), continuous mining machines, roadway driers, loaders, handlers, people carriers, rescue vehicles, shuttle cars, flexible haulers, coal planers, or any other machine equipment with or without an extraction device to remove material from a mine.

An underground mining vehicle comprising:

a tri-axial MEMS gyroscope mounted on a rotator configured to rotate the tri-axial MEMS gyroscope into a plurality of different orientations about an axis of rotation;

a gyroscope interface connected to the tri-axial MEMS gyroscope and configured to receive rotation measurement data from the tri-axial MEMS gyroscope with respect to the plurality of different orientations, and further configured to:

calculating a first rotation rate bias of the tri-axis MEMS gyroscope relative to a first axis different from the axis of rotation based on the rotation measurement data;

calculating a second rotation rate bias of the tri-axial MEMS gyroscope relative to a second axis different from the first axis and the rotation axis based on the rotation measurement data;

calculating a rotation rate about the rotation axis based on an earth rotation rate vector by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias; and is

Calculating a third rotation rate bias of the tri-axis MEMS gyroscope relative to the rotational axis based on the calculated rotation rate about the rotational axis; and

a navigation unit connected to the gyroscope interface and configured to:

receiving the first, second, and third rotation rate biases from the gyroscope interface; and

calculating a pose of the underground mining vehicle based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

The structure of MEMS gyroscopes facilitates the large-scale manufacture of such gyroscopes in semiconductor processes, which makes them much cheaper than Fiber Optic Gyroscopes (FOG). This allows for a wider range of MEMS applications. Further, using a gyroscope having three axes, the posture can be continuously maintained during the movement. More specifically, the three-axis MEMS gyroscope is rotated about an axis of rotation to calculate first and second rotational rate biases. These deviations can be used to calculate the deviation about the axis of rotation by inference using the statically sensed earth rotation rate vector. This results in a significantly more accurate calculation of the deviation about the axis of rotation than if other methods including iterative numerical optimization methods were used. This compensates for the lower accuracy of the MEMS gyroscope, which in particular solves the problem of poor repeatability of MEMS biases. Furthermore, only a single axis of rotation is required, which reduces the need for gimbals and other complex mechanical structures. At the same time, the yaw rate deviations about a single rotational axis are precisely determined.

The underground mining vehicle preferably includes a vehicle controller connected to the vehicle and configured to:

stopping the vehicle;

calibrating an inertial navigation system while the vehicle is stopped;

rotating the three-axis gyroscope by the rotator while the vehicle is stopped;

causing the gyroscope interface to calculate the first, second, and third rotation rate biases; and is

Restoring movement of the vehicle based on the calculated pose.

The underground mining vehicle also preferably includes a three-axis accelerometer communicatively coupled to the navigation unit, wherein the navigation unit is configured to calculate the pose based on acceleration data from the three-axis accelerometer.

The navigation unit is preferably an inertial navigation unit and is configured to determine an absolute position of the underground mining vehicle based on the pose.

The gyroscope interface is preferably further configured to calculate the third rotation rate bias by calculating a rotation rate about the rotational axis and to calculate the third rotation rate bias based on the calculated rotation rate about the rotational axis and the earth rotation rate vector.

Preferably, the gyroscope interface is further configured to calculate the rotation rate by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias.

Preferably, the gyroscope interface is configured according toCalculating a rotation rate omega about a third rotation axis_thirdWherein ω is_EarthIs the magnitude of the vector of the rotation rate of the earth, ω_firstAnd ω_secondAre the rotation rate about the first axis and the rotation rate about the second axis corrected by the calculated first rotation rate deviation and second rotation rate deviation, respectively.

Preferably, the underground mining vehicle further comprises a filter connected to the gyroscope interface and configured to continuously track the third rotation rate bias.

Preferably, the underground mining vehicle further comprises a further three-axis MEMS gyroscope mounted in a fixed position and attitude relative to the underground mining vehicle, wherein the navigation unit is connected to the further three-axis MEMS gyroscope and is configured to receive further rotation measurement data from the further three-axis MEMS gyroscope and to calculate the attitude of the underground mining vehicle based on the further rotation measurement data.

It is advantageous to have a second fixed three-axis gyroscope that operates in a "continuous integration mode" and is calibrated by means of a rotational setting, particularly when the vehicle is stationary.

Preferably, the gyroscope interface is configured according toCalculating a rotation rate omega about a third rotation axis_thirdWherein ω is_EarthIs the magnitude of the vector of the rotation rate of the earth, ω_firstAnd ω_secondIs a rotation rate about the first axis and a rotation rate about the second axis corrected by the calculated first and second rotation rate deviations, respectively;

the underground mining vehicle further includes a three-axis accelerometer communicatively coupled to the navigation unit, the navigation unit configured to calculate the pose based on acceleration data from the three-axis accelerometer; and is

The underground mining vehicle further includes a further three-axis MEMS gyroscope mounted in a fixed position and attitude relative to the underground mining vehicle, and the navigation unit is connected to the further three-axis MEMS gyroscope and is configured to receive further rotation measurement data from the further three-axis MEMS gyroscope and to calculate the attitude of the underground mining vehicle based on the further rotation measurement data.

