The present invention relates to a MEMS or NEMS inertial unit with several axes intended to detect angular velocities and/or accelerations along several axes.
Inertial units exist with five detection axes, three axes in acceleration and two axes in rotation. Such a unit is described in the document Y. Watanabe et al., “Five axis motion sensor with electrostatic drive and capacitive detection fabricated with silicon bulk micromachining”, Sensors and Actuators A, Vol. 97-98, p 109-115, 2002. On the one hand, this unit is made using volume technology, which makes its manufacture long and costly. On the other hand, the detection of all the detection and excitation axes are entangled. The processing of the data is then complex and a high risk of introducing errors exists, said risk being higher the more the capacitive detection by variation of air gap is non-linear, this non-linearity being amplified when the movements are of high amplitude, which is the case of excitation movements. Moreover, the measurement precision is higher the lower the capacitive air gap, yet the actuation requires important air gaps to obtain significant amplitudes.
The document 20100132463 describes an inertial unit formed of three integrated gyrometers, comprising three pairs of inertial masses and an oscillating disc that makes it possible to have two directions of excitation movement in the plane. This unit has the advantage compared to a unit comprising three separate gyrometers of having simpler electronics, in particular it only requires a single excitation control loop. Two pairs of masses are dedicated to the detection of the Coriolis forces around axes contained in the plane of the unit and a pair of masses is dedicated to the detection of the Coriolis forces around the axis normal to the plane. However these pairs of masses are also sensitive to accelerations, which perturbs the measurement of the Coriolis forces. Moreover, this unit does not enable the measurement of acceleration.
It is consequently an aim of the present invention to offer an inertial unit capable of detecting rotations and/or accelerations along at least four axes with simplified electronics in excitation and in reading and in which the detection of rotation and the detection of acceleration are relatively independent of each other.
The aforementioned aim is attained by an inertial unit comprising at least four inertial sensors, two sensors being aligned on a first axis and two sensors being aligned on a second axis.
Each sensor comprises a pair of inertial masses, the masses of each pair being connected together by a mechanical coupling and the sensors are mechanically coupled together. Advantageously, the mechanical coupling between the masses of each pair is a strong coupling.
The sensors may be gyrometers or dynamic accelerometers. Due to the implementation of two inertial masses in each sensor, they are relatively insensitive to rotation in the case of an accelerometer and relatively insensitive to acceleration in the case of a gyrometer. Thus, the processing of the detection signals is simplified, it is not necessary to separate the signals to obtain the acceleration and/or rotation velocity values. Moreover, by mechanically coupling the sensors together, a single excitation control loop is required. The control electronics of the unit is thus simplified and the precision of the measures made is good.
In other words, an inertial unit is formed with at least four inertial sensors mechanically coupled so as to transmit the excitation to the four sensors; the sensors may be accelerometers chosen so as to be relatively insensitive to Coriolis forces and/or gyrometers, these being chosen so as to be relatively insensitive to accelerations.
Thanks to the invention, an inertial unit may be formed with six detection axes capable of measuring the accelerations in the three axes of the space and the Coriolis forces around three axes of the space in a quasi-independent manner, with a reduced surface compared to a structure with six independent sensors.
Advantageously, the coupling means are formed so as to have a centre of symmetry and the sensors are chosen so as to have a similar structure and operation. Moreover, their relative arrangement is chosen preferably to offer a substantially symmetrical inertial unit.
The subject matter of the present invention is then an inertial unit for the measurement of accelerations and/or rotations comprising at least four sensors, distributed by pairs on at least two intersecting axes, each sensor comprising two inertial masses, each inertial mass comprising a moveable excitation part and a moveable detection part, the moveable excitation parts of a sensor being separate from those of the other sensors, said unit also comprising means for detecting the excitation parts of the sensors and means for detecting the movement of the detection parts, and a control unit of the excitation means and detection means, the two inertial masses of each sensor being coupled mechanically by coupling means, called intra-sensor coupling means, said unit also comprising mechanical coupling means between the four sensors, called inner-sensor coupling means, said inner-sensor coupling means being elastically deformable and able to transmit the excitation vibrations from one axis to the other. At least one of the sensors is an accelerometer.
The intra-sensor coupling means and the inner-sensor coupling means are separate.
The excitation means may be shared for all of the sensors or independent from one sensor to the other.
