Strapdown inertial navigation systems are the most common form of inertial navigation system due ... more Strapdown inertial navigation systems are the most common form of inertial navigation system due to its potential benefits of lower cost, reduced size, and greater reliability compared with equivalent gimbal systems. This chapter introduces fundamentals on the strapdown inertial navigation mechanism. In the n‐frame, navigation data are expressed by velocity components along the North, East, and Down directions, and latitude, longitude, and altitude. Therefore, it is more commonly used in navigation applications on the Earth or in the vicinity of the Earth surface. The Coriolis acceleration can be neglected in cases where the navigation error caused by the inertial measurement units (IMU) measurement error is much greater than the Coriolis effect. Attitude initialization, unlike position and velocity initialization, can be achieved by inertial sensors if the IMU is stationary with respect to the Earth. Gyrocompassing requires the IMU to measure the Earth's rotation, which is around 15 °/h, to obtain the yaw angle information.
Glassblown 3D micro shells are released by a multi-step mechanical lapping process, introducing s... more Glassblown 3D micro shells are released by a multi-step mechanical lapping process, introducing structural imperfections to the sensing element. We present an analytical model predicting the frequency mismatch in micro shells due to imperfections induced by the release process. The analytical model of such micro shells was derived based on inextensional assumption. Mode shapes and the resonant frequencies were calculated and compared to results from both Finite Element Analysis (FEA) and experiments. The predictive frequency of the n=2 wineglass mode shape was within 10% of the finite element results and 20% of the experimental results for thin shells, showing the fidelity of the model. The effect of imperfections of the shell release process on frequency mismatch was studied both analytically and experimentally.
We developed a learning-based calibration algorithm for a vestibular prosthesis with the long-ter... more We developed a learning-based calibration algorithm for a vestibular prosthesis with the long-term goal of reproducing error-free vestibular system dynamic responses. Our approach uses an additional IMU to detect the head acceleration of a patient and to correct the corresponding drift in the vestibular prosthesis. The algorithm includes four major parts. First, we extract features from the shoe-mounted IMU to classify human activities through convolutional neural networks. Second, we fuse data from the head-mounted IMU (vestibular prosthesis). Third, we artificially create additional data samples from a small pool of training data for each classification class. Fourth, we use the classified activities to calibrate the reading from the head-mounted IMU. The results indicate that during daily routine activities the firing rate baseline of a vestibular prosthesis system without calibration fluctuates between 100 pulses/s to 150 pulses/s; in contrast, an appropriate calibration to human activity results in correction of 4 pulses/s in extreme cases, providing a stable baseline firing rate while the head is not moving. In this work, we specifically study the contribution of gyroscope scale factor on the drift of the vestibular prosthesis system and propose a corresponding calibration method.
Pedestrian navigation has been of high interest in many fields, such as human health monitoring, ... more Pedestrian navigation has been of high interest in many fields, such as human health monitoring, personal indoor navigation, and localization systems for first responders. Due to the potentially complicated navigation environment, selfcontained types of navigation such as inertial navigation, which do not depend on external signals, are more desirable. Pure inertial navigation, however, suffers from sensor noise and drifts and therefore is not suitable for long-term pedestrian navigation by itself. Zero-velocity update (ZUPT) aiding technique has been developed to limit the navigation error growth, but adaptivity of algorithms, model fidelity, and system robustness have been major a concern if not properly addressed. In this paper, we attempt to establish a common approach to solve the problem of self-contained pedestrian navigation by identifying the critical parts of the algorithm that have a strong influence on the overall performance. We first review approaches to improve the na...
2019 IEEE International Symposium on Inertial Sensors and Systems (INERTIAL), 2019
Presented a customizable laboratory testbed for the sensor fusion of multiple Inertial Measuremen... more Presented a customizable laboratory testbed for the sensor fusion of multiple Inertial Measurement Units (IMU) and SOund Navigation And Ranging sensors (SONAR), for self-contained navigation experiments. We described the architecture and communication interfaces for the simultaneous collection of data from two IMUs and two SONARs, however, parallel acquisition from a larger variety of sensors is also feasible. Representative human gait patterns are experimentally acquired to demonstrate the functionality of the testbed.
