Showing papers on "Inertial measurement unit published in 2004"

Design of multi-sensor attitude determination systems

Abstract: The design of inexpensive multi-sensor attitude determination systems is discussed. The systems discussed fuse information from a triad of solid state rate gyros with an aiding system mechanized using GPS or magnetometers and accelerometers. Euler angle and quaternion-based sensor fusion algorithms are developed. Methods for gain scheduling and estimator pole placement are presented. Using simulation and flight test results, it is shown that quaternion-based algorithms simplify gain scheduling and improve transient response.

Integrating Inertial Sensors With Global Positioning System (GPS) for Vehicle Dynamics Control

Abstract: This paper demonstrates a method of estimating several key vehicle states-sideslip angle, longitudinal velocity, roll and grade-by combining automotive grade inertial sensors with a Global Positioning System (GPS) receiver. Kinematic Kalman filters that are independent of uncertain vehicle parameters integrate the inertial sensors with GPS to provide high update estimates of the vehicle states and the sensor biases. Using a two-antenna GPS system, the effects of pitch and roll on the measurements can be quantified and are demonstrated to be quite significant in sideslip angle estimation. Employing the same GPS system as an input to the estimator, this paper develops a method that compensates for roll and pitch effects to improve the accuracy of the vehicle state and sensor bias estimates. In addition, calibration procedures for the sensitivity and cross-coupling of inertial sensors are provided to further reduce measurement error The resulting state estimates compare well to the results from calibrated models and Kalman filter predictions and are clean enough to use in vehicle dynamics control systems without additional filtering.

Vision-based road-following using a small autonomous aircraft

Abstract: This paper describes the vision-based control of a small autonomous aircraft following a road. The computer vision system detects natural features of the scene and tracks the roadway in order to determine relative yaw and lateral displacement between the aircraft and the road. Using only the vision measurements and onboard inertial sensors, a control strategy stabilizes the aircraft and follows the road. The road detection and aircraft control strategies have been verified by hardware in the loop (HIL) simulations over long stretches (several kilometers) of straight roads and in conditions of up to 5 m/s of prevailing wind. Hardware experiments have also been conducted using a modified radio-controlled aircraft. Successful road following was demonstrated over an airfield runway under variable lighting and wind conditions. The development of vision-based control strategies for unmanned aerial vehicles (UAVs), such as the ones presented here, enables complex autonomous missions in environments where typical navigation sensor like GPS are unavailable.

Adaptive Kalman filtering algorithms for integrating GPS and low cost INS

Abstract: GPS and Inertial Navigation Systems are used for positioning and attitude determination in a wide range of applications. Over the last few years, a number of low cost inertial sensors have become available. Although they exhibit large errors, GPS measurements can be used correct the INS and sensor errors to provide high accuracy real-time navigation. The integration of GPS and INS measurements is usually achieved using a Kalman filter. The measurement and process noise matrices used in the Kalman filter represent the stochastic properties of the GPS and INS systems respectively. Traditionally they are defined a priori and remain constant throughout a processing run. In reality, the stochastic properties of the system vary depending on factors such as vehicle dynamics and environmental conditions. This is particularly an issue for low cost inertial sensors where the initial sensor errors can be large, and experience significant temporal variation. This paper investigates three adaptive Kalman filtering algorithms that can be used to improve the estimation of the stochastic properties of a low cost INS. The algorithms are tested using a low cost Crossbow MEMS IMU integrated with carrier phase GPS for a marine application. The adaptive Kalman filtering algorithms are shown to reduce the dependence on the a priori information used in the filter. This results in a reduction in the time required to initialise the sensor errors and align the INS, and results in an improvement in navigation performance.