A method for calculating a pose of an underground mining vehicle, the method comprising:

rotating a three-axis MEMS gyroscope about a rotation axis into a plurality of different orientations;

receiving rotation measurement data from the tri-axial MEMS gyroscope for the plurality of different orientations;

calculating a first rotation rate bias of the tri-axis MEMS gyroscope relative to a first axis different from the axis of rotation based on the rotation measurement data;

calculating a second rotation rate bias of the tri-axial MEMS gyroscope relative to a second axis different from the first axis and the rotation axis based on the rotation measurement data;

calculating a rotation rate about the rotation axis based on the earth rotation rate vector by correcting the rotation measurement data using the first rotation rate bias and the second rotation rate bias;

calculating a third rotation rate bias of the tri-axis MEMS gyroscope relative to the rotational axis based on the calculated rotation rate about the rotational axis;

calculating a pose of the underground mining vehicle based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

Preferably, the method further comprises:

stopping the vehicle;

calibrating an inertial navigation system while the vehicle is stopped;

rotating the three-axis gyroscope while the vehicle is stopped;

calculating the first, second, and third rotation rate deviations; and is

Restoring movement of the vehicle based on the calculated pose.

Preferably, stopping the vehicle includes stopping the vehicle for 10 seconds or more and 300 seconds or less before resuming the movement of the vehicle.

Preferably, restoring movement of the vehicle comprises restoring movement greater than or equal to zero meters and less than 100 meters.

Preferably, the method further comprises calculating a heading-independent gesture component based on acceleration data from the three-axis accelerometer.

Preferably, the method further comprises determining an absolute position of the underground mining vehicle based on the pose.

Preferably, the method further comprises calculating the third rotation rate bias by calculating a rotation rate about the rotation axis and calculating the third bias based on the calculated rotation rate about the rotation axis and the earth rotation rate vector.

Preferably, calculating the rotation rate comprises correcting the rotation measurement data using the first rotation rate offset and the second rotation rate offset.

Preferably, a rotation rate ω about the third rotation axis is calculated_thirdIncluding calculation ofWherein ω is_EarthIs the magnitude, ω, of the vector of the rate of rotation of the earth_firstAnd ω_secondAre the rotation rate about the first axis and the rotation rate about the second axis corrected by the calculated first rotation rate deviation and second rotation rate deviation, respectively.

Preferably, the method further comprises continuously tracking the first, second and third rotation rate deviations.

Use of a three axis MEMS gyroscope for vehicle navigation is provided by rotating the three axis MEMS gyroscope, calculating a three axis rate bias based on an earth rotation rate vector and calculating a pose of a vehicle based on the earth rotation rate vector and the three axis rate bias.

Optional features described for any aspect of the vehicle or method are similarly applicable to other aspects described herein, where appropriate.

Drawings

Fig. 1 shows a local frame of reference of a vehicle on earth according to the prior art.

Embodiments will be described with reference to the following drawings:

figure 2a schematically shows an underground mine.

Fig. 2b shows the main measurements of yaw, pitch and roll.

Fig. 3a, 3b, 3c and 3d show the principle of rotating the flag (maytag).

Fig. 4a, 4b, 4c and 4d show rotation markers for earth rotation axis singularities.

Fig. 5 shows an underground mining vehicle.

Fig. 6 shows a three-axis MEMS gyroscope.

Fig. 7 illustrates a method for calculating a pose of an underground mining vehicle.

Detailed Description

North seeking (or more professionally, "gyro platform north") is a sufficiently sensitive and accurate gyroscope's ability to detect the rate of rotation of the earth present in any observation frame 105 that is fixed relative to the rotating earth 100. Furthermore, the quantity is usually expressed as a vector, which contains the rotation rate magnitude as the magnitude of the vector, the rotation axis 101 as the direction of the vector, and the rotation direction 102 defining the rotation direction with respect to the vector using the right-hand rule. The earth rotation rate axis 101 and direction is fixed, superficially, relative to the earth's surface such that the vector observed at the north pole points directly up from the local horizontal plane, the vector observed at the south pole points directly down from the local horizontal plane, and parallel to the local horizontal plane at the equator (as shown in fig. 1). Thus, for local observation of this vector, this direction can be used for "north finding", and therefore for the gyro platform north pointing function.