2019 IEEE International Symposium on Inertial Sensors and Systems (INERTIAL), 2019
We present a ranging-based aiding method for pedestrian inertial navigation, utilizing not only r... more We present a ranging-based aiding method for pedestrian inertial navigation, utilizing not only ranging readouts, but also orientation of the ranging sensors. Both numerical and experimental results demonstrated improvements in navigation accuracy using the method. Kalman Filter (KF) was implemented to merge inertial navigation with Zero-Velocity-Update (ZUPT) algorithm and foot-to-foot directional ranging information. The improvement of navigation accuracy was achieved purely algorithmically without any increase in complexity of the hardware. We demonstrated experimentally that the navigation errors can be improved by about two times using the method.
2020 DGON Inertial Sensors and Systems (ISS), 2020
In this paper, we proposed a novel zero velocity detector, the Dynamic-Vision-Sensor (DVS)-aided ... more In this paper, we proposed a novel zero velocity detector, the Dynamic-Vision-Sensor (DVS)-aided Stance Phase Optimal dEtection (SHOE) detector, for Zero-velocity-UPdaTe (ZUPT)-aided Inertial Navigation Systems (INS) augmented by a foot-mounted event-based camera DVS128. We observed that the firing rate of the DVS consistently increased during the swing phase and decreased during the stance phase in indoor walking experiments. We experimentally determined that the optimal placement configuration for zero-velocity detection is to mount the DVS next to an Inertial Measurement Unit (IMU) and face the sensor outward. The DVS-SHOE detector was derived in a General Likelihood Ratio Test (GLRT) framework, combining statistics of the conventional SHOE detector and the DVS firing rate. This paper used two methods to evaluate the proposed DVS-SHOE detector. First, we compared the detection performances of the SHOE detector and the DVS-SHOE detector. The experimental results showed that the DVS-SHOE detector achieved a lower false alarm rate than the SHOE detector. Second, we compared the navigation performance of the ZUPT-aided INS using the SHOE detector and the DVS detector. The experimental results showed that the Circular Error Probable (CEP) of the case using DVS-SHOE was reduced by around 25 % from 1.2 m to 0.9 m, as compared to the case of the SHOE detector.
This paper reports on the design and implementation of high efficiency, nonmetallic reflectors in... more This paper reports on the design and implementation of high efficiency, nonmetallic reflectors integrated on the sidewalls of micromachined cavities. Due to shadowing from deposition within a cavity, significant variation in the thicknesses of the dielectric thin films composing the reflectors are encountered when the layers are deposited by Plasma Enhanced Chemical Vapor Deposition (PECVD). These gradients in thickness limit the performance of the reflector at the intended design wavelength. An optimized design procedure is described to maximize the performance of the reflector at the D 1 absorption wavelength of 87 Rb of 795 nm for use in micromachined atomic vapor cells. The reflector design is based on multiple shifted quarter wave Bragg reflectors in series, which extends the reflective bandwidth for increased robustness to film thickness gradients. The extended reflectance range maintains high reflection at the design wavelength despite greater than 70% decrease in film thickness across the reflector surface. The reflector technology is ideally suited for use in atomic MEMS vapor cell applications by achieving high reflectance while maintaining light polarization. We demonstrate less than 2 dB of return loss with circular polarization ellipticity maintained to ±2 • .
