Quadrotors are widely used for surveillance, mapping, and deliveries. In several scenarios the quadrotor operates in pure inertial navigation mode resulting in a navigation solution drift. To handle such situations and bind the navigation drift, the quadrotor dead reckoning (QDR) approach requires flying the quadrotor in a periodic trajectory. Then, using model or learning based approaches the quadrotor position vector can be estimated. We propose to use multiple inertial measurement units (MIMU) to improve the positioning accuracy of the QDR approach. Several methods to utilize MIMU data in a deep learning framework are derived and evaluated. Field experiments were conducted to validate the proposed approach and show its benefits.
For the reliable control of unmanned aerial and underwater vehicles, a robust attitude solution is needed in order to control the vehicle correctly along its three axes. Emerging sensor faults endanger the attitude estimates, and thus the control of the vehicle, and must be detected as early as possible in order to counteract and mitigate, or possibly even prevent, any negative effects. A common approach to detect erroneous sensor data is the use of sensor redundancies in which measurements of three or more sensors are compared. However, this concept is not applicable in all cases, especially with regard to unmanned vehicles that are often limited by size, weight, cost and other constraints. This paper describes an algorithm, developed to detect faults and degradation in navigation sensor data for an Autonomous Underwater Vehicle (AUV) with a limited navigation sensor suit, i.e. without sensor redundancies. The algorithm aims to calculate an integrity value for each sensor channel, based on the sensor faults or degradations detected. On the one hand, the calculated integrity data is meant to be used by Bayes filter based navigation algorithms (e.g. Kalman Filter) to adjust their confidence in the measured data used to correct the state predictions. On the other hand, the integrity values are used to monitor the overall health condition of the navigation sensor suit, enabling downstream software modules to come to decisions regarding the continuation of the mission.
In this paper, the influence of phase shift induced by the front-end amplifier and the digital control circuits on the quadrature suppression and the force-to-rebalance control loops is theoretically analyzed. Due to the existence of a phase shift, the output of the resonator gyroscope is affected not only by the damping mismatch, but also by the frequency splitting of the gyroscope which is the main contribution to the nonlinearity of scale factor of a RG. Based on the analytical results and established mathematical model of scale factor, the phase delay was characterized and compensated in the room temperature. Detailed experiments of phase compensation were conducted and results were analyzed. Experimental results show that with phase compensation, the quadrature suppression signal changes far less with the variation of external angular rate than that before phase compensation. This means that the coupling between the quadrature suppression signal and the force-to-rebalance control signal is greatly decreased. In addition, residual errors of the scale factor of the resonator gyroscope were also greatly decreased. It is shown that after compensation of the phase delay the residual error is reduced by a factor of 45 and the linearity of the scale factor of the resonator gyroscope was extremely increased by more than 50 times compared with results before compensation.
Inertial measurement units and navigation systems range from low-end tactical to strategic grade, spanning a very large spectrum of sensor performance. EMCORE makes use of its own quartz MEMS gyros for tactical grade systems, while for higher-end systems we use in-house optical gyroscopes (fiber optic and ring laser gyros). These gyroscopes and systems range from 10 °/hr down to 0.0001 °/hr performance over dynamic environment. These systems require accelerometers with suitable levels of performance for each IMU/INS grade. In a previous DGON ISS 2021 paper we described the quartz-MEMS IMU progress toward navigation grade performance [1]. This improvement was based on use of a modified conventional calibration method over a dynamic thermal environment. Here we report the latest experimental results (consistent with a high-end tactical grade accelerometer) which are much-improved due to a method of self-calibration incorporating multiple resonant modes of the accelerometer structure. The results we present are for the quartz MEMS accelerometer, which demonstrate a velocity random walk of $0.1\ \mu \mathrm{g}/\surd{\text{Hz}}$ , and bias instability of 30 ng. The accelerometer bias and scale factor drift over a full temperature range (-55 °C to +90 °C) were $25\ \mu \mathrm{g}$ and 25 ppm, respectively, after we improved the conventional calibration. Using an innovative advanced operational regime, called multiple-mode operation, a new self-calibration algorithm was developed demonstrating accelerometer bias over a dynamic thermal profile of less than $2\ \mu \mathrm{g}$ , thus, making the accelerometer suitable for high-end navigation grade systems.