The magnitude of the vector represents the rate of rotation of the earth, which is 7.292x 10 per second^-5Radian, or about 15.04 degrees per hour. This is not 15 degrees per hour as would be expected for 24 hours a day, since it is to accommodate the earth's orbit around the sun; the sun is measured at 15.00 degrees per hour, while the velocity at 15.04 degrees per hour is the sidereal velocity observed by the inertial measurement system.

However, the sensitivity of the gyroscope used for the north-seeking of the gyro platform should be significantly better than this rate in order to be able to resolve the components of the earth rotation rate vector. At the equator, all of these observed rates are present in the local horizontal plane horizontal component, and thus represent the maximum sensitivity achievable. This decreases with increasing latitude (as cosecant of latitude) until the vector has a very small component in the horizontal plane near the pole. If the orientation of the vehicle reference frame 105 is not horizontal and can approach either extreme (parallel or perpendicular to the earth rotation rate vector), the north-seeking sensitivity of the gyroscope can vary greatly.

Fig. 2a schematically shows an underground mine 200. The underground mine 200 comprises a hoistway 201 and an access ramp 202 as well as the mining vehicle 103 in fig. 1. As can be seen in the figure, the attitude of the underground mine vehicle 103 changes as it travels through the access ramp 202. In some embodiments, the access ramp 202 has a helical shape, which means that the attitude of the underground mining vehicle 103 changes significantly.

Fig. 2b shows an exemplary pose of the underground mining vehicle 103. In this embodiment, the pose is described by the standard euler angles conventionally reported by inertial measurement systems (INS) to describe the orientation of the INS. Euler angle pitch 251, roll 252 and yaw 253 are defined in turn about coordinate system axes 'x', 'y', 'z', respectively. The order represents a selected convention and uniquely defines the orientation of the INS with respect to local horizontal and geographical north. However, in other embodiments, different labels and appointment selections may be used as well.

Sensitivity of north finding

Returning to the issue of sensitivity, the accuracy of north pointing for a given latitude using a gyro platform is a function of the sensitivity of the gyro component in the sensing plane.

The approximation provides the expected accuracy of the quadrant angle as a function of gyroscope component noise:

where δ is the gyroscope rate uncertainty in degrees/hour and θ is the angular pointing accuracy that can be expected. Note that when the gyroscope rate uncertainty is equal to the rate of rotation of the earth, the accuracy level is ± 45 degrees, meaning that the measurement cannot determine the direction within the compass quadrant. To achieve an accuracy on the order of 1 degree, the gyroscope rate uncertainty needs to be better than about 0.2 degrees/hour by this estimation.

Technical parameters of MEMS inertial measurement unit

With respect to north finding applications, the most relevant aspect of Inertial Measurement Unit (IMU) technology parameters is gyroscope technology parameters. The strength of the locally measured gravity vector is usually large (9.8 m/s)²Also known as 1g) so that even a grade of 10 milli-g (98 mm/s) is obtained²) Also represents an accuracy of the order of 0.5 degrees (for the determination of the level). In contrast, a MEMS gyroscope with an uncertainty on the order of 20 degrees per hour can provide a pointing accuracy of no more than 50 degrees, which is insufficient for north-seeking applications.

At this point, some clarification of various uncertainties in both the accelerometer and gyroscope is necessary. Two key technical parameters are bias instability and bias repeatability. The bias on both the accelerometer and the gyroscope are the primary uncertainties that are used to add a positive or negative signal to the measurement signal in a manner that may be unpredictable. The deviation can be compensated in some way in order to retrieve a signal that enables the direction to be determined to the required accuracy. Bias instability refers to the ultimate stability of the bias when all other effects (e.g., drift and white noise) are removed. This indicates the degree of stability of the deviation and the speed of the expected change. In other controlled environments (e.g., temperature stable environments), it is expected that gyroscopes with a bias instability of 0.05 degrees/hour will not change faster than this rate. However, bias repeatability describes the repeatability of the bias under a multitude of control factors (e.g., turn-on repeatability, warm-up, temperature, and aging). Bias instability can limit the variation of the bias without these factors, but in other cases this is often overwhelmed by repetitive factors and therefore represents a functional limitation. For MEMS gyroscopes with bias repeatability of no more than 20 degrees per hour, small bias instabilities may not ensure pointing accuracy unless the repeatability is compensated for by measurement or control. For reference, a low-grade FOG with a very small fiber loop area exhibits a deviation repeatability of the order of 2 degrees per hour.

As mentioned above, to achieve FOG level pointing performance, poor bias repeatability specifications should be compensated for in some way. Two options are generally considered: (1) controlling the temperature; and (2) rotating the marker (maytaging), also known as indexing.

Temperature control

Due to the nature of MEMS, these systems (both throughout the cell and within the cell) are very sensitive to temperature changes depending on temperature gradients. Thus, temperature control is performed in the MEMS IMU as part of normal operation. The residuals after this control are typically reported in the manufacturer's technical parameters, so any further external compensation needs to be significantly improved according to the basic technical parameters.