Wavelet de-noising for IMU alignment

Abstract: Inertial navigation system (INS) is presently used in several applications related to aerospace systems and land vehicle navigation. An INS determines the position, velocity, and attitude of a moving platform by processing the accelerations and angular velocity measurements of an inertial measurement unit (IMU). Accurate estimation of the initial attitude angles of an IMU is essential to ensure precise determination of the position and attitude of the moving platform. These initial attitude angles are usually estimated using alignment techniques. Due to the relatively low signal-to-noise ratio of the sensor measurement (especially for the gyroscopes), the initial attitude angles may not be computed accurately enough. In addition, the estimated initial attitude angles may have relatively large uncertainties that may affect the accuracy of other navigation parameters. This article suggests processing the gyro and accelerometer measurements with multiple levels of wavelet decomposition to remove the high frequency noise components. The proposed wavelet de-noising method was applied on a navigational grade inertial measurement unit (LTN90-100). The results showed that accurate alignment procedure and fast convergence of the estimation algorithm, in addition to reducing the estimation covariance of the three attitude angles, could be obtained.

Global Positioning System (GPS): A Low-Cost Velocity Sensor for Correcting Inertial Sensor Errors on Ground Vehicles

Abstract: This paper demonstrates the ability of a standard low-cost Global Positioning System (GPS) receiver to reduce errors inherent in low-cost accelerometers and rate gyroscopes used on ground vehicles. Specifically GPS velocity is used to obtain vehicle course, velocity, and road grade, as well as to correct inertial sensors errors, providing accurate longitudinal and lateral acceleration, and pitch, roll, and yaw angular velocities. Additionally, it is shown that transient changes in sideslip (or lateral velocity), roll, and pitch angles can be measured. The method utilizes GPS velocity measurements to determine the inertial sensor errors using a kinematic Kalman Filter estimator Simple models of the inertial sensors, which take into account the sensor noise and bias drift properties, are developed and used to design the estimator. Based on the characteristics of low-cost GPS receivers and IMU sensors, this paper presents the achievable performance of the combined system using the covariance analysis from the Kalman filter. Subsequent simulations and experiments validate both the error analysis and the methodology for utilizing GPS as a velocity sensor for correcting low-cost inertial sensor errors and providing critical vehicle state measurements.

FlightTracker: A Novel Optical/Inertial Tracker for Cockpit Enhanced Vision

Abstract: One of the earliest fielded augmented reality applications was enhanced vision for pilots, in which a display projected on the pilot's visor provides geo-spatially registered information to help the pilot navigate, avoid obstacles, maintain situational awareness in reduced visibility, and interact with avionics instruments without looking down. This requires exceptionally robust and accurate head-tracking, for which there is not a sufficient solution yet available. In this paper, we apply miniature MEMS sensors to cockpit helmet-tracking for enhanced/synthetic vision by implementing algorithms for differential inertial tracking between helmet-mounted and aircraft-mounted inertial sensors, and novel optical drift correction techniques. By fusing low-rate inside-out and outside-in optical measurements with high-rate inertial data, we achieve millimeter position accuracy and milliradian angular accuracy, low-latency and high robustness using small and inexpensive sensors.

An unscented Kalman filter for in-motion alignment of low-cost IMUs

Abstract: This paper describes the alignment of low-cost inertial measurement units (IMUs) using an unscented Kalman filter (UKF), which allows large initial attitude error uncertainties. The state vector includes position, velocity, attitude, and sensor biases and scale factors. Position information from the differential global positioning system (DGPS) solutions is used as measurements. Test results with a micro-electrical-mechanical-systems (MEMS) IMU showed that the alignment converged within 50 s with RMS values of 0.093/spl deg/, 0.094/spl deg/ and 0.388/spl deg/ for roll, pitch and heading, respectively. The UKF works well even in cases of large initial attitude errors (about 30/spl deg/) not only for heading but also for roll and pitch. Therefore, the UKF is a unified approach to handle large and small attitude errors of an inertial navigation system (INS) seamlessly.

Magic wand: a hand-drawn gesture input device in 3-D space with inertial sensors

Abstract: This paper presents a gesture input device, magic wand, with which a user can input gestures in 3-D space, inertial sensors embedded in it generate acceleration and angular velocity signals according to a user's hand movement. A trajectory estimation algorithm is employed to convert them into a trajectory on 2-D plane. The recognition algorithm based on Bayesian networks finds the gesture model with the maximum likelihood from it. The recognition performance of the proposed system is quite promising; the writer-independent recognition rate was 99.2% on average for the database of 15 writers and 13 gesture classes.

