Patent Application
20240167845
The present application claims the benefit under 35 U.S.C. § 119 of German Patent Application No. DE 10 2022 212 081.7 filed on Nov. 15, 2022, which is expressly incorporated herein by reference in its entirety.
The present invention relates to a method for adjusting an inertial sensor component by means of a calibration device.
Microelectromechanical sensors, in particular in the form of inertial sensors, are described in the related art in a variety of embodiments. The operating principle of such sensors is usually that an external linear acceleration or rotation applied to the sensor causes inertial forces that cause displacements or deflections of a movable structure or force effects to be compensated for on such a movable structure in the sensor core, via which the applied acceleration or rotation can be measured. In this case, the movable structure can be etched from a single layer or be built from a plurality of layers. Inertial sensor modules—(IMUs—inertial measurement units) consisting of or comprising typically three acceleration and three angular rate sensors along each of the three spatial axes of x, y, and z—are used in a wide variety of applications, such as automotive and consumer applications and military, aerospace and space applications. Typical applications are motion detection and stabilization applications, for example in gaming controllers or photo and video cameras. A further major field is platform stabilization—used in deep drilling, for example—and traditional inertial navigation.
Navigation and platform stabilization typically place high demands on sensor performance, in particular on the precise alignment of the sensors with respect to the coordinate system to be stabilized or the coordinate system of the vehicle to be navigated, and thus on the stability of the sensor offsets. When used in navigation applications, an offset error of a gyroscope, for example, increases to the third power over time.
Up to now, sensor systems used for inertial navigation have often been adjusted in an already soldered (for example, on a printed circuit board or a circuit board) and installed (for example, in a control device) state. The results of the calibration, in particular deviations from the desired behavior found, are typically compensated for by compensation devices such as powerful microcontrollers or external signal processors. In addition, redundancy is applied to the sensors and performance is guaranteed over selection even after calibration of the entire IMU. In particular when using MEMS-based sensors, performance is generated by selection. The adjustment requires that the IMUs are calibrated—typically in all three spatial directions—by applying a defined acceleration or rotation rate.
In this case, in accordance with conventional systems, calibration must be carried out after the components of the IMU—in particular the sensors—have already been installed or soldered on. Otherwise, a sufficiently low installation tolerance of the overall system, related to the fixed-body coordinate system required for navigation, for example of the vehicle, is not present.
In this case, suitable calibration devices can generally measure only individual modules or a small number of modules in parallel. The limiting factor is the maximum feasible installation space that such calibration devices can provide to accommodate IMUs. For applications in larger vehicles, for example, which cannot be calibrated as a whole, sensor systems—which form the IMU—are built into elaborate housings, which are then installed in the larger vehicle in a defined manner using suitable mechanical methods such as stop edges or aligning pins.
Currently, there are no solutions available that allow sensor adjustment—for example in a base at the sensor manufacturer—such that corresponding tolerances are so small that automotive applications for automated driving level 5 would be possible without a renewed adjustment in the installed state. The reason for this is in particular that the tolerances on the alignment of the sensors in the measuring base are not sufficiently small and the installation position differs between mounted measurement and, for example, an insert soldered on a printed circuit board.
Against this background, it is an object of the present invention to provide a method for adjusting an inertial sensor component by means of a calibration device, with which it is possible achieve to very small tolerances or errors of alignment of the inertial sensor component inside its housing relative to external dimensions of the housing. In particular, it is possible according to the present invention to specify a method that allows MEMS-based sensors (or inertial sensor components) that are particularly well suited for automotive applications to be adjusted already at the manufacturer (of the inertial sensor components) and tolerances to be reduced to such an extent that no further time-consuming subsequent adjustment is required. According to the method according to the present invention, a novel measuring base (or base device) designed for this application is used. In this way, cost and performance targets for automotive applications, for example for automated driving, can be achieved. In accordance with an example embodiment of the present invention, the method achieves such low tolerances with regard to the alignment of the inertial sensor component that no further time-consuming subsequent adjustment is required and, in particular, tolerances for applications of automated driving at level 5 are enabled. In particular, it is provided in this connection according to the present invention that, for example with respect to a rotation about the sensor Z-axis (i.e., the “vertical” axis of the sensor coordinate system), an alignment tolerance of less than 1.5° is realizable, in particular of less than 1° and particularly preferably of less than 0.8° (i.e., of less than 1.396×10−2 rad).