If the system continues to maintain a constant temperature, any changes from the initial power-up of the MEMS will eventually equilibrate and remain constant within the fundamental stability limits (bias instability in the case of a bias). Unless the deviation is measured independently in this constant temperature regime, even this value may drift at a rate at which the deviation is unstable, so that it may grow to the limit specified by the repeatability in sufficient time.

Furthermore, this represents only one problem to be solved in the biased temperature control. In practice, maintaining the constancy of temperature and temperature gradients in MEMS in real environments can be very problematic. Mechanisms to heat and cool the MEMS may be used, or alternatively controlled at one extreme where thermal gain/loss to the environment may be warranted, allowing only cooling/heating, respectively, to be used. This increases the volume of the instrument and other factors to ensure that the temperature control is operating properly.

Typically, the unit is also restarted, so that a warm-up transient exists. Even if the same temperature is reached each time, the opening deviation repeatability can result in unknown deviations. In general, these problems make compensating for deviations by this method challenging. Therefore, a second selectable rotational marker (maytaging) may be preferred.

(rotating mark)

This process refers to physically rotating the system about a selected axis, which is the rotation marker axis in the presence of a known rotation rate signal (typically the earth rotation rate), by a precisely known amount. This is also referred to as "indexing", wherein different rotational orientations represent different and precisely known index positions. Typically, there are two positions representing a precise 180 degree rotation. This process has the effect of: any signal present in an axis perpendicular to the axis of rotation of the indicia will be sign-inverted. The mine vehicle 103 may remain stationary for the rotary marker rotary marking process.

In one embodiment, the first, second, and third (rotational) axes herein are perpendicular to each other and fixed relative to the vehicle 103 or in a known orientation relative to the vehicle 103, e.g., the first axis is parallel to the forward direction of the vehicle 103; the first axis is parallel to the lateral direction of the vehicle 103 and the rotation axis is parallel to the vertical axis of the vehicle. However, other orientations may be suitable, and the three axes may not be exactly perpendicular but linearly independent. Similarly, the rotation about the axis of rotation may be in 180 degree increments or 90 degree increments, for example, in the direction of the first and second axes, or may be in any other step size, so long as it results in linearly independent measurements.

Fig. 3a, 3b, 3c and 3d show the rotary marking principle. In positions 3a to 3d, the rotation marker axis is a vertical dashed line 301, the sensing axis is a horizontal dashed arrow 302, the measurement signal is a black arrow, and the deviation is a thick arrow. In 3a and 3b there is no offset, so when the sense axis 302 is rotated 180 degrees around the rotation mark axis 301, it will be seen that the horizontal signal along that axis changes sign, but remains the same magnitude. Note that the vertical signal remains unchanged because there is no change in this direction. The same signals measured in 3c and 3d have deviations with the shaft movement. The sensing axis changes direction and the offset moves with this axis. Note again that the signal and offset with respect to the rotating marker axis does not change. Thus, in one position, a larger signal is seen on the sensing axis, while in another position a smaller signal is seen. Thus, the deviation can be unambiguously inferred from the difference between the two signals in each position. However, the deviation on the axis of the rotating mark is still unknown because it does not change. It is also noted that although north seeking is generally the target for measuring the rate of rotation of the earth, as an intermediate step in the north seeking process, the rate of rotation of the earth itself is used as a reference signal to reveal the bias present in the system using the principle of rotational tagging.

Thus, in a 3D system, a single axis of rotation marker can be used to infer the deviation on the other two perpendicular axes. For example, a rotational marker about the x-axis can be used to account for deviations in the y-axis and z-axis, where the deviations in the x-axis remain unknown. In applications where a full navigation solution is required, this may not be sufficient, as it is desirable to know the signals about all three axes accurately in order to produce a solution. In the context of north finding applications, using only two of the three axes is sufficient to determine the north direction. However, there are pathological conditions using this approach.

Rotational marker pathologic condition-ground axis alignment singularity

Fig. 4a, 4b, 4c and 4d show rotation markers for earth rotation axis singularities. Consider the case where the instrument rotation marker axis may or may not be aligned along the earth rotation rate vector for a given position. In fig. 4a to 4d, the sensed earth rotation rate vector is shown in a thin solid line 401, the rotation marker axis is the z-axis in a thick line 402, and the x and y axes are dashed lines 403 and 404, respectively. To illustrate the 3D dependency, fig. 4a to 4D are shown with "bottom views" (4a and 4b) and "side views" (4c and 4D), where the bottom views are viewed from below the x-y plane in the direction of the positive z-axis and the side views are viewed along the negative x-axis towards the z-y plane. The earth rotation rate vector component in the x-y plane allows the north angle (as previously described in fig. 2) to be easily measured if there is a signal on all axes as shown in the two views given in fig. 4a and 4 c. This will occur when the rotation marker axis is not aligned with the earth rotation rate vector. However, when the rotation mark axis is closely aligned to the vector, the situation is similar to the views given in 4b and 4d, where there is little signal in the x-y plane. North alignment is difficult to determine because it becomes very noisy since the signal in this plane is small and comparable to the gyro noise level. This is not a problem if z-axis signals can be used, as angles formed in other z-x and z-y planes can be referenced to north. However, the rotation marker about the z-axis only provides knowledge of the signal in the x-y plane, and the unknown deviations in the z-axis can completely mask the true signal and thus the north direction.