Said intra-sensor coupling means are in a preferential manner strong coupling means.
Said inner-sensor coupling means can be situated at the intersection of axes bearing the two pairs of sensors, the two sensors of each pair being arranged on either side of said inner-sensor coupling means.
Preferably, the two axes are perpendicular.
Advantageously, the inner-sensor coupling means are strong coupling means.
The inner-sensor coupling means are for example connected to the excitation part of one of the masses of each sensor.
The inner-sensor coupling means may comprise an assembly of beams comprising at least one closed pattern.
In an embodiment example, the closed pattern has the shape of a diamond, each top of said diamond being connected to a mass of a sensor.
The intra-sensor coupling means connect for example the two excitation parts of each sensor.
Preferably, the excitation part of each sensor is excitable in a single direction.
The inertial unit may comprise one or more gyrometers and/or one or more accelerometers, said at least gyrometer and/or accelerometer being able to measure a rotation and/or an acceleration along an axis called detection axis.
The inertial unit may comprise a rigid rod in rotation around an axis orthogonal to the detection axis, said rigid rod connecting the two detection parts of the masses of the accelerometer.
Preferably, the intra-sensor coupling means between two masses of a gyrometer form a strong coupling in antiphase, and the intra-sensor coupling means between two masses of an accelerometer form a strong coupling in phase.
In an embodiment example, the intra-sensor coupling means in antiphase comprise a closed assembly of beams.
For example, the intra-sensor coupling means in phase comprise a rigid connection, the two masses of a same accelerometer acting as a single mass.
In an embodiment example, the gyrometer(s) comprise a rigid rod in rotation around the detection axis around which the rotation is to be detected connecting the detection parts of the inertial masses.
At least one of the sensors may be replaced by an object not supplying a measurement on the acceleration or the rotation and ensuring a balancing of the inertial unit.
The inertial unit according to the present invention may comprise three pairs of sensors, the sensors of each pair being aligned on an axis, two of the axes being parallel and the other axis being intersecting to the two parallel axes, first inner-sensor coupling means being provided at the level of each intersection of one of the axes parallel with the other axis and second inner-sensor coupling means being provided between the first inner-sensor coupling means.
In one embodiment example, the inertial unit comprises three gyrometers and three accelerometers, the three gyrometers being connected directly to first inner-sensor coupling means different to the first inner-sensor coupling means to which are connected directly the three accelerometers.
In another embodiment example, the inertial unit comprises eight sensors, the inner-sensor coupling means comprising an assembly of eight patterns of juxtaposed diamond shape, the patterns being connected by one of their tops in a point constituting a centre of symmetry of the coupling structure, so as to have opposite patterns and aligned two by two, the tops of the patterns opposite to the centre of symmetry each being intended to be integral with a mass of eight sensors.
The excitation means are for example of electrostatic, piezoelectric or magnetic type and the detection means are of capacitive, piezoelectric, magnetic or piezoresistive type.
In an advantageous manner, the detection means are chosen of same type for all of the sensors.
The control unit may advantageously comprise a unique excitation control loop for all of the sensors and means for detecting the displacement of the detection parts.
The control unit may comprise means for reading the detection means of the different sensors by time division multiplexing.
The present invention will be better understood by means of the description given hereafter and the appended drawings in which:
In
The axes X, Y and Z are defined as follows:
The inertial unit 2 comprises four sensors 4a, 4b, 4c and 4d distributed two by two on the two perpendicular axes X and Y.
A unit in which the sensors are distributed on intersecting but non perpendicular axes does not go beyond the scope of the present invention.
The sensors 4a to 4d may be gyrometers or accelerometers.
The unit 2 comprises mechanical coupling means 6 between the four sensors 4a to 4d so as to link in movement the masses of all of the sensors.
In the example represented, the mechanical coupling means 6 are formed by a structure of four beams 8 forming a diamond, each top 10a to 10d of which is connected to a sensor 4a to 4d. Each of the diagonals of the diamond is aligned with one of the axes X and Y of the inertial unit.
In a variant, the coupling means could form a square or an ellipse, in the latter case the large axis would be aligned on the axis X or Y, and the small axis would be aligned on the axis Y or X.
Thus, a vibratory movement generated by an excitation signal at one of the tops or at two tops opposite one of the axes X, Y is transmitted to two other tops situated on the other axis Y, X.