Motion capture from inertial sensing for untethered humanoid teleoperation

Abstract: We describe the design of a modular system for untethered real-time kinematic motion capture using sensors with inertial measuring units (IMU) Our system is comprised of a set of small and lightweight sensors Each sensor provides its own global orientation (3 degrees of freedom) and is physically and computationally independent, requiring only external communication Orientation information from sensors is communicated via wireless to host computer for processing We present results of the real-time usage of our untethered motion capture system for teleoperating the NASA Robonaut We also discuss potential applications for untethered motion capture with respect to humanoid robotics

Performance Test Results of an Integrated GPS/MEMS Inertial Navigation Package

Abstract: This paper describes the design, operation and performance test results of a miniature, low cost integrated GPS/inertial navigation system (INS) designed for use in UAV or UGV guidance systems. The system integrates a miniaturized commercial GPS with a low grade Micro-Electro-Mechanical (MEMS) inertial measurement unit (IMU). The MEMS IMU is a small self-contained package (< 1 cu inch) and includes a triad of accelerometers and gyroscopes with additional sensors integrated for temperature compensation and baro pressure altitude aiding. The raw IMU data is provided through a serial interface to a processor board where the inertial navigation solution and integrated GPS/inertial Kalman filter is generated. The GPS/Inertial software integration is performed using NAVSYS’ modular InterNav software product. This allows integration with different low cost GPS chips sets or receivers and also allows the integrated GPS/inertial navigation solution to be embedded as an application on a customer’s host computer. This modular, object oriented architecture facilitates integration of the miniature MEMS GPS/INS navigation system for embedded navigation applications. Test results are presented in this paper showing the performance of the integrated MEMS GPS/INS navigation system when used to perform guidance for a small Unmanned Ground Vehicle (UGV). Data is provided showing the position, velocity and attitude accuracy when operating with GPS aiding and also for periods where GPS dropouts occur and alternative navigation update sources are used to bound the MEMS inertial navigation error growth.

A quaternion-based orientation estimation algorithm using an inertial measurement unit

Abstract: This paper presents a real-time orientation estimation algorithm based on signals from a low-cost inertial measurement unit (IMU). The IMU consists of three MEMS accelerometers and three MEMS rate gyros. This approach is based on relationships between the quaternion representing the platform orientation, the measurement of gravity from the accelerometers, and the angular rate measurement from the gyros. Process and measurement models are developed, based on these relations, in order to implement them into an extended Kalman filter. The performance of each filter is evaluated in terms of the roll, pitch, and yaw angles. These are derived from the filter output since this orientation representation is more intuitive than the quaternion representation. Extensive testing of the filters with simulated and experimental data show that the filters perform very accurately in the roll and pitch angles, and even significantly corrects the yaw angle error drift.

Initial calibration of an inertial measurement unit using an optical position tracking system

Abstract: A reliable calibration procedure of a standard six degree-of-freedom inertial measurement unit (IMU) is presented. Mathematical models are derived for the three accelerometers and three rate gyros, taking into account the sensor axis misalignments, accelerometer offsets, electrical gains, and biases inherent in the manufacture of an IMU. The inertial sensors are calibrated using data from a 3D optical tracking system that measures the position coordinates of markers attached to the IMU. Inertial sensor signals and optical tracking data are obtained by manually moving the IMU. Using vector methods, the quaternion corresponding to the IMU platform orientation is obtained, along with its acceleration, velocity, and position. Given this kinematics information, the sensor models are used in a nonlinear least squares algorithm to solve for the unknown calibration parameters. The calibration procedure is verified through extensive experimentation.