This paper addresses the requirements on IMUs for future small and affordable short range missiles. Current developments in this area of missile systems are presented and also some functional requirements on the missiles are explained. From these requirements, the impact on missile design and missile functions are derived and finally the impact on the requirements on the used inertial sensor systems are presented. At the end, we will have to conclude some contradictory requirements for the IMUs: smaller, better, cheaper and last but not least safer.
The quaternion and dual quaternion have been widely used in describing the three-dimensional rigid-body motion. This work starts by introducing their Clifford algebra representations, followed by proposing the Clifford algebra formulation of the trident quaternion, fulfilling the simultaneous representation of attitude, velocity and position. Besides, the exponential/logarithmic map of trident quaternion is developed. Finally, the kinematic model of the trident quaternion is presented, and the left/right-error kinematic models of the inertial navigation system are used to design extended Kalman filters (EKFs). Land vehicle experiments demonstrate that the EKF developed with the Clifford algebra has better performance in odometer-aided in-motion alignment.
The results of an observability analysis of a CAI-based IMU sensor fusion model, based on findings from an extensive analysis of inertial datasets collected over the last 10 years at the Institut für Erdmessung of the Leibniz University of Hannover, are presented. Datasets are analysed with respect to characteristic peaks occurring in the body frame during acceleration, deceleration, and turn maneuvers. This is done for IMU datasets recorded on board trains and cars. Based on these findings, “characteristic” maneuvers are derived for the forward (x) and right (y) axis of the accelerometer, and the z axis of the gyroscope in the body frame. Maneuvers are derived by ranking multiple possible function fits on a RMSE-based evaluation method. This results in best fitting functions which are used to confirm the observability of different systematic IMU error terms with respect to a CAI-based reference sensor. Turn maneuvers result in dynamics across both accelerometer and the gyroscope axes, which in turn leads to observability of misalignments. For acceleration and deceleration maneuvers, only the longitudinal axis of the vehicle exhibits changes in acceleration, which should also be sufficient to estimate the misalignment terms between the conventional IMU and the CAI-based sensor. Meanwhile, the lever arm (displacement) between the CAI and IMU cannot be reliably estimated by maneuvers considered here, as it requires significant angular rates along two axes. A solution to this problem could be the oscillation due to suspension visible in the car-based datasets, which have a frequency of 0.2-1 Hz, and an amplitude of up to 0.1 rad/s. Based on these results, a follow-up study is suggested with real CAI sensor measurements to estimate the impact of such slow oscillations on the sensor solution.
The gravity gradiometer is an instrument that continuously measures the small gravity gradient changes on the earth's surface. The actual output signal of the system contains a lot of noise, and the signal-to-noise ratio is extremely low due to the limitations of the process and performance level of the core sensitive element accelerometer, and the installation errors of multiple links. In order to effectively extract the real gravity gradient signal from the strong noise, a gravity gradient noise reduction method based on the equivalent source model is proposed by using the basic principle that the earth's gravity field is a conservative force field. First, according to the relationship between the rectangular prism and the gravity gradient, the analytical model of the equivalent source is derived, and the regression equation of the equivalent source is established. Aiming at the ill conditioned problem of regression operators, Tikhonov regularization method is proposed, and the optimal regularization parameters are determined by L-curve method, which can reduce the noise of gravity gradient measurement by an order of magnitude in simulation. Using this method to process the measured data of the shipborne test, the results show that this method has an obvious effect on suppressing the dynamic measurement noise of the gravity gradiometer, and can reduce the internal coincidence mean square error of the horizontal tensor gravity gradient measurement signal by 58%, which is of engineering significance.
Satellite navigation systems, inertial navigation systems and integrated systems (satellite navigation systems and inertial navigation systems) are used to determine navigation parameters. But satellite navigation systems are not autonomous. Although inertial navigation systems are autonomous, they accumulate errors over time that require external correction. A new method of autonomous determination of latitude and longitude of moving objects is proposed. Unlike standard methods, this method does not use double integration of accelerometer signals. To determine latitude and longitude, gyroscope signals are used, which are part of the inertial measurement unit (IMU). To implement the method of autonomous determination of latitude, it is enough to have the signals of the IMU gyroscopes, the elements of the direction cosines matrix, the angles of rotation of the object, and their derivatives. For the autonomous determination of longitude, it is necessary, in addition, to know the initial value of the longitude. The use of the Luenberger observer to determine of latitude and longitude is considered.