In the remainder of the description, the expressions “strong coupling” and “weak coupling” will be used to qualify the mechanical couplings between the sensors, and between the masses of a same sensor.
In the present application, “strong coupling” between two moveable masses is taken to mean a mechanical connection such that if a first mass of an amplitude A0 is displaced, leading to a displacement of the second mass by an amplitude A1, then the second mass is displaced by an amplitude A1 leading to a movement of an amplitude A2 on the first mass, the amplitudes A0 and A2 are substantially the same.
“Weak coupling” between two moveable masses is taken to mean a mechanical connection such that A2 is less than A0.
Examples of sensors 4a to 4d that can be implemented in the inertial unit will now be described.
In
The gyrometer comprises two inertial masses 12, 14 moveable with respect to a substrate.
The mass 12 comprises two parts 16, 18 moveable with respect to each other. In the example represented, the first part 16 is box shaped and surrounds on three sides the second part 18.
The box 16 is suspended from the substrate and the second part is arranged inside the box 16 and is suspended from it by beams 20.
The first moveable part 16 is anchored on the substrate for example at the level of the points A1, A2.
Elastic suspension means 22 are provided between the anchoring points A1, A2 and the first part 16 enabling it to vibrate along the axis X.
The sensor comprises means 24 for exciting the first moveable part. In the example represented, the excitation means 24 are provided laterally. Said excitation means may be of the electrostatic type with interdigitated combs, piezoelectric, magnetic means, etc.
Advantageously, the first moveable excitation part 16 has a single degree of freedom in displacement along the excitation axis, along the axis X in the example represented. This moveable part 16 is connected to the coupling means. This only moves advantageously in excitation, which makes it possible to only couple the excitations and not to have detection couplings which would induce perturbations between sensors.
The coupling means are fixed to the excitation part.
The second moveable part 18 is, for its part, mounted articulated around a pivot 26 of axis parallel to the axis Z orthogonal to the plane of the sensor. In the example represented, the second moveable part 18 is connected to the pivot 26, this liaison will be described later. Preferably, anchoring points A3, A4 are provided in the unit part of the second moveable part 18 to ensure the guiding. Elastic means are then provided between said anchoring points A3, A4 and the second part 18.
Means for detecting (not represented) the displacement of the second part 18 are provided, they are for example of capacitive, piezoelectric, magnetic, piezoresistive type. For example, piezoresistive gauges may be provided on either side of the pivot 26 and to measure the variation in electrical resistance of the gauges and thereby detect the variation in length and deduce thereof the amplitude of displacement of the second parts 18. Preferably, the same types of detection means are provided for all of the sensors in order to simplify the realisation and the processing of the detection signals.
In an alternative manner, the second part may, itself, be formed of several sub-parts. For example it may comprise a first frame connected to the first excitation part, said first frame moves in excitation in an integral manner with the first excitation frame and has a degree of freedom in detection. A second frame, anchored or not according to the excitation mode, sees all the detection movement of the second frame or a part thereof transmitted.
The second mass 14 has substantially the same structure as the first and will not be described in detail.
Moreover the sensor comprises strong coupling means 30 between the two masses 12, 14 along the axis X and connecting the two first parts 16. The strong coupling means 30 are of the type in antiphase, i.e. when the masses are excited, they move apart or closer together in phase. Moreover, the coupling being strong, they have the same amplitude of displacement.
In the example represented, the strong coupling means 30 comprise a closed structure of beams in diamond shape, two opposite tops 30.1, 30.3 being connected to the first parts 16 of each of the masses.
The two other tops 30.2, 30.4 are connected to beams 32, perpendicular to a straight line connecting the tops 30.2 and 30.4 at the level of the central beam area. The beams are anchored on the substrate at their ends. The coupling means 30 deform along the axis X and enable the masses to move apart and closer together along the axis X.
Advantageously, the two second parts are connected to the pivot 26 via a rigid rod 38 in rotation around an axis parallel to the detection axis, which makes it possible to reduce very considerably the sensitivity of the gyrometers to the accelerations.
In
Preferably, an accelerometer is chosen in which the structure and the mode of operation are similar to those of gyrometers. The accelerometer is of the dynamic type and is excited in detection, in a similar manner to a gyrometer. Thus, the same excitation control loop may be applied to all of the sensors and use the same detection means.