An Inertial and Visual Sensing System for a Small Autonomous Helicopter

Abstract: This paper describes the design and architecture of a low-cost and light-weight inertial and visual sensing system for a small-scale autonomous helicopter. A custom 6-axis IMU and a stereo vision system provide vehicle attitude, height, and velocity information. We discuss issues such as robust visual processing, motion resolution, dynamic range, and sensitivity. © 2004 Wiley Periodicals, Inc.

A car test for the estimation of GPS/INS alignment errors

Abstract: Misalignment can be an important error source in the integration of the global positioning system (GPS) and inertial navigation systems. This paper presents car test results on the estimation of alignment errors in the integration of a low-grade inertial measurement unit (IMU) with accurate GPS measurement systems. The car test was conducted with a low-cost solid-state IMU and carrier-phase differential GPS measurement systems. Test results showed that changes in the angular velocity improve the estimation of the lever arm between the GPS antenna and IMU. They also showed that changes in acceleration improve the estimation of the relative attitude between the GPS antenna array and IMU. The lever arm was estimated with a 10-cm error. The relative attitude was estimated with a half-degree error. An iterative scheme was used to improve the estimation of the alignment errors during postprocessing. The scheme was shown to be useful when the test car could not have sufficient changes in motion due to limitations in its path. With the given set of test data, the estimation error decreased as the number of iterations increased.

Adaptive filter for a miniature MEMS based attitude and heading reference system

Abstract: A strapdown Inertial Navigation System (INS) can provide attitude and heading estimates after initialization and alignment. Many factors affect the accuracy and the performance of the system. They mainly are: sensor noise, bias, scale factor error, and alignment error. The Inertial Measurement Unit (IMU) based on the newly developed MEMS technology has wide applications due to its low-cost, small size, and low power consumption. However, the inertial MEMS sensors have large noise, bias and scale factor errors due to drift. The traditional strapdown algorithm using a low-cost MEMS sensor ONLY is difficultly satisfying the attitude and heading performance requirements. An extended Kalman filter with adaptive gain was used to build a miniature attitude and heading reference system based on a stochastic model. The adaptive filter has six states with a time variable transition matrix. The six states are three tilt angles of attitude and three bias errors for the gyroscopes. The filter uses the measurements of three accelerometers and a magnetic compass to drive the state update. When the system is in the non-acceleration mode, the accelerometer measurements of the gravity and the compass measurements of the heading have observability and yield good estimates of the states. When the system is in the high dynamic mode and the bias has converged to an accurate estimate, the attitude calculation will be maintained for a long interval of time. The adaptive filter tunes its gain automatically based on the system dynamics sensed by the accelerometers to yield optimal performance. The paper presents the methodology of the technique, performs the analysis, and gives the testing results of the system based on the adaptive filter. The whole system can be fitted within the size of 5cm /spl times/ 5cm /spl times/ 5cm with analog to digital conversion and digital signal processing boards.

Modeling Inertial Sensor Errors Using Autoregressive (AR) Models

Abstract: The accelerometer and gyro sensor errors of a Strapdown Inertial Navigation System (SINS) consist of two parts: a deterministic part and a random part. The deterministic part includes biases and scale factors, which are determined by calibration and then removed from the raw measurements. The random part is correlated over time and is basically due to the variations in the SINS sensor bias terms. Therefore, these errors are modeled stochastically so that they can be included in the SINS error model. For most of the navigation-grade SINS systems, a first-order Gauss-Markov model with a fairly large correlation time is usually used to describe the random errors associated with inertial sensors. By studying the autocorrelation sequences of the noise components at the outputs of inertial sensors, we have determined that a first-order Gauss-Markov is not adequate to model such noise behavior. This paper offers a new method to model the inertial sensor noise as a higher order autoregressive (AR) process and adaptively estimates the AR model parameters. Three different methods of determining the AR model parameters are investigated, namely: the Yule-Walker (autocorrelation) method, the covariance method and the Burg’s method. The three algorithms were tested with different AR model orders with real SINS static data. The results showed that the Burg’s method gives the minimum prediction mean square estimation error and hence this method was used for generating the applied AR models. In this paper, the current algorithms for modeling inertial sensor errors are introduced. The different algorithms for AR modeling are discussed and their results are presented. The results after applying the Burg’s AR models with different orders are also analyzed.