This situation represents a pathological condition of the single axis rotation marker. When the system aligns the rotation marker axis with the earth rotation rate vector, the accuracy is reduced to the point where no direction can be determined. This may occur at any latitude. For example, if operating at the equator, the earth rotation rate vector is horizontal and points north. Any system with a direction of the axis of the rotating marker closely aligned to this direction cannot clearly resolve north. Similarly, at some intermediate latitude, the earth rotation rate vector will point north but tilt toward local horizontal due to the current latitude. Any system with a rotation mark axis aligned with this direction will prevent accurate orientation.

A final comment on this question can be made. The quality of the azimuth solution degrades as the axis of rotation of the earth is closely aligned. This in effect provides the information: the alignment of the vehicle must therefore be close to north. Nevertheless, the problem still exists in the accuracy of approaching alignment north. This is not directly solvable.

Completely upright pathological conditions

Another situation may occur when the vehicle is fully level. However, this has no relation to the disadvantage of the single-axis rotary marker; this is a problem that exists in all INS and is only relevant to defining singularities inherent in orientation with respect to the local vertical. Using the IMU accelerometer to measure the level of the instrument, the IMU accelerometer unambiguously determines the direction of the locally measured gravity vector without other motional forces or error sources in the instrument. The same problem as shown in fig. 4 occurs when the system is very close to the vertical, in which case the locally measured gravity vector signal is aligned with the vertical accelerometer axis. Here, however, the north direction determined by the instrument does not become uncertain, but is very sensitive to tilt. The reported orientation can swing around the vertical direction depending on the tilt direction, which can change direction significantly due to the small component of the gravity vector in the horizontal plane. To a large extent, this can be circumvented simply by carefully selecting the definition of the orientation.

Underground mining vehicle

Fig. 5 shows an underground mining vehicle 500 comprising a three-axis MEMS gyroscope 501 mounted on a rotator 502. Rotator 502 is configured to rotate the tri-axial MEMS gyroscope 501 about an axis of rotation 503 into a plurality of different rotational orientations, which are indicated by arrow 504. The mining vehicle 500 further comprises a gyroscope interface 505 connected to the three-axis MEMS gyroscope 501. The gyroscope interface 505 is configured to receive rotation measurement data from the three-axis MEMS gyroscope 501 for a plurality of different rotational orientations. The rotation measurement data may include the rotation rate (including the corresponding offset) measured by the gyroscope. For example, the rotation measurement data may include the currently measured rotation rates about the three axes of three-axis gyroscope 501, and may be a data triplet (w1, w2, w3) for each sample time. Note that the rotation measurement data may also be pre-processed by gyroscope 510, e.g., smoothed. In this sense, receiving rotation measurement data for different rotational orientations means receiving a first data triplet with respect to a first orientation (e.g. 0 degrees), receiving a second data triplet in a second orientation (e.g. 90 degrees), and so on. Gyroscope interface 505 is further configured to calculate a first rotation rate bias of three-axis MEMS gyroscope 501 about a first axis different from rotation axis 503 based on the rotation measurement data. Gyroscope interface 505 is further configured to calculate, based on the rotation measurement data, a second rotation rate bias of three-axis MEMS gyroscope 501 about a second axis different from the first axis and different from rotation axis 503, and to calculate, based on the first rotation rate bias, the second rotation rate bias, and the Earth rotation rate vector, a third rotation rate bias of three-axis MEMS gyroscope 501 about the rotation axis.

One embodiment of a MEMS gyroscope is Apogee-a from SBG Systems, which reports a bias operational instability <0.08 degrees/hour.

The mine vehicle 500 further comprises a navigation unit 506 connected to the gyroscope interface 505. Navigation unit 506 is configured to receive the first, second, and third rotation rate biases calculated by gyroscope interface 505. The navigation unit 506 is further configured to calculate a pose of the underground mining vehicle 500 based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

When the term "construction" is used, this may refer to a variety of different ways to construct the system. For example, rotator 502 may be a microprocessor, FPGA, ASIC, or other configurable device programmed to rotate a gyroscope, or may be a chassis and motor that automatically rotate a gyroscope without further input. Similarly, the gyroscope interface and navigation unit 506 may be a processor programmed to perform the described functions, or may be an ASIC, FPGA, or other configurable device configured to perform those functions. It should also be noted that the rotator 502, the gyroscope interface 505, and the navigation unit 506 may be implemented across one or more hardware devices. In particular, rotator 502, gyroscope interface 505, and navigation unit 506 may be implemented on the same microprocessor using separate or overlapping programming code, processes, or threads to represent these units.