MIMU-M (MEMS based IMU Module) is an ITAR-free Inertial Navigation System (INS) which matches high performance into miniaturized dimensions (35 × 45 mm) and lower than 20gr weight. MIMU-M architecture is based on 4 MEMS IMUs, each one containing 3-axis gyroscopes and 3-axis accelerometers with a total of 24 axis, whose fusion provides the system orientation and rotation along the body axes. To enhance the IMU performance, single MEMS IMUs are subjected to a substantial screening process to preliminary discriminate anomalies and undesired behavior. The screened MEMS IMU is in-house fully calibrated with proprietary temperature and motion profiles to properly characterize the system in all operational environments. This calibration process, together with the IMU fusion, allows us to obtain exceptional results especially in terms of ARW (<0.17 deg/√hr) and gyroscope bias stability (< 0.7 deg/hr) that are essential requirements for navigation. The extracted compensation is embedded into the system through an ARM-based microprocessor, responsible also for the fusion algorithm. The MIMU-M includes a GNSS receiver, whose output is combined with the sensor block to estimate the hybrid INS output (attitude, heading, velocities and position) by means of a Kalman filter algorithm. In support of this, software real-time built-in tests are executed in the INS at power-on and continuously to check the correct working mode of the system. Moreover, a proprietary algorithm of IMUs fault detection (FDI) and recovery (FDIR) can be included to boost the system reliability in critical operations. This FDI algorithm identifies faults in navigation system caused by defective sensor (slowly ageing or suddenly broken) or variation in the operational environment that compromises the navigation model. This strategy is based on the placement of redundant IMUs (RIMU) which are used to isolate out-of-specification sensors allowing the system to continuously operate in a fail-safe mode (FDIR) with the remaining IMUs. MIMU-M system experienced an extensive test phase for validating data accuracies and testing the versatility in different environments. Successful land and flight tests were executed in different conditions included the most challenging that being GNSS outages. Test results showed that the MIMU-M offers excellent accuracy, especially on flight with errors (expressed as RMS) for attitude less than 0.2° and true heading accuracy drift equal to 1°/10min with GNSS aiding. Thanks to the combination of high performance and fault tolerance with the miniaturized volume and weight, MIMU-M aims at competitively occupying a high-level position into the MEMS INS product space.
MEMS IMUs are the most common units for data collection in the field of pedestrian inertial navigation, whereas on the data processing side a Kalman filter is a broadly used algorithm for the required pedestrian motion state estimation. The paper investigates the influence of several IMU-related parameters on the accuracy of this state estimation. It complements a former study of the authors that discussed algorithmic enhancements of the data processing leading to noteworthy accuracy improvements compared to the state of the art of pedestrian inertial navigation [1]. Therefore, the IMU-related parameters will be reassessed against this enhanced numerical background. The paper is based on test walks employing a special test track and a wearable overshoe equipped with two MEMS IMUs of different accuracy levels. The IMU data were then fused with a continuous-discrete total-state Kalman filter. The only aiding techniques were Zero Velocity Update (ZUPT) and Zero Angular Rate Update (ZARU) to exclude influences from other sensors than the IMUs. The paper considers the following points concerning their influence on the position and on the yaw estimation accuracy, which are especially error-prone motion states in pure ZUPT- and ZARU-aided inertial navigation: •Influence of the IMU sample rate for inertial data collection. •Influence of a pretest IMU calibration with respect to bias, scale factor, and sensor axes misalignment. •Incorporation of Allan variance sensor parameters in the Kalman filter. The parameters were used to model all inertial sensor biases by a random walk or a Gauss-Markov process of first order. The results on these points can be outlined as follows. •Higher sample rates show a significantly improved capture of the step dynamics leading to smoother and more accurate estimates of the walking path. •Against the enhanced numerical background mentioned above, the IMU calibration leads typically (but not always) to further accuracy improvements. •The state of the art in ZUPT-based pedestrian navigation does not show a clear advantage of using a Gauss-Markov process for the IMU sensor biases. This seems also to be the case for the algorithmic improvements mentioned above.