2016 Solid-State, Actuators, and Microsystems Workshop Technical Digest, 2016
This paper presents experimental results on the noise performance of an ultra high quality factor... more This paper presents experimental results on the noise performance of an ultra high quality factor (Q-factor) micromachined silicon Quad Mass Gyroscope (QMG). The device was vacuum sealed at below 0.1mTorr in an LCC package with activated getters, demonstrating the Q-factor better than 1.7 million, along the drive and sense modes, at the 1.6 kHz center frequency. A bias stability of 0.09 deg/hr and Angle Random Walk (ARW) of 0.015 deg/rt-hr were demonstrated, without any thermal compensation or post-processing of data. To the best of our knowledge, the result presented in this paper is the highest reported Q-factor for MEMS silicon gyroscopes, now demonstrating a true potential to achieve the navigation-grade performance.
2020 IEEE/ION Position, Location and Navigation Symposium (PLANS), 2020
In this paper, we investigate how self-contained pedestrian navigation can be augmented by the us... more In this paper, we investigate how self-contained pedestrian navigation can be augmented by the use of foot-tofoot visual observations. The main contribution is a measurement model that uses Zero velocity UpdaTe (ZUPT) and relative position measurements between the two shoes obtained from shoe-mounted feature patterns and cameras. This measurement model provides directly the compensation measurements for the three position states and three velocity states of a pedestrian. The involved features for detection are independent of surrounding environments, thus, the proposed system has a constant computational complexity in any context. The performance of the proposed system was compared to a standalone ZUPT method and a relative-distance-aided ZUPT method. Simulation results showed an improvement in accumulated navigation errors by over 90%. Real-world experiments were conducted, exhibiting a maximum improvement of 85% in accumulated errors, verifying validity of the approach.
56th Electronic Components and Technology Conference 2006
Large 6mm x 9mm silicon dice have been successfully bonded on alumina substrate with electroplate... more Large 6mm x 9mm silicon dice have been successfully bonded on alumina substrate with electroplated Au80Sn20 eutectic alloy. Eutectic AuSn is one of the best known hard solders having excellent fatigue-resistance and mechanical properties. A fluxless bonding process in 50militorrs of vacuum environment is presented. Vacuum environment is employed to prevent tin oxidation during the process. The oxygen content is expected to be reduced by a factor of 15,200, comparing to bonding in air. One of the challenges in silicon-to-alumina bonding is the large mismatch in thermal expansion between silicon of 2.7x10-6 ppm/°C and alumina of 7x10-6 ppm/°C. Electroplating method is used to build multi-layer solder. It is an economical alternative to vacuum deposition method and can produce thick solders. Joints fabricated are examined using Scanning Electron Microscope (SEM), and Energy Dispersive X-ray Spectroscopy (EDX). It is found that proper bonding condition is needed to turn the stacked layers into a uniform AuSn eutectic alloy. Nearly void-free joints are achieved and confirmed by a Scanning Acoustic Microscope (SAM). To evaluate the reliability of the solder joint and the bonded structure, samples will go through thermal cycling test to determine failure modes. Microstructural changes of the solder joints during thermal cycling test will also be investigated.
Proceedings of the 33rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2020), 2020
He is currently pursuing the Ph.D. degree in mechanical engineering in University of California I... more He is currently pursuing the Ph.D. degree in mechanical engineering in University of California Irvine. His research interests include design, fabrication, and control of vibratory inertial sensors and capacitive micromachined ultrasonic transducers (CMUT). Andrei M. Shkel received the Diploma (Hons.
2020 IEEE/ION Position, Location and Navigation Symposium (PLANS), 2020
We present an analytical expression for estimating the position error in the vertical direction i... more We present an analytical expression for estimating the position error in the vertical direction in pedestrian navigation using an Inertial Measurement Unit (IMU) aided by an altimeter. The analytical expression is found to be directly proportional to the altimeter resolution, the square root of the altimeter sampling rate, IMU Velocity Random Walk (VRW), and a ratio of time intervals in the stance phase and the swing phase during the gait cycle, and inversely proportional to the square root of IMU sampling rate. The analytical analysis was verified by numerical simulations and supported by experimental results. The numerical simulation of the ZUPT-augmented navigation aided by an altimeter was conducted based on a mathematically simulated foot motion trajectory. Experiments were implemented on a foot-mounted platform containing IMUs and altimeters. The results from both simulations and experiments matched our analytical prediction, with an error less than 20%.