The MEMS gyroscope 501, the rotator 502, the gyroscope interface 505 and the navigation unit 506 together form a navigation system 507, which navigation system 507 can be integrated into a single device and used on mining and civil engineering vehicles and other applications with or without access to GPS signals.

MEMS gyroscope

Following the introduction of the MEMS gyroscope above, fig. 6 shows the three-axis MEMS gyroscope 501 in more detail. Gyroscope 501 includes a central portion that holds processing electronics and four proof weights (weights) 602, 603, 604, and 605. The weight elements 602, 603, 604 and 605 vibrate in the drawing as indicated by arrows. When the vehicle 500 turns and the gyroscope rotates, the checkweights 602, 603, 604, and 605 tend to maintain their vibration planes while the gyroscope 501 itself rotates. As a result, the checkweights 602, 603, 604, and 605 vibrate out of plane, which can be detected by a capacitor (not shown). The capacitance measurement represents a rotation rate, which can be measured in degrees/second, radians/second, or hertz.

In some embodiments, the MEMS sensors referenced herein may be single axis systems, each individually mounted on an electronic circuit board with supporting electronics. Three separate circuit boards are mounted on a common chassis such that any circuit board sense axis is orthogonal to the other two circuit board sense axes. This forms a 3-axis cluster of single-axis sensors. Single axis MEMS sensors typically have higher performance than three axis MEMS sensors and can approach similar bias instability performance of FOG.

Typically, almost every gyroscope has a bias, which is an additional error in the reported rotation rate. The offset may be different for different axes and may change over time (e.g., as the gyroscope changes temperature) or over lifetime.

One of the main factors limiting the performance of MEMS gyroscopes is gyroscope bias instability, which refers to the ultimate stability of the bias when all other effects (such as drift and white noise) are removed. This indicates the degree of stability of the deviation, and the speed of the expected change. In an otherwise controlled environment, it is expected that a gyroscope with a bias instability of 0.05 degrees/hour will not change faster than this rate. For comparison, the medium to high FOG bias instability can be 0.01 degrees per hour, while the medium MEMS sensor can be 0.2 degrees per hour or higher.

Another limiting factor is bias repeatability, which describes the repeatability of the bias under numerous control factors (e.g., turn-on repeatability, warm-up, temperature, and aging). Bias instability can limit the variation of the bias without these factors, but in other cases this is often overwhelmed by repetitive factors and therefore represents a functional limitation. For MEMS with offset repeatability of no more than 20 degrees per hour, slight offset instability may not ensure pointing accuracy unless the repeatability is compensated for by measurement or control. For reference, low-grade FOG with a very small fiber ring area exhibits a deviation repeatability of the order of 2 degrees per hour, and medium-grade to high-grade FOG exhibits a deviation repeatability of 0.01 degrees per hour.

Determining third axis deviation

To solve the problem of misalignment, the gyroscope 501 may be rotated about a rotation axis 503 by means of the rotator 502. As described above, multiple rotational orientations about the rotational axis 503 allow for determining the offset with respect to the first and second axes perpendicular to each other and with respect to the rotational axis 503 parallel to the z-axis in fig. 3. Embodiments provided herein refer to the x-axis as the "first axis," the y-axis as the "second axis," and the z-axis as the "third axis. It should be noted, however, that these three axes may be permuted or rearranged.

A constant rate of rotation (e.g., the rate of rotation of the earth when the mining vehicle 500 is stopped) may be detected at a zero degree rotational orientation (or index position) about the axis of rotation 503, and then again at a 180 degree rotational orientation (or index position) about the axis of rotation 503. For these two rotational orientations, the difference between the two measured rotation rates on the first axis is the deviation of the first axis. Similarly, for two rotational orientations, the difference between the two rotational rates measured on the second axis is the deviation of the second axis. More specifically, the measurement of the first axis when defined perpendicular to the north direction 205 is given:

ω(0)＝ω_h cosα+b，

ω(180)＝-ω_h cosα+b，

wherein, ω is_hIs the rate of rotation of the earth at a given latitude, α is the azimuth angle (heading) of the first axis relative to north, and b is the gyroscope bias (zero offset). Summing the two terms results in a deviation of ω (0) + ω (180) ═ 2 b. While this process may provide for deviations in two axes, it is difficult to determine deviations in the third axis.