The widely used Error State Kalman Filter (ErKF) is known to deliver a suitable navigation solution for most navigation applications. A commonly used architecture is an Inertial Measurement Unit (IMU) that delivers system input to a strapdown algorithm while aiding sensors like Global Navigation Satellite Systems (GNSS) and Barometric Altitude Sensors are used to update the error state. Following the evolution from single-sensor to multi-sensor integration leads to the concept of collaborative sensors to improve the navigation capability of a group of users. The idea is that group members collaboratively share information about their state estimates and sensor data which enables the group to determine an optimal navigation solution for all users. In real-time collaborative navigation system, there are two challenges that need to be solved and they cannot be sharply separated. First, a dynamic sensor network must be established and maintained that, in addition to communication, allows precise time synchronization and at least measurement of the range between nodes (inter-nodal ranging). Second, the estimation of the distributed sensors must be solved, which can be underdetermined and highly nonlinear. To achieve the optimal solution one would have to process the data of all users and additional inter-nodal information (e.g. ranges or direction) in one centralized ErKF. This would however impose significant communication costs on all users and computation effort on the master node that conducts the calculations. A cheaper approach in terms of communication and calculation efforts is a decentralized filter algorithm where only certain information are available to each member of the network. The filter calculations are not conducted by one master node but each member uses their available information to enhance their own position estimate. The navigation solution in this approach will not be as accurate as the centralized solution but more efficient and more robust to disturbances in communication. This paper shows simulation results of a network-based collaborative navigation scenario featuring a decentralized approach for the state estimation. Additional to his own state vector, a network member is assumed to have knowledge about the inter-nodal range of itself to a changing number of other members of the same network as well as their position estimates. Certain nodes have independent knowledge about their absolute position, i.e. are considered to have GNSS information available. On the basis of these information, each member can enhance their position estimate. The behavior of the network in cases of GNSS and/or inter-nodal range outages are shown with a special focus on the influence of IMU performance parameters. For demonstration several IMU performance classes are simulated and results compared. As a benchmark, we compare our results with results generated by the above mentioned centralized Kalman filter.
Nowadays, GNSS signal spoofing has become an increasingly concrete threat. Low-cost consumer electronics, software defined radios and ADS-B are some of the keys that made GNSS spoofing technology accessible. A function for detecting, or even mitigating, the occurrence of a spoofing attack on GNSS signals has therefore become mandatory for civilian equipment such as the SkyNaute. Thus, the new standard for GNSS aided inertial equipment (RTCA DO384) includes a dedicated appendix specifying how to claim and evaluate the performance of such a function. The solution patented by Safran consists of analyzing the statistical behavior of the shifts computed by the hybridization filter. Several new detectors based on shift magnitude and direction have been tested. Under normal conditions, this shift direction is random with a relatively large standard deviation. In the event of GNSS signals coherent spoofing, the shift direction becomes constant; the standard deviation tends towards zero. This is the principle of the detector implemented by Safran. This article aims to present these new detectors, their reaction under spoofing condition and the results of the first evaluation campaign. Further evaluation campaigns and development will be presented in the conclusion.
Thanks to the breakthrough brought by the Hemispherical Resonator Gyroscope (HRG), Safran Electronics & Defense has been introducing the most efficient navigation-grade Inertial Navigation Systems (INS) in terms of C-SWaP (Cost, Size, Weight and Power), with products such as: •Onyx, high-performance land and sea inertial reference system ([1]) •Skynaute, ultra-compact hybrid navigation system for helicopters, UAVs and aircrafts ([2]) •Spacenaute, high-end inertial measurement unit for space launchers, qualified on Ariane6. By pairing HRG Crystal™ and macro-sized pendulous accelerometers, those products have already proven their ability to achieve the highest navigation performance, with better C-SWaP characteristics than optical gyroscope technologies (FOG and RLG), and much higher reliability. In order to move forward and push back the limits of integration capability of inertial navigation systems, Safran Electronics & Defense has been working for several years on the development of a new generation of navigation-grade, MEMS-based closed loop accelerometers ([3], [4]). Through their combined expertise, Safran Electronics & Defense and its daughter-company Safran Sensing Technologies Switzerland (formerly known as Safran Colibrys) have now brought this new sensor design to life, and the sensor is already in production in Safran's factories. Previous presentations ([3]) showed the performance of this new accelerometer at sensor level, along with Safran's capability to mass-manufacture such accelerometer. Following the outstanding results obtained at sensor level, this new generation of accelerometers has been paired with HRG Crystal™ to design a new generation of Inertial Navigation Systems. This paper presents the key design criteria of such inertial core, along with experimental performance evaluations results of the Inertial Navigation Systems in which it is integrated, on various types of platforms and mission profiles, such as: –North Finding –Navigation on airborne platforms –Land navigation In particular, the results will focus on orientation and position accuracy in pure inertial modes, this showing the absolute resilience of such position and navigation solution to GNSS-denied conditions. A particular focus will also be given to robustness tests on the accelerometer, and associated experimental results.