Tightly Coupled INS/GPS with Bias Estimation for UAV Applications

Abstract: This paper presents an analysis of a proposed tightly coupled Inertial/GPS navigation system for UAV applications. An indirect, error state extended Kalman filter is employed to estimate vehicle position, velocity, attitude and IMU bias errors. The indirect configuration linearises the state transition matrix and greatly reduces the required feedback frequency. Standard GPS C/A code pseudo-ranges are used directly to update the Kalman filter. The advantage of this configuration arises in situations of poor GPS availability. Traditional loosely coupled filters can deliver no new information to the filter when the observed satellite number falls below four. In the tightly coupled configuration, all available information is delivered to the filter even in situations where only one satellite remains observable. The extension of the vehicle states to include IMU biases further improves navigation accuracy by constraining drift during the INS alone cycle and in periods of low GPS observability. Both algorithms are suited to a low cost implementation. Results from flight trials of the Brumby Mk. III UAV are presented.

INS/GPS integration using neural networks for land vehicular navigation applications

Abstract: Most of the present navigation sensor integration techniques are based on Kalman filtering estimation procedures. Although Kalman filtering represents one of the best solutions for multi-sensor integration, it has some drawbacks in terms of stability, computation load, immunity to noise effects and observability. Furthermore, Kalman filters perform adequately only under certain predefined dynamic models. Neuron computing, a technology of Artificial Neural Network (ANN), is a powerful tool for solving nonlinear problems that involve mapping input data to output data without having any prior knowledge about the mathematical process involved. This article suggests a multi-sensor integration approach for fusing data from an Inertial Navigation system (INS) and Differential Global Positioning System (DGPS) utilizing multi-layer feed-forward neural networks with a back propagation learning algorithm. The performance of the proposed architecture was tested using two different INS systems(tactical grade IMU and navigation grade IMU) in a land vehicle.

Multi-rate fusion with vision and inertial sensors

Abstract: This work presents a multi-rate fusion model, which exploits the complimentary properties of visual and inertial sensors for egomotion estimation in applications such as robot navigation and augmented reality. The sampling of these two sensors is described with size-varying input and output equations without assumed synchronicity and periodicity of measurements. Data fusion is performed with two different multi-rate (MR) filter models, an extended (EKF) and an unscented Kalman filter (UKF). A complete dynamic model for the 6D-tracking task is given together with a method to calculate the dependencies of the covariance matrices. It is further shown that a centripetal acceleration model and the precise description of quaternion prediction for a constant velocity model highly improve the estimation error for rotary motions. The comparison demonstrates that the MR-UKF provides better estimation results at higher computational costs.

Vehicle Model Aided Inertial Navigation For A Uav Using Low-Cost Sensors

Abstract: This paper considers two different methods of using a dynamic vehicle model in order to aid pose estimates provided by an Inertial Navigation System (INS) for a Un-manned Aerial Vehicle (UAV). We consider low-cost inertial sensors in which errors from sensor noise, bias and scale-factor errors cause a significant growth in pose estimate errors when the navigation system is un-aided by external positioning information such as from GPS or terrain observations. It is shown that an accurate dynamic vehicle model can considerably reduce the error growth in vehicle pose estimates by enhancing the observability of the inertial sensor error sources. Furthermore we analyse the effectiveness of using the vehicle model information given that there are errors in the vehicle model.

An inertial and visual sensing system for a small autonomous helicopter

Abstract: This paper describes the design and architecture of a low-cost and light-weight inertial and visual sensing system for a small-scale autonomous helicopter. A custom 6-axis IMU and a stereo vision system provide vehicle attitude, height, and velocity information. We discuss issues such as robust visual processing, motion resolution, dynamic range, and sensitivity. © 2004 Wiley Periodicals, Inc.