According to the present invention, the following advantages arise compared to conventional calibration methods: Possible errors after installation of the module are minimized (since such errors can be detected earlier). There is no need for extensive measurements in the (finished) (control) device (“IMU”, DCU, etc.) or even in the vehicle. The main load of the adjustment is on a small mechanical volume, since only the sensor (or the inertial sensor component and the base (or the base device), and not the IMU/DCU, have to be mechanically stimulated; this also means that a parallelizability of the measurement (or a higher degree thereof) is possible. Furthermore, performance-relevant failures are detected early in the process chain, specifically at the sensor level or the inertial sensor component level. This also improves sustainability, because less resource loss follows from a faulty sensor (only one sensor vs. entire device or only one sensor 6DOF (6 degrees of freedom) vs. a plurality of sensors). Furthermore, according to the present invention, a cost framework sufficient for mass production is achieved; otherwise, a sufficiently accurate calibration would be many times more expensive.
According to the present invention, a calibration device is used to adjust the inertial sensor component. The inertial sensor component comprises at least one micromechanical sensor element and a housing that is generally wider and deeper than high, i.e., has a main extension plane and an upper side extending substantially in parallel therewith and a lower side and with side surfaces extending substantially perpendicularly to the main extension plane in each case. The housing surrounds or encloses the micromechanical sensor element. According to the present invention, the housing is in particular a ceramic housing. To describe the geometric relationships, reference is also made below to the main extension plane of the housing of the inertial sensor component. The direction perpendicular to the main extension plane is also referred to as the vertical direction, without a reference to the direction of gravity intended to be implied. The novel base concept (or base device) according to the present invention is based on the fact that, by means of a suitable mechanism, the sensor housing is positioned at a defined stop edge before the sensor (or housing) is fixed in the base (or base device). This achieves a high positional accuracy of the housing to the base.
In particular, according to an example embodiment of the present invention, the sensor housing is positioned at the defined stop edge by a mechanical fixing device, i.e., the housing of the inertial sensor component is positioned and/or aligned relative to the base device in particular by means of a fixing element that is movable relative to the base device in a fixing direction, wherein by means of the fixing element a part of a particular side surface of the housing of the inertial sensor component, which part extends perpendicularly to the fixing direction, is pressed against the base device (or its stop edge). In this case, in particular in the course of the first step according to the present invention, a first positioning element of the housing of the inertial sensor component, which element is located on the particular side surface of the housing, is pressed against a second positioning element of the base device by means of the fixing element or due to the movement of the fixing element. The edge (or the side surfaces of the housing, in particular the particular side surface) of the ceramic housing is usually characterized by various zones: A first zone has a low roughness; it is defined by the manufacturing process, which generates a predetermined break edge in this region. A second zone has a higher roughness, but is many times higher than the first zone; this corresponds to the breaking region of the ceramic (i.e., the breaking region of the ceramic housing, in particular when carrying out a singulation operation during the manufacture of the inertial sensor component or its housing). A third zone joins the (particular side surface of the housing to the) upper side of the housing; this in turn serves to define the predetermined breaking point (of the ceramic housing) and again has a lower roughness. Thus, the particular side surface of the housing has a first snapline region (lower roughness, first zone) extending substantially along the lower side of the housing and generally also a second snapline region (lower roughness, third zone) extending substantially along the upper side of the housing and a break edge region therebetween extending substantially in the central region of the particular side surface between the upper side and the lower side. Typically, the height extension of the first snapline region (first zone) is approximately 10% to 20% of the (“vertical”) height extension of the housing, the height extension of the second snapline region (third zone) is approximately 10% to 20% of the height extension of the housing, and the height extension of the break edge region (second zone) is in the range of 60% to 80% of the height extension of the housing. For example, with an exemplary overall thickness (or height extension of the housing) of 1.75 mm, the first snapline region (first zone) could have a height extension of approximately 0.5 mm (longer smooth zone, for example, bottom), the second snapline region (third zone) could have a height extension of approximately 0.2 mm (shorter smooth zone, for example, top) and the break edge region (second zone) could have a height extension of 1.05 mm.