The structure of the accelerometer of
In the example represented, the strong coupling means in phase are formed by a rigid rod type connection, connected to each of the first moveable parts 16′, simplifying the manufacture of the sensor.
The coupling means 6 between the sensors are, in the example represented, weak coupling means. These are particularly visible in
For reasons of simplification, in
For example, the gyrometer of
In the remainder of the description, the gyrometers will be designated by Gy and the accelerometers by Acc.
In
In
The coupling means between the sensors are advantageously strong coupling means, ensuring there is a same amplitude at the level of the sensors, even in the case where the unit comprises gyrometers and accelerometers.
According to the invention, the two differential measurement parts of each sensor are situated side by side on the same side of the inner-sensor coupling means, which makes it possible to couple the two detection parts and in particular to use strong coupling means, such as a rigid rod in rotation around a pivot, which makes the sensor very insensitive to parasite signals.
By way of example, if the gyroscope of
The implementation of strong intra-sensor couplings enables the amplitudes of the two masses of each sensor to be equal, the couples exerted for example on the strong coupling rigid rod 38 are then equal, which makes the sensors the most robust possible to parasite signals.
In
The rectangular elements represent the masses. The grey level of the masses illustrates their amplitude of displacement, the higher the grey level, the greater the amplitude. In the case of the structure of the unit of
In the case of the unit of
In a variant, it could be envisaged to replace a sensor by an object having the function of balancing the structure. Said object would have the same mass, the springs would have the same stiffness. However, it would not comprise a detection part, i.e. the excitation and detection parts would be merged and the structure would not comprise a rod 38, or detection means such as electrodes, gauges, etc.
In a variant, the unit may comprise three gyrometers and an accelerometer, two gyrometers and two accelerometers, or one gyrometer and three accelerometers.
In
In
In
In
It may be observed that the masses of the three gyrometers have substantially the same amplitude of displacement, and that the two masses of the accelerometer have the same amplitude of displacement which is different to that of the masses of the gyrometers. This situation is interesting since the masses in each sensor always have the same amplitude of displacement.
The inventors have noted that an inertial structure in which the masses of each sensor were coupled together by a weak coupling did not make it possible to operate with simplified electronics because the masses of each sensor did not move with the same amplitude.
The operation of such a unit will now be explained.
The excitation means of the sensors are activated, for example those of the sensors 4a and 4c are activated, which brings about a vibration of the first moveable parts of the sensors 4a, 4c directly excited and the second moveable parts of the sensors 4a, 4c. In the case of gyrometers, the two masses of sensors 4a, 4c vibrate in antiphase, in the case of accelerometers they vibrate in phase. Moreover, due to the coupling means 6 between the two sensors 4a, 4c aligned on the axis X and the two sensors 4b, 4d aligned on the axis Y, their masses directly connected to the coupling means are made to vibrate, as well as the masses coupled by the strong coupling means. The masses of the sensors 4b, 4d then vibrate in antiphase in the case of gyrometers and in phase in the case of accelerometers.
In a practical manner, the excitation means are arranged in a symmetrical manner on the structure and more particularly at the level of the masses, however it is possible to only excite one mass, i.e. only one excitation part 16, which is sufficient to make the whole assembly move in the desired manner.
Consequently, a single control loop is implemented to excite the assembly of moveable masses of the sensors.
The moving masses are excited by the excitation means acting on the first parts.
Due to the accelerations and rotations or Coriolis forces, the second parts of the moveable masses are brought into movement; in the case of the sensors represented, the moving parts pivot around pivots 26. This displacement is detected by the detection means, for example piezoresistive gauges. Since each sensor has a good robustness, due to their structure with two inertial masses, the detection of the rotations discriminate relatively well the accelerations and conversely, the processing of the signals detected and the detection electronics are simplified. In the case of three or four gyrometers, the Coriolis forces in the three directions of the space are determined, and the same is true for the acceleration in the case of three or four accelerometers.
In the case where the unit comprises two gyrometers and two accelerometers, preferably the two gyrometers are aligned and the two accelerometers are aligned in order to improve the balancing of the structure.
In the case where the coupling means between the sensors are strong coupling means, and whatever the type of sensors, they all have the same amplitude as is represented in
In the case of a weak coupling, the sensors of the same type have the same amplitude. For example in the case where the unit comprises three gyrometers and an accelerometer, the accelerometer does not have the same amplitude of displacement.