It is noted that, in general, the rate of rotation of the earth ω_eCan be written as the magnitude of the vector componentWherein ω is_x、ω_y、ω_zThe vector components of the earth rotation rate vector measured along the gyroscope x, y and z axes, respectively. Thus, the rate of rotation about the axis of rotation (the z-axis in this example) can be written as:the same may apply for the deviations, which means that the above-mentioned ω is used_zThe equation of (a) may calculate a rotation rate bias about the rotation axis based on the bias about the x and y axes, respectively, and the earth rotation rate. In one embodiment, ω in the above equation_xAnd ω_yIndicating corrected rotation measurement data after removing the calculated offset. The calculated rate of rotation about the axis of rotation ω may then be subtracted from the measured rate of rotation about the axis of rotation_zTo calculate the yaw rate deviation about the rotational axis.

Navigation unit 506 may then receive the rotation rate deviation and calculate the attitude of vehicle 500 by subtracting the deviation from the rate measurement, and calculate the attitude using the corrected rate measurement. This may involve reading acceleration sensors to determine locally measured gravity vectors, i.e. attitude components related to pitch (251) and roll (252) and not to yaw (253). Navigation unit 506 may then calculate a Direction Cosine Matrix (DCM) that mathematically embodies the complete 3-component pose orientation.

The technique allows for the use of lower cost sensors to adequately control a mining vehicle (e.g., a longwall miner) in terms of face alignment or to control autonomous navigation of other vehicles. In this mode, the vehicle uses a "dead reckoning" (DeadReckoned) solution that approximates the true path by employing multiple linear subsamples of orientation and distance traveled. For example, projecting the current heading in 0.5m increments will result in a profile that is substantially the same as the true path taken.

For low accuracy IMU/INS, the calculated heading drifts significantly during operation of the mine vehicle and may change significantly during any power cycles (restarts) or long standing still of the mine vehicle. The present disclosure introduces a number of methods to correct these heading changes, thereby improving the dead reckoning solution.

Bias estimation and tracking

The rotation mark can correct the misalignment error on two of the three axes using the rotation mark explained above. This improves the accuracy of the sensor by an order of magnitude or more. These two measurements may provide enough information to perform the azimuth calculation at the ideal position, but there are a number of exceptions. The offset estimation and tracking of the third (rotating marker) axis enables standard inertial navigation equations to be used on sensors that are typically too noisy to perform such calculations.

Recalibration

During standstill, the vehicle may more accurately calculate the deviation on the various sensors and thus calculate a more accurate heading estimate. During the period after stopping motion, the vehicle may compare the calculated accurate heading with the heading before standing still. If there is a difference, the vehicle may transmit the difference back to the contour generated at this point.

Discontinuous segments

After a power outage or a long period of inactivity, the computed heading may be significantly different from the heading determined at the end of the movement. In this case, the vehicle may apply a constant offset of the difference to the previous motion period or the current motion segment. The absolute profile will be corrected using a capture process (snapping process).

Zero speed update

It should be noted that the disclosed north-seeking solution may be combined with other techniques (e.g., zero-velocity update) to increase the accuracy of inertial navigation. To this end, the underground mining vehicle 500 also includes a vehicle controller, which may be integrated with other components described herein. The vehicle controller is configured to stop the vehicle and correct the inertial navigation system while the vehicle is stopped based on learning that the vehicle speed is zero. The vehicle controller also causes the rotator to rotate the three-axis gyroscope while the vehicle is stopped and causes the gyroscope interface 505 to calculate first, second, and third rotational rate biases. Finally, the vehicle controller resumes movement of the vehicle based on corrections obtained from the zero speed update and the calculated yaw rate bias.

Offset tracking filter

It should also be noted that vehicle 500 may also include a filter connected to gyroscope interface 505 and configured to continuously track the third rotation rate bias. The filter may compare the current value of the third deviation with the previous value to ensure that there is no abrupt jump. In particular, when passing through an ill-conditioned pose, the tracking filter may ensure that the deviation remains within a reasonable range band while the pose is in an ill-conditioned condition. In one embodiment, the tracking filter disables the bias calculation and infers the bias until the vehicle exits the ill-posed angle.

To solve the ambiguity

In some cases, there may be ambiguity arising from the inference of a deviation in the very flat z-axis (third axis). More specifically, the inference of the deviation in the z-axis (third axis) of the extraordinary flat frame is based on And (6) solving. This may provide two roots of the equation, and thus two candidate solutions, which represent ambiguity.

The gyroscope interface 505 may remove an initial estimate of z-axis (third axis) bias from the measured z-axis (third axis), for example, during a calibration process. The gyroscope interface 505 then selects the candidate solution closest to this estimate as the correct solution, providing maximum accuracy since this solution is determined from the exact xy-axis (first and second axis) deviations measured as described above.