A signal-sensor-based indoor pedestrian navigation system is developed. The proposed system: (i) ... more A signal-sensor-based indoor pedestrian navigation system is developed. The proposed system: (i) utilizes a foot-mounted inertial navigation system (INS), in which the accumulated errors are mitigated via a zero velocity update (ZUPT) approach and (ii) exploits opportunistically cellular long-term evolution (LTE) signals in a deep neural network (DNN)-based synthetic aperture navigation (SAN) framework, in which the pedestrian's motion is utilized to suppress multipath-induced errors. The proposed DNN-SAN-LTE-ZUPT-INS (DUALS) indoor pedestrian navigation system utilizes the complementary the desirable characteristics of both subsystems, coupled via two architectures: (a) loosely-coupled and (b) tightly-coupled. This paper designs and assesses both architectures experimentally in an indoor environment. The experimental study demonstrates a pedestrian traversing a trajectory of 600 m in 14 minutes, including a stationary period, straight segments, up/down the stairs, and riding in an elevator, while receiving signals from 4 LTE base stations (also known as evolved node B (eNodeB)). The proposed tightly-coupled DUALS system exhibited a 3-D position root mean-squared error (RMSE) of 1.34 m, outperforming the loosely-coupled DUALS, ZUPT-aided INS, and LTE-DNN-SAN which achieved a position RMSE of 1.38, 1.49, and 1.97 m, respectively.
In this paper, we propose a Foot-Instability-Based Adaptive (FIBA) covariance to dynamically adju... more In this paper, we propose a Foot-Instability-Based Adaptive (FIBA) covariance to dynamically adjust the covariance matrix for the pseudo-zero-velocity measurements in the Zero velocity UPdaTe (ZUPT)-aided Inertial Navigation Systems (INS). The proposed ZUPT-aided INS using the FIBA covariance is implemented in an Adaptive Extended Kalman Filter (AEKF) framework, where the measurement covariance matrix is updated in each iteration according to the FIBA covariance. The FIBA covariance is designed to have a very high value during the swing phases in a gait cycle, and the value significantly decreases during the stance phases. As a result, the proposed method eliminates a need to use a binary stance phase detector in implementation of the ZUPT-aided INS. Two series of indoor pedestrian navigation experiments were conducted to investigate the navigation performance of the algorithm. In the first series of experiments, which included cases of walking and running, localization solutions produced by the system using the FIBA covariance demonstrated 36% and 64% improvements in navigation accuracy along the horizontal and vertical directions, respectively. In the second series of experiments, which included a pedestrian walking on different indoor terrains, such as flat planes, stairs, and ramps, the navigation accuracy of the system using the FIBA covariance reduced horizontal and vertical position errors by 12% and 45%, respectively, as compared to the conventional ZUPT-aided INS.
Proceedings of 1995 34th IEEE Conference on Decision and Control
Control schemes in real-time sensor-based systems often operate under tight time constraints dete... more Control schemes in real-time sensor-based systems often operate under tight time constraints determined by the system sampling rate. One area where uch constraints are especially severe is the sensorbased motion planning with dynamics in robotics. Though very important both theoretically and in practice, this problem has not been addressed so far. The typical sampling rates in such systems are 20 to 50 %>er second. This leaves only 20 to 50 msec for the whole cycle, including sensing, complex geometric (intelligence) analysis, calculations due to dynamics and control, and motion execution. As a Arst attempt to solve the problem, we show that the time constraints an be met by a combination of a simple model with a carefully chosen analytical solution of the dynamic equations. The resulting control scheme guarantees Convergence and safety of motion, and blends well with kinematic path planning algorithms. Two such strategies are discussed, one using a simple heuristic and the other with local optimization.