However, thanks to the strong coupling between the masses of each sensor, the masses of a same sensor have the same amplitude of displacement, as is represented in
In
The unit comprises six sensors, three gyrometers Gy1, Gy2, Gy3 and three accelerometers Acc1, Acc2, Acc3.
Two gyrometers Gy1 and Gy3 are aligned on an axis Y1, two accelerometers Acc1 and Acc3 are aligned on an axis Y2 parallel to the axis Y1, and a gyrometer Gy2 and an accelerometer Acc2 are aligned on an axis X.
Coupling means 106.1 connect the three gyrometers and coupling means 106.2 connect the three accelerometers, and coupling means 106.3 connect the three gyrometers and the three accelerometers.
In the example represented, the coupling means 106.1 and 106.2 are strong coupling means having the same structure, i.e. diamond shape. The coupling means 106.3 are also strong coupling means and are formed by a beam or rod type rigid connection.
Other strong coupling structures may be provided between the three gyrometers and the three accelerometers. For example, a structure of beams laid out in diamond shape connecting the two tops facing the diamonds of coupling means 106.1, 106.2 or in ellipse could be provided.
In another embodiment example, the coupling means 106.1, 106.2 and 106.3 may also be weak coupling means. Or alternatively certain coupling means may be strong coupling means and others weak coupling means.
The structure of
In
In
In
The four trapezes are connected to each other by their non parallel sides by means of elastic elements 210.
Each trapezoidal element is, for its part, connected to the sensors 4a to 4d by a longitudinally rigid beam type connection 212.
In
In
In this embodiment example, the coupling means 40 between sensors have a structure formed of beams laid out so as to form eight identical diamonds 410.1 to 410.8 having a shared top forming the centre of symmetry of the structure, and each diamond 410.1 to 410.8 having a side shared with the adjacent diamond. The coupling structure then has the shape of a star with eight branches.
The sensors 4.1 to 4.8 are then integral with the structure at the level of the tops of the branches of the star.
By way of example, the sensors may be spread out as follows:
It may be noted that, in this case, the axes X, Y of the gyrometer and X′, Z′ of the accelerometer are shifted by an angle of 45°.
The sensors described in
The star structure with eight branches is balanced. In a variant, the redundant gyrometer Z and the accelerometer Z′ could be replaced by objects uniquely ensuring the balancing of the structure.
The present invention offers an inertial unit capable of detecting up to 6 axes and offering a high level of integration. It enables an important surface gain compared to a unit composed of sensors simply assembled and without coupling between them. Moreover, it offers a high alignment precision compared to assembled sensors.
In addition, due to the coupling of the sensors and their structure and similar mode of operation, it is possible to manage all of the sensors by a single electronic, with a single excitation loop and a same reading electronic, this both for the accelerometers and the gyrometers.
As regards the detection reading, it is possible to carry out a reading of the different axes by time division multiplexing, i.e. by carrying out a reading of the detection means of the different sensors one after the other at a frequency substantially higher than the frequency of the excitation movement.
Moreover, the “cross axis” sensitivities are reduced and they can be easily optimised thanks to the use of separate structures dedicated to the detection of each axis, these structures each having two moveable masses.
Preferably, the structure of the unit is balanced by arranging an even number of structures of sensors such that the structures facing each other have the same coupling (either in phase or in phase opposition) and the same mass actuated in excitation, for example an accelerometer is aligned with an accelerometer and a gyrometer is aligned with a gyrometer as in the case of the inertial unit of
An example of method of making an inertial unit according to the present invention, with electrostatic excitation and with capacitive detection, will now be described.
In
For example, an SOI substrate is used, represented in
During a first step represented in
During a following step represented in
A deep etching of the layer I is then carried out with stoppage on the buried oxide II.
The resin remaining on the layer of silicon I is eliminated.
A release of the moveable masses and coupling means IV is then carried out by elimination of the oxide of the layer II situated underneath the etched silicon, for example with hydrofluoric acid vapour. The released structure is shown schematically in
This method is given uniquely by way of example, and is in no way limiting.
Number | Date | Country | Kind |
---|---|---|---|
11 53731 | May 2011 | FR | national |