Further, with this initial accurate measurement of z-axis (third axis) bias, gyroscope interface 505 can thereafter track changes in value through sufficiently regular pose estimation as described herein.

Second MEMS gyroscope

In one embodiment, the mining machine 500 includes a second three-axis MEMS gyroscope mounted in a fixed position and attitude relative to the vehicle 500. In this case, the navigation unit 506 is connected to the second three-axis MEMS gyroscope and is configured to receive the second rotation measurement data from the second three-axis MEMS gyroscope. The navigation unit 506 may then calculate the pose of the underground mining vehicle based on the second rotation measurement data. This enables the functionality to be divided between the first and second triaxial MEMS gyroscopes, i.e. the first triaxial MEMS gyroscope produces a bias estimate involving the use of the rotation marker as previously described, and the second triaxial MEMS gyroscope allows continuous monitoring of the motion of the mining machine 500 and compensation for errors by the bias estimate from the first triaxial MEMS gyroscope. The bias estimates from the first three-axis MEMS gyroscope occur at the appropriate time when the miner 500 is stationary, however the miner 500 is not constrained to be stationary in order to avoid corrupting the bias estimates because the second three-axis MEMS gyroscope provides continuous attitude tracking. The roles of these divisions may be reversed. The additional data also introduces redundancy that can be used to minimize errors.

Method for computing gestures

Fig. 7 illustrates a method 700 for calculating a pose of an underground mining vehicle 500. The method may be performed by a plurality of units in cooperation, such that each unit performs some of the steps of the method. In other embodiments, the method is performed by a single controller, such as a microprocessor programmed to perform method 700.

Method 700 begins by rotating 701 a tri-axial MEMS gyroscope to a plurality of different orientations about an axis of rotation. Again, this may be performed by the microprocessor sending a signal to the rotator (e.g., a signal to energize a stepper motor). The processor then receives 702 rotation measurement data from the tri-axis MEMS gyroscope with respect to a plurality of different orientations. This allows the processor to calculate 703 a first rotation rate bias of the tri-axis MEMS gyroscope relative to a first axis different from the axis of rotation based on the rotation measurement data and calculate 704 a second rotation rate bias of the tri-axis MEMS gyroscope relative to a second axis different from the first axis and the axis of rotation based on the rotation measurement data.

Further, the processor may now use the calculated first and second rotation rate deviations to correct the rotation measurement data. Using the calibration data, the processor calculates 705 the rate of rotation about the axis of rotation. Importantly, the processor then calculates 706 a third rotation rate bias of the three-axis MEMS gyroscope relative to the rotational axis based on the first rotation rate bias, the second rotation rate bias rate, and the Earth rotation rate vector as described above. Finally, the processor calculates 707 a pose of the underground mining vehicle based on the earth rotation rate vector, the first rotation rate bias, the second rotation rate bias, and the third rotation rate bias.

As described above, the processor may perform the zero speed update by: stopping the vehicle; calibrating the inertial navigation system while the vehicle is stopped; rotating the three-axis gyroscope while the vehicle is stopped; calculating first, second and third rotational rate deviations; and resumes the movement of the vehicle according to the calculated posture. In one embodiment, stopping the vehicle includes stopping the vehicle for 10 seconds or more and 300 seconds or less before resuming movement of the vehicle. In another embodiment, stopping the vehicle includes stopping the vehicle for more than 50 seconds and less than 200 seconds before resuming movement of the vehicle. In yet another embodiment, stopping the vehicle includes stopping the vehicle for more than 100 seconds and less than 170 seconds before resuming movement of the vehicle. Further, restoring movement of the vehicle may include, for example, restoring movement greater than or equal to zero meters and less than 100 meters. In another embodiment, restoring movement of the vehicle includes restoring movement of less than 500 meters. In another example, restoring movement of the vehicle includes restoring movement of less than 50 meters. In yet another embodiment, restoring movement of the vehicle includes restoring movement of less than 200 meters. In yet another embodiment, restoring movement of the vehicle includes restoring movement of less than 30 meters.

Vehicle control

Once the attitude of the vehicle is available, the vehicle controller adjusts the steering angle to guide the vehicle 500 to a desired location within the underground mine. Further, based on prior calibration of the fixed reference points, the underground mining vehicle 500 may determine an absolute position of the underground mining vehicle based on the pose using an inertial navigation unit. That is, the navigation unit adds the posture vector calibrated by the current velocity to the current position. The mine vehicle 500 may also transmit the current position, attitude and speed to an above-ground control room where the data may be displayed to a remote operator.

Note that the navigation systems and methods disclosed herein may be used for navigation of underground machines and underground mining equipment. This may include mining vehicles, shuttle cars, continuous miners, drilling rigs, rig alignment tools, etc. in the field of mining, civil construction, etc.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the above-described embodiments without departing from the broad general scope of the disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.