Positioning along the sensor edge (i.e., the particular side surface of the housing) ensures a definition of the alignment in the (“horizontal”) x-y plane of the sensor. Positioning in the (“vertical”) z-direction is performed by a stop surface on which the bottom of the sensor housing is placed.
According to an example embodiment of the present invention, the positioning accuracy in the base or in the base device is achieved by optimally matching the height of the stop edge (i.e., the height of the second positioning element (of the base device)) to the height and roughness of the corresponding regions (i.e., the first positioning element (of the housing)) of the break edge (or the particular side surface) of the housing. In this case, the first positioning element is not necessarily (this is, however, a possibility according to the present invention) a particular structure in the particular side surface of the housing, but possibly only that region (or those regions in the case of a plurality of first or second positioning elements) of the particular side surface of the housing that corresponds to the second positioning element (or a plurality thereof) of the base device (or the stop edge).
According to a preferred embodiment of the positioning of the housing relative to the base device of the present invention, the first positioning element has a comparatively low height compared to the height extension of the housing (between its lower and upper sides) and is arranged exclusively in the first snapline region or exclusively in the second snapline region. According to this embodiment, a stop edge at exactly the height of the first, smooth sensor zone is thus used as the second positioning element of the base device. According to the present invention, this achieves the highest positional accuracy. According to a further variant of this embodiment, the housing is positioned in the base device rotated by 180° (about an axis in the x-y plane), such that the stop edge of the base device (or the second positioning element) contacts the housing along the upper zone three of the housing; thus, in this embodiment, the z-positioning of the housing is achieved by placing the housing cover (or the upper side of the housing) on the stop surface of the base device.
According to another preferred embodiment of the positioning of the housing relative to the base device according to the present invention, the first positioning element has a comparatively large height compared to the height extension of the housing (between its lower and upper sides) and is arranged either exclusively in the break edge region or both in the break edge region and in the first snapline region or both in the break edge region and in the second snapline region. According to this embodiment, a stop edge is thus used as the second positioning element of the base device up to zone two of the break edge. Here, the key is careful matching of the height of the edge, such that averaging is achieved across the surface. This also leads to a high overall accuracy due to averaging.
According to the an example embodiment of the present invention, the aim of the adjustment operation or the calibration measurement of the inertial sensor component is a sufficiently low effective torsion of the MEMS fixed coordinate system (i.e., the coordinate system of the sensor element), in which the inertial measured quantities of angular rate and acceleration are determined along the x-, y- and z-axes, relative to an installation of the sensor in an IMU/DCU and a vehicle coordinate system firmly connected thereto. Upon installation, tolerances with regard to the vehicle coordinate system and torsion of the IMU/DCU housing and sensor printed circuit board relative to this coordinate system can be kept sufficiently low by mechanical design elements. The position reference of the sensor (i.e., the housing of the inertial sensor component) is its contacting pads. Upon a soldering operation of the sensor onto a printed circuit board, the component centers itself along the coordinate system of the contacting pads. The assembly and connection technology used (AVT) ensures a maximum torsion of the pads relative to the housing edges of the ceramic or the housing, which is sufficiently small in relation to the required overall tolerance. However, the largest unknown in the tolerance chain remains the torsion of the micromechanical sensor element relative to the sensor ceramic or the coordinate system of the contacting pads. This torsion and possibly also structure-related distortions of the MEMS coordinate system must be compensated for by means of calibration. If the position of the sensor (i.e., the housing of the inertial sensor component) in the calibration device is known, taking into account the complete tolerance chain—i.e., starting from the machine (i.e., the calibration device) to the measuring fixture of the same to the pad (device) and to the sensor ceramic (or housing)—misalignments between pad coordinate system and MEMS coordinate system can be determined. Therefore, according to the present invention, the individual torsion of the measuring base used is determined in relation to the workpiece carrier or the measuring fixture by optical measurement; tolerances determined in this way can be taken into account in the adjustment. Available optical measuring methods can be optimized by suitable adaptation to the specific application in such a way that sufficiently accurate detection of such tolerances is possible. Furthermore, according to the present invention, the exact positioning of the measuring fixture (or the base device) in the machine (i.e., the calibration device) is determined. This is achieved by mechanical design elements such as aligning pins and aligning edges, analogous to the procedure for the housing of an IMU/DCU, i.e., the control device housing. Experience shows that the largest alignment error is caused by a possible torsion of the sensor ceramic (i.e., the housing) in the base device, since the external tolerances of the ceramic or the housing are large in relation to the required tolerances due to the process, and an alignment of the ceramic or the housing with respect to the pad coordinate system can thus also have large errors. According to an example embodiment of the present invention, minimization of the possible torsion is made possible in particular by a mechanical pusher or the fixing device, which presses the sensor (or the housing) against a reference edge (of the base device).