In Coriolis vibratory Rate-Integrating Gyroscopes (RIG), a higher electrical Signal-to-Noise Ratio (SNR) provides an improved resolution for measuring the orientation of the oscillation pattern. Therefore, precession due to a smaller angular rotation can be detected by maximizing the vibration amplitude. On the other hand, the contribution of non-linear mechanisms to the resonator dynamics, the measurements, and the control grows in complexity with the oscillation amplitude. This paper studies the trade-offs between achieving a higher SNR through increasing the oscillation amplitude and the adverse effects of non-linear mechanisms in RIG. As part of our study, we derived analytical equations to demonstrate that by operating a RIG in the non-linear regime, electrostatically-induced amplitude-frequency coupling results in an angle-dependent anisoelasticity. Anisoelasticity was shown to reduce angular gain and introduce angle-dependent biases in measurements despite utilizing a quadrature control for suppressing ellipticity. We proposed a modified Whole Angle (WA) control, which utilizes a non-linear feedback loop to compensate for the amplitude-frequency coupling along the X and Y axes. By implementing the control with a Micro-Electro-Mechanical System (MEMS) Dual Foucault Pendulum (DFP) gyroscope, we were able to operate the RIG with a vibration amplitude equal to 15% of the capacitive gap size and reduce ellipticity by 100 times. The magnitude of control voltages for compensation of nonlinearity was shown to depend on the biasing voltages, which were intentionally applied for mode-matching. Based on our observations, we conjectured that impractically large control voltages coupled with phase drifts in electronics degrade the stability of RIG when operated in the non-linear regime. Our results highlight the limitations of feedback compensation methods in RIG and the need for fabrication of structurally symmetric micro-gyroscopes for high-resolution and high-accuracy angle measurements.
ARIETIS-NS is a Rad-Tolerant, space qualified 3-axis gyro, whose main applications are Telecom (15+ years GEO), Earth Observation as well as Science and Exploration missions. It mostly uses commercial EEE components that are upscreened by means of radiationtesting. ARIETIS-NS is in the process of being qualified to ESA ECSS standards, meeting the most stringent space quality requirements. ARIETIS-NS, as his Hi-Rel equivalent ARIETIS gyro, is based on Innalabs proprietary Coriolis Vibratory Gyroscope (CVG) technology which is currently used in commercial products for land, marine, and aerospace applications. Innalabs CVG gyros, as presented previously at this conference, have already accumulated more than 2,500,000 hours in space on board a Low Earth Orbit constellation with no failures and no deviation from the specified performance, confirming the suitability of the technology to space environment. ARIETIS-NS has already been selected for a variety of space applications and environments such as LEO, GEO, and heliocentric orbits. Thanks to the support of the European Space Agency, the development of ARIETIS-NS is now complete and qualification testing is ongoing with target completion date in Q4 2022, ahead of delivery of the first flight models. ARIETIS-NS is able to provide high performance (ARW <0.005 deg/ $\sqrt{\text{hr}}$ ) and high reliability (~1000 FITs) in a very compact design (~1.2 kg for LEO application) and very low power consumption (<7W). To achieve this, several targeted enhancements and optimisations to InnaLabs gyroscope's design have been introduced, mainly in the component selection, the control loops, the compensation algorithms, and a smaller Sensing Element design. After a brief description of the InnaLabs CVG basic principles and an overview of the CVG technical strengths in comparison to competing technologies, this paper describes the specification, key design features, and budgets of ARIETIS-NS. The paper will present the flexibility embedded in the design and the different configuration of the unit. These include different box thickness for LEO vs GEO applications, different data buses supported, the possibility of changing gyro bandwidth depending on customer needs. Testing and verification approach is then introduced. A number of Engineering Models (EM) as well as Engineering Qualification Models (EQM, practically identical to Flight Models) have been built as part of the verification campaign. Test results obtained with both models are presented to validate ARIETIS-NS specification. These include performance tests (bias stability, scale factor stability, noise, and misalignment), mechanical tests (vibration and shocks) as well as thermal vacuum tests. Test results show that ARIETIS-NS meets and exceeds its specification.