Benefits of Using a Tactical-Grade IMU for High-Accuracy Positioning

Abstract: Integration of GPS with inertial sensors can provide many benefits for navigation, from improved accuracy to increased reliability. The extent of such benefits, however, is typically a function of the quality of the inertial system used. Traditionally, high-cost, navigation-grade inertial measurement units (IMUs) have been used to obtain the highest position and velocity accuracies. However, the work documented in this paper uses a Honeywell HG-1700 IMU (1 deg/h) to assess the benefits of a tactical-grade IMU in aiding GPS for high-accuracy (centimeter-level) applications. To this end, the position and velocity accuracy of the integrated system during complete and partial GPS data outages is investigated. The benefit of using inertial data to improve the ambiguity resolution process after such data outages is also addressed in detail. Centralized and decentralized filtering strategies are compared in terms of system performance.

Kalman filtering for real-time orientation tracking of handheld microsurgical instrument

Abstract: This paper presents the theory and modeling of a quaternion-based augmented state Kalman filter for real-time orientation tracking of a handheld microsurgical instrument equipped with a magnetometer-aided all-accelerometer inertial measurement unit (IMU). The onboard sensing system provides two complementary sources of orientation information. The all-accelerometer IMU provides a high resolution but drifting angular velocity estimate, while the magnetic north vector is combined with the estimated gravity vector to yield a non-drifting but noisy orientation estimate. Analysis of the dominant stochastic noise components of the sensors and derivation of the noise covariance are presented. The proposed Kalman filter obtains a non-drifting orientation estimate with improved resolution by incorporating the motion dynamics of the instrument during microsurgery and models the angular velocity drift explicitly as extra dynamic states.

Error analysis and stochastic modeling of MEMS based inertial sensors for land vehicle navigation applications

Abstract: Although GPS measurements are the essential information for currently developed land vehicle navigation systems (LVNS), the situation when GPS signals are unavailable or unreliable due to signal blockages must be compensated to provide continuous navigation solutions. In order to overcome the unavailability or unreliability problem in satellite based navigation systems and also to be cost effective, Micro Electro Mechanical Systems (MEMS) based inertial sensor technology has pushed the development of lowcost integrated navigation systems for land vehicle navigation and guidance applications. In spite of low inherent cost, small size, low power consumption, and solid reliability of MEMS based inertial sensors, the errors in the observations from the MEMS-based sensors must be appropriately treated in order to turn the observations into useful data for vehicle position determination. The error analysis would be conducted in the time domain specifying the stochastic variation as well as error sources of systematic nature. This thesis will address the above issues and present algorithms to identify and model the error sources in MEMS-based inertial sensors. A Kalman filter will be described and applied to analyze the performance of a minimum configured GPS/IMU system for vehicle navigation applications. The performance of the testing system has been assessed via a comparison to Precise Point Position (PPP) reference data. The testing results indicate the effectiveness of the discussed error analysis and modeling method. iii

Inertial augmentation for GPS navigation on ground vehicles

Abstract: An inertial compensation assembly suitable for use in a GPS based navigation system for a ground vehicle, in particular, an agricultural ground vehicle such as a tractor, combine, or the like, provides inertial augmentation to compensate GPS navigation information such as position, course, and track spacing for errors caused by variation of ground vehicle attitude (i.e., roll and yaw) over non-level terrain.

Implementation of Deeply Integrated GPS/Low-Cost IMU for Reacquisition and Tracking of Low CNR GPS Signals

Abstract: A deeply integrated GPS/IMU system is developed to acquire and continuously track low Carrier-to-Noise Ratio (CNR) GPS signals, while maintaining carrier tracking performance at the mm-level Reacquisition and tracking are demonstrated for 15 dB-Hz GPS signals without receivers for a number of applications (eg indoor applications or navigation in the presence of wideband interference) The deeply integrated GPS/IMU employs inertial aiding of the GPS signal integration to significantly increase the integration time interval as compared to conventional unaided receivers As a result, requiring knowledge of navigation data bits Initial, lowdynamic test results demonstrate the feasibility of the deeply integrated GPS/IMU system