Further objects of the present invention are a calibration device or a base device for carrying out the method according to the present invention or an inertial sensor component that is adjusted according to the method according to the present invention.
Alternatively or in addition to the example embodiments of the calibration device according to the present invention or of the base device according to the present invention or of the inertial sensor component according to the present invention described above, the advantageous embodiments and features explained in connection with the method can also be applied to these alone or in combination.
In
In
Since the housing 20 provided in accordance with the present invention is in particular a ceramic housing, the housing edge (or the particular side surface 24) has surfaces with low roughness on the lower side of the ceramic in the region of the so-called snapline (first snapline 24′), an irregularly shaped break edge (the break edge region 24″), which is produced by the breaking apart of the ceramic components, and surfaces with low roughness on the upper side in the region of a further scribe mark for defining the predetermined breaking points (second snapline 24″). The design of the stop edge, i.e., the second positioning elements 35, is decisive for the positioning accuracy, while ensuring the process reliability of the positioning operation.
With the exemplary embodiments according to
With the second variant, the height of the stop edge (or the second positioning elements 35) is designed such that the break region of the sensor edge (i.e., the break edge region 24″) is also used. This allows sufficient positioning accuracy to be achieved, although this is lower than with the first variant. On the other hand, process reliability is significantly improved, since the stop edge is more robust with regard to wear and possible incorrect positioning due to the “slipping” of the ceramic (or the housing 20) from the stop edge due to the larger stop surface. The position and length of the stop edges are designed to circumvent known inaccuracies of the sensor edge, for example, breakouts at corners of the sensor and in the region of the typically four indentations on the sensor ceramic. Overall, this variant is to be considered the optimal variant. All variants have in common that the stop edge is not executed across the entire length of the side edge (i.e., the particular side surface 24) of the sensor ceramic, but only in defined regions. Optimization of the position and effective stop surface is then crucial for positional accuracy and process reliability of the calibration method.
When adjusting in a 2-axis measuring system, the sensor can be stimulated in each of its 3 axes at least 2 support points without re-assembling. This allows the sensitivity vectors seen by the sensor to be measured in magnitude and spatial orientation. Additional tolerances due to re-assembling are eliminated. Equipment manufacturers for such 2-axis measuring systems can guarantee sufficient accuracy of axis positioning and movement. Thus, in the case of the calibration method according to the present invention, using a 2-axis measuring system is proposed.
Overall, tolerances of the entire calibration device can thus be minimized according to the present invention to such an extent that the remaining unknown torsion of the MEMS element relative to the contacting pads can be determined.
According to the present invention, the two components that are matched to one another—the housing 20 or sensor ceramic and measuring base or the base device 30—create the basic prerequisite for achieving a tolerance chain of low tolerance from the actual MEMS-based sensor element 15 via the ceramic housing 20 to the printed circuit board and the control device. Furthermore, the present invention comprises calibrating the calibration device (or the measuring equipment), in particular the alignment of the measuring base with respect to the measuring device, by means of optical methods. Overall, a highly accurate calibration is thus already achieved at the sensor or housing level. This achieves accuracy for applications in the field of HAD, highly automated driving, in particular level 5 applications, without additional adjustment and compensation mechanisms in the downstream control device.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2022 212 081.7 | Nov 2022 | DE | national |