In recent years, vibrating sensors have become ubiquitous [1]–[3]. This is because of their simple design (no moving parts), robustness, and continuous improvements in algorithms and electronics. Nowadays, vibrating gyroscopes are replacing well-established optical technologies like ring laser and fiber optics gyroscopes [4]. Vibrating gyroscopes are based on the Coriolis Effect [5]. For instance, when a standing wave is excited in an axisymmetric structure (resonator) and an angular rate is applied; the Coriolis acceleration that appears on the structure causes the vibrating pattern to rotate around its symmetry axis, with a velocity proportional to the external angular rate [6]. Non-ideal vibrating sensors suffer from a well-known effect called lock-in. For external angular velocities smaller than the lock-in rate, the standing wave does not rotate and the sensor fails to operate [7]. For instance, Coriolis Vibrating Gyroscopes (CVGs), working in the Whole Angle Mode (WAM), have a lock-in threshold proportional to the difference between the inverse of the maximum and the inverse of the minimum values of the damping time constant [8]. Recently, a Slow Variables (SV) method has been used to demonstrate this effect as well as to calculate the lock-in angle [7]. In this work, using a Fast Variables (FV) method, we derive the lock-in condition without making any assumptions and/or approximations. We solve the dynamics exactly and we find equivalent conditions for the system's eigenvalues to those known for the Damped Harmonic Oscillator (DHO), i.e., angular rate decreasing to zero - over-damped dynamic, lock-in - critically damped dynamic, and free rotation - under-damped dynamic. Therefore, we can explain the lock-in effect without resorting to a nonlinear theory (SV method) but by the form of the roots of an eigenvalue problem of a linear system (FV method).
The purpose of navigation is to determine the position, velocity, and orientation of manned and autonomous platforms, humans, and animals. Obtaining accurate navigation commonly requires fusion between several sensors, such as inertial sensors and global navigation satellite systems, in a model-based, nonlinear estimation framework. Recently, data-driven approaches applied in various fields show state-of-the-art performance, compared to model-based methods. In this paper we review multidisciplinary, data-driven based navigation algorithms developed and experimentally proven at the Autonomous Navigation and Sensor Fusion Lab (ANSFL) including algorithms suitable for human and animal applications, varied autonomous platforms, and multi-purpose navigation and fusion approaches
A sensor fusion algorithm of a conventional MEMS IMU with semi-simulated cold atom interferometer (CAI) sensor values, based on an ESEKF, is demonstrated, and its performance is analyzed under different boundary conditions. CAI sensor values are generated from real navigation grade RLG IMU measurements. The MEMS IMU sensor data is used as input to a navigation strapdown algorithm. The developed ESEKF is able to track different systematic error terms, such as bias components, relative to the CAI reference. Dynamic and static experiments are recorded, with 10 repetitions. The drift of the MEMS strapdown + ESEKF (filtered) navigation solution is compared to the drift of a MEMS-only (unfiltered) strapdown navigation solution, with respect to a reference solution based on the navigation grade sensor data. The drift in position, velocity and attitude of the filtered solution is reduced by a factor of 30 (in position) to 100 (in attitude) with respect to the unfiltered solution. In the dynamic case, the drift is reduced from 9848.56 ± 297.27 m to 335.93 ± 138.51 m. A sensor delay study is performed as well. The system is stable for delays of up to 0.25 seconds between both sensors. For larger delays, the variance components of the bias terms in the process noise matrix need to be adjusted, as the large noise of the MEMS sensor causes divergence of the system. After this adjustment, the filter does not diverge for a higher delay, a smoothed error estimate results, which exhibits a worse performance when compared to the previously mentioned filter solutions with smaller delays. On the other hand, it performs better than the unfiltered solution. These results validate the applicability of the algorithm to real data. Furthermore, it is shown that the algorithm is stable even if sensors of different grades are used. Finally, divergence of the system only occurs if the sensor differences between two filtering steps are too large, which indicates that the grade of the conventional sensor, and the performed dynamics, are important characteristics of this type of filter.
Beside the camera, MEMS IMUs (Micro Elector-Mechanical Systems) belong today to the standard sensor conglomerate that every smartphone should have. In the future, the importance of MEMS-IMUs will increase more and more, especially if we talk about smart cities or internet of things (IoT). Since the MEMS manufacturer care only about numbers, i.e., low-cost, size and power consumption, some hardware...