Strapdown inertial navigation systems are the most common form of inertial navigation system due ... more Strapdown inertial navigation systems are the most common form of inertial navigation system due to its potential benefits of lower cost, reduced size, and greater reliability compared with equivalent gimbal systems. This chapter introduces fundamentals on the strapdown inertial navigation mechanism. In the n‐frame, navigation data are expressed by velocity components along the North, East, and Down directions, and latitude, longitude, and altitude. Therefore, it is more commonly used in navigation applications on the Earth or in the vicinity of the Earth surface. The Coriolis acceleration can be neglected in cases where the navigation error caused by the inertial measurement units (IMU) measurement error is much greater than the Coriolis effect. Attitude initialization, unlike position and velocity initialization, can be achieved by inertial sensors if the IMU is stationary with respect to the Earth. Gyrocompassing requires the IMU to measure the Earth's rotation, which is around 15 °/h, to obtain the yaw angle information.
Glassblown 3D micro shells are released by a multi-step mechanical lapping process, introducing s... more Glassblown 3D micro shells are released by a multi-step mechanical lapping process, introducing structural imperfections to the sensing element. We present an analytical model predicting the frequency mismatch in micro shells due to imperfections induced by the release process. The analytical model of such micro shells was derived based on inextensional assumption. Mode shapes and the resonant frequencies were calculated and compared to results from both Finite Element Analysis (FEA) and experiments. The predictive frequency of the n=2 wineglass mode shape was within 10% of the finite element results and 20% of the experimental results for thin shells, showing the fidelity of the model. The effect of imperfections of the shell release process on frequency mismatch was studied both analytically and experimentally.
We developed a learning-based calibration algorithm for a vestibular prosthesis with the long-ter... more We developed a learning-based calibration algorithm for a vestibular prosthesis with the long-term goal of reproducing error-free vestibular system dynamic responses. Our approach uses an additional IMU to detect the head acceleration of a patient and to correct the corresponding drift in the vestibular prosthesis. The algorithm includes four major parts. First, we extract features from the shoe-mounted IMU to classify human activities through convolutional neural networks. Second, we fuse data from the head-mounted IMU (vestibular prosthesis). Third, we artificially create additional data samples from a small pool of training data for each classification class. Fourth, we use the classified activities to calibrate the reading from the head-mounted IMU. The results indicate that during daily routine activities the firing rate baseline of a vestibular prosthesis system without calibration fluctuates between 100 pulses/s to 150 pulses/s; in contrast, an appropriate calibration to human activity results in correction of 4 pulses/s in extreme cases, providing a stable baseline firing rate while the head is not moving. In this work, we specifically study the contribution of gyroscope scale factor on the drift of the vestibular prosthesis system and propose a corresponding calibration method.
Pedestrian navigation has been of high interest in many fields, such as human health monitoring, ... more Pedestrian navigation has been of high interest in many fields, such as human health monitoring, personal indoor navigation, and localization systems for first responders. Due to the potentially complicated navigation environment, selfcontained types of navigation such as inertial navigation, which do not depend on external signals, are more desirable. Pure inertial navigation, however, suffers from sensor noise and drifts and therefore is not suitable for long-term pedestrian navigation by itself. Zero-velocity update (ZUPT) aiding technique has been developed to limit the navigation error growth, but adaptivity of algorithms, model fidelity, and system robustness have been major a concern if not properly addressed. In this paper, we attempt to establish a common approach to solve the problem of self-contained pedestrian navigation by identifying the critical parts of the algorithm that have a strong influence on the overall performance. We first review approaches to improve the na...
2019 IEEE International Symposium on Inertial Sensors and Systems (INERTIAL), 2019
Presented a customizable laboratory testbed for the sensor fusion of multiple Inertial Measuremen... more Presented a customizable laboratory testbed for the sensor fusion of multiple Inertial Measurement Units (IMU) and SOund Navigation And Ranging sensors (SONAR), for self-contained navigation experiments. We described the architecture and communication interfaces for the simultaneous collection of data from two IMUs and two SONARs, however, parallel acquisition from a larger variety of sensors is also feasible. Representative human gait patterns are experimentally acquired to demonstrate the functionality of the testbed.
2019 IEEE International Symposium on Inertial Sensors and Systems (INERTIAL), 2019
We present a ranging-based aiding method for pedestrian inertial navigation, utilizing not only r... more We present a ranging-based aiding method for pedestrian inertial navigation, utilizing not only ranging readouts, but also orientation of the ranging sensors. Both numerical and experimental results demonstrated improvements in navigation accuracy using the method. Kalman Filter (KF) was implemented to merge inertial navigation with Zero-Velocity-Update (ZUPT) algorithm and foot-to-foot directional ranging information. The improvement of navigation accuracy was achieved purely algorithmically without any increase in complexity of the hardware. We demonstrated experimentally that the navigation errors can be improved by about two times using the method.
2020 DGON Inertial Sensors and Systems (ISS), 2020
In this paper, we proposed a novel zero velocity detector, the Dynamic-Vision-Sensor (DVS)-aided ... more In this paper, we proposed a novel zero velocity detector, the Dynamic-Vision-Sensor (DVS)-aided Stance Phase Optimal dEtection (SHOE) detector, for Zero-velocity-UPdaTe (ZUPT)-aided Inertial Navigation Systems (INS) augmented by a foot-mounted event-based camera DVS128. We observed that the firing rate of the DVS consistently increased during the swing phase and decreased during the stance phase in indoor walking experiments. We experimentally determined that the optimal placement configuration for zero-velocity detection is to mount the DVS next to an Inertial Measurement Unit (IMU) and face the sensor outward. The DVS-SHOE detector was derived in a General Likelihood Ratio Test (GLRT) framework, combining statistics of the conventional SHOE detector and the DVS firing rate. This paper used two methods to evaluate the proposed DVS-SHOE detector. First, we compared the detection performances of the SHOE detector and the DVS-SHOE detector. The experimental results showed that the DVS-SHOE detector achieved a lower false alarm rate than the SHOE detector. Second, we compared the navigation performance of the ZUPT-aided INS using the SHOE detector and the DVS detector. The experimental results showed that the Circular Error Probable (CEP) of the case using DVS-SHOE was reduced by around 25 % from 1.2 m to 0.9 m, as compared to the case of the SHOE detector.
This paper reports on the design and implementation of high efficiency, nonmetallic reflectors in... more This paper reports on the design and implementation of high efficiency, nonmetallic reflectors integrated on the sidewalls of micromachined cavities. Due to shadowing from deposition within a cavity, significant variation in the thicknesses of the dielectric thin films composing the reflectors are encountered when the layers are deposited by Plasma Enhanced Chemical Vapor Deposition (PECVD). These gradients in thickness limit the performance of the reflector at the intended design wavelength. An optimized design procedure is described to maximize the performance of the reflector at the D 1 absorption wavelength of 87 Rb of 795 nm for use in micromachined atomic vapor cells. The reflector design is based on multiple shifted quarter wave Bragg reflectors in series, which extends the reflective bandwidth for increased robustness to film thickness gradients. The extended reflectance range maintains high reflection at the design wavelength despite greater than 70% decrease in film thickness across the reflector surface. The reflector technology is ideally suited for use in atomic MEMS vapor cell applications by achieving high reflectance while maintaining light polarization. We demonstrate less than 2 dB of return loss with circular polarization ellipticity maintained to ±2 • .
2016 Solid-State, Actuators, and Microsystems Workshop Technical Digest, 2016
This paper presents experimental results on the noise performance of an ultra high quality factor... more This paper presents experimental results on the noise performance of an ultra high quality factor (Q-factor) micromachined silicon Quad Mass Gyroscope (QMG). The device was vacuum sealed at below 0.1mTorr in an LCC package with activated getters, demonstrating the Q-factor better than 1.7 million, along the drive and sense modes, at the 1.6 kHz center frequency. A bias stability of 0.09 deg/hr and Angle Random Walk (ARW) of 0.015 deg/rt-hr were demonstrated, without any thermal compensation or post-processing of data. To the best of our knowledge, the result presented in this paper is the highest reported Q-factor for MEMS silicon gyroscopes, now demonstrating a true potential to achieve the navigation-grade performance.
2020 IEEE/ION Position, Location and Navigation Symposium (PLANS), 2020
In this paper, we investigate how self-contained pedestrian navigation can be augmented by the us... more In this paper, we investigate how self-contained pedestrian navigation can be augmented by the use of foot-tofoot visual observations. The main contribution is a measurement model that uses Zero velocity UpdaTe (ZUPT) and relative position measurements between the two shoes obtained from shoe-mounted feature patterns and cameras. This measurement model provides directly the compensation measurements for the three position states and three velocity states of a pedestrian. The involved features for detection are independent of surrounding environments, thus, the proposed system has a constant computational complexity in any context. The performance of the proposed system was compared to a standalone ZUPT method and a relative-distance-aided ZUPT method. Simulation results showed an improvement in accumulated navigation errors by over 90%. Real-world experiments were conducted, exhibiting a maximum improvement of 85% in accumulated errors, verifying validity of the approach.
56th Electronic Components and Technology Conference 2006
Large 6mm x 9mm silicon dice have been successfully bonded on alumina substrate with electroplate... more Large 6mm x 9mm silicon dice have been successfully bonded on alumina substrate with electroplated Au80Sn20 eutectic alloy. Eutectic AuSn is one of the best known hard solders having excellent fatigue-resistance and mechanical properties. A fluxless bonding process in 50militorrs of vacuum environment is presented. Vacuum environment is employed to prevent tin oxidation during the process. The oxygen content is expected to be reduced by a factor of 15,200, comparing to bonding in air. One of the challenges in silicon-to-alumina bonding is the large mismatch in thermal expansion between silicon of 2.7x10-6 ppm/°C and alumina of 7x10-6 ppm/°C. Electroplating method is used to build multi-layer solder. It is an economical alternative to vacuum deposition method and can produce thick solders. Joints fabricated are examined using Scanning Electron Microscope (SEM), and Energy Dispersive X-ray Spectroscopy (EDX). It is found that proper bonding condition is needed to turn the stacked layers into a uniform AuSn eutectic alloy. Nearly void-free joints are achieved and confirmed by a Scanning Acoustic Microscope (SAM). To evaluate the reliability of the solder joint and the bonded structure, samples will go through thermal cycling test to determine failure modes. Microstructural changes of the solder joints during thermal cycling test will also be investigated.
Proceedings of the 33rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2020), 2020
He is currently pursuing the Ph.D. degree in mechanical engineering in University of California I... more He is currently pursuing the Ph.D. degree in mechanical engineering in University of California Irvine. His research interests include design, fabrication, and control of vibratory inertial sensors and capacitive micromachined ultrasonic transducers (CMUT). Andrei M. Shkel received the Diploma (Hons.
2020 IEEE/ION Position, Location and Navigation Symposium (PLANS), 2020
We present an analytical expression for estimating the position error in the vertical direction i... more We present an analytical expression for estimating the position error in the vertical direction in pedestrian navigation using an Inertial Measurement Unit (IMU) aided by an altimeter. The analytical expression is found to be directly proportional to the altimeter resolution, the square root of the altimeter sampling rate, IMU Velocity Random Walk (VRW), and a ratio of time intervals in the stance phase and the swing phase during the gait cycle, and inversely proportional to the square root of IMU sampling rate. The analytical analysis was verified by numerical simulations and supported by experimental results. The numerical simulation of the ZUPT-augmented navigation aided by an altimeter was conducted based on a mathematically simulated foot motion trajectory. Experiments were implemented on a foot-mounted platform containing IMUs and altimeters. The results from both simulations and experiments matched our analytical prediction, with an error less than 20%.
A signal-sensor-based indoor pedestrian navigation system is developed. The proposed system: (i) ... more A signal-sensor-based indoor pedestrian navigation system is developed. The proposed system: (i) utilizes a foot-mounted inertial navigation system (INS), in which the accumulated errors are mitigated via a zero velocity update (ZUPT) approach and (ii) exploits opportunistically cellular long-term evolution (LTE) signals in a deep neural network (DNN)-based synthetic aperture navigation (SAN) framework, in which the pedestrian's motion is utilized to suppress multipath-induced errors. The proposed DNN-SAN-LTE-ZUPT-INS (DUALS) indoor pedestrian navigation system utilizes the complementary the desirable characteristics of both subsystems, coupled via two architectures: (a) loosely-coupled and (b) tightly-coupled. This paper designs and assesses both architectures experimentally in an indoor environment. The experimental study demonstrates a pedestrian traversing a trajectory of 600 m in 14 minutes, including a stationary period, straight segments, up/down the stairs, and riding in an elevator, while receiving signals from 4 LTE base stations (also known as evolved node B (eNodeB)). The proposed tightly-coupled DUALS system exhibited a 3-D position root mean-squared error (RMSE) of 1.34 m, outperforming the loosely-coupled DUALS, ZUPT-aided INS, and LTE-DNN-SAN which achieved a position RMSE of 1.38, 1.49, and 1.97 m, respectively.
In this paper, we propose a Foot-Instability-Based Adaptive (FIBA) covariance to dynamically adju... more In this paper, we propose a Foot-Instability-Based Adaptive (FIBA) covariance to dynamically adjust the covariance matrix for the pseudo-zero-velocity measurements in the Zero velocity UPdaTe (ZUPT)-aided Inertial Navigation Systems (INS). The proposed ZUPT-aided INS using the FIBA covariance is implemented in an Adaptive Extended Kalman Filter (AEKF) framework, where the measurement covariance matrix is updated in each iteration according to the FIBA covariance. The FIBA covariance is designed to have a very high value during the swing phases in a gait cycle, and the value significantly decreases during the stance phases. As a result, the proposed method eliminates a need to use a binary stance phase detector in implementation of the ZUPT-aided INS. Two series of indoor pedestrian navigation experiments were conducted to investigate the navigation performance of the algorithm. In the first series of experiments, which included cases of walking and running, localization solutions produced by the system using the FIBA covariance demonstrated 36% and 64% improvements in navigation accuracy along the horizontal and vertical directions, respectively. In the second series of experiments, which included a pedestrian walking on different indoor terrains, such as flat planes, stairs, and ramps, the navigation accuracy of the system using the FIBA covariance reduced horizontal and vertical position errors by 12% and 45%, respectively, as compared to the conventional ZUPT-aided INS.
Proceedings of 1995 34th IEEE Conference on Decision and Control
Control schemes in real-time sensor-based systems often operate under tight time constraints dete... more Control schemes in real-time sensor-based systems often operate under tight time constraints determined by the system sampling rate. One area where uch constraints are especially severe is the sensorbased motion planning with dynamics in robotics. Though very important both theoretically and in practice, this problem has not been addressed so far. The typical sampling rates in such systems are 20 to 50 %>er second. This leaves only 20 to 50 msec for the whole cycle, including sensing, complex geometric (intelligence) analysis, calculations due to dynamics and control, and motion execution. As a Arst attempt to solve the problem, we show that the time constraints an be met by a combination of a simple model with a carefully chosen analytical solution of the dynamic equations. The resulting control scheme guarantees Convergence and safety of motion, and blends well with kinematic path planning algorithms. Two such strategies are discussed, one using a simple heuristic and the other with local optimization.
Uploads
Papers by Andrei Shkel