Search

This paper presents the first part of a study aiming at error state selection in Kalman filters applied to the stationary self-alignment and calibration (SSAC) problem of strapdown inertial navigation systems (SINS). Estimation algorithms are derived through the analytical manipulation of the full SINS error model, thereby enabling us to investigate the dynamic coupling existing between the state variables. As contributions of this work, we demonstrate that the vertical velocity error is very important for the estimation of almost all error states. Latitude and altitude errors, in turn, are shown to uniquely affect the inertial sensor bias estimates. Besides, the longitude error is found to be totally detached from the system. As straightforward consequence, Bar-Itzhack and Berman's error model turns out to be inadequate for real implementations, and a 12-state Kalman filter is shown to be the optimal error state selection for SSAC purposes. Simulated and experimental tests confirm the adequacy of the outlined conclusions.

Low cost inertial measurement units (IMU) are comprised of micro-electro-mechanical systems (MEMS) gyroscopes and accelerometers. These inertial sensors have high random noise and time varying bias. Hence, an accurate stochastic error model is necessary to predict performance and also to implement an attitude and heading reference system or a navigator. The parameters of this stochastic model are classically obtained via Allan Variance analysis. In this paper, two modern approaches, based on autoregressive moving average model and Kalman Filter, respectively, are investigated and their performances are unveiled via sensitivity analysis, realistic simulations with typical MEMS parameters and also with experimental data. Basic equations for the estimation problem are developed and the solution sensitivity to data and parameters is discussed. It is shown that the deficiency with the online methods can be traced back to the difficulty in estimating the parameters B (bias instability) and K (rate random walk for gyros, acceleration random walk for accelerometers) separately under high measurement noise N (angular random walk for gyros, velocity random walk for accelerometers), which is typical of MEMS IMU.

This paper presents the design of a complete multi-rate self-alignment algorithm for strapdown inertial navigation systems (SINS), readily amenable to be implemented in a digital computer. The proposed algorithm is divided in two steps, the coarse and the fine self-alignments. As main contribution of this paper, one demonstrates that, for the particular case of stationary SINS, the fine alignment is not able to improve the orientation estimates derived from the coarse alignment. Instead, the main role of the fine alignment consists in estimating some of the uncompensated inertial sensor biases, which may be used for compensation into the navigation algorithms, greatly improving the accuracy of autonomous inertial navigators.

This paper presents an adaptive algorithm to synchronize a hyperchaotic finance system in the presence of unknown system parameters and bounded disturbances. Based on Lyapunov-like analysis, an adaptive scheme is proposed to make the synchronization error asymptotically null. Simulation results are provided to demonstrate the effectiveness and feasibility of the proposed synchronization method.

This paper implements and evaluates experimentally a procedure for automatically georeferencing images acquired by unmanned aerial vehicles (UAV's) in the sense that ground control points (GCP) are not necessary. Since the camera model is necessary for georeferencing, this paper also proposes a completely automatic procedure for collecting corner pixels in the model plane image to solve the camera calibration problem, i.e., to estimate the camera and the lens distortion parameters. The performance of the complete georeferencing system is evaluated with real flight data obtained by a typical UAV.

A novel Temporal Motion Vector Filter (TF) is presented and evaluated for real-time object detection on compressed videos in MPEG-2, MPEG-4 or H.264/AVC formats. The filter significantly reduces the noisy motion vectors that do not represent a real object movement . The filter analyses the temporal coherence of block motion vectors to determine if they are likely to represent true motion in the recorded scene. Experiments are performed using the CLEAR metrics for object detection and public available video datasets from CAVIAR, PETS and CLEAR. These experiments demonstrate that the TF outperforms the Vector Median Filter, by providing better object detection accuracy with reduced computational complexity. The good results obtained by the TF make it suitable as a first step towards implementing systems that aim to detect and track objects from compressed video by using motion vectors. The TF could also be used to improve other techniques based on motion vectors such as Global Motion Estimation (GME) and Motion-Compensated Frame Interpolation (MCFI).

This paper presents the second part of a study aiming at the error state selection in Kalman filters applied to the stationary self-alignment and calibration (SSAC) problem of strapdown inertial navigation systems (SINS). The observability properties of the system are systematically investigated, and the number of unobservable modes is established. Through the analytical manipulation of the full SINS error model, the unobservable modes of the system are determined, and the SSAC error states (except the velocity errors) are proven to be individually unobservable. The estimability of the system is determined through the examination of the major diagonal terms of the covariance matrix and their eigenvalues/eigenvectors. Filter order reduction based on observability analysis is shown to be inadequate, and several misconceptions regarding SSAC observability and estimability deficiencies are removed. As the main contributions of this paper, we demonstrate that, except for the position errors, all error states can be minimally estimated in the SSAC problem and, hence, should not be removed from the filter. Corroborating the conclusions of the first part of this study, a 12-state Kalman filter is found to be the optimal error state selection for SSAC purposes. Results from simulated and experimental tests support the outlined conclusions.

This paper discusses the black-box identification of a real aerospace electrohydraulic actuator. A Hammerstein model structure is adopted due to its simplicity and capacity to represent a wide set of nonlinear phenomena. The identification strategy begins with a frequency-domain analysis so that a polynomial NARX model structure can be defined. Then, an extended least squares procedure is applied to obtain the coefficient values.

Reinforcement learning techniques are based on experience acquisition. In most realistic applications, experience is time-consuming: it implies sensor reading, actuator control and algorithmic update, constrained by the learning system dynamics. In fact, the information crudeness upon which classical learning algorithms operate can make such problems too difficult and unrealistic: embedding of prior structural knowledge, learning of control laws (instead of low level control actions) and full use of experience are required if control of real systems is at stake. In this paper, we show how a robust formulation (with respect to convergence properties) of the Qlearning method that considers prior information about the structure of the state space can be combined with an instance-based technique as a mechanism to accelerate learning. We then demonstrate it both in simulation (with a statistical consistency test) and in a real robot for a guidance task defined by a combination of a predefined control law and learned action policies.

This paper describes a modeling procedure for tandem cold metal rolling, including the linearization step and system identification for control. The tandem cold rolling process is described by a mathematical model based on algebraic equations developed for control purposes and empirical relations. A state-space model is derived and detailed analyses in open loop are presented, concerning the sensitivity with regard to the variations in process parameters and results for the application of a new subspace identification method are compared with classical methodologies. Therefore, this work intents to be a contribution for developments in new control strategies for tandem cold rolling process that offer the potential to reduce the design efforts, the commissioning time and maintenance in rolling mills. The preliminary results obtained with this model have shown reasonable agreement with operational data presented at literature for industrial cold rolling process.

PurposeThe purpose of this paper is to employ an extended Kalman filter for implementing an AHRS (attitude and heading reference system) with acceleration compensation, thereby improving the reliability of such systems, since this removes the usual restrictive assumption that the vehicle is undergoing a non‐accelerated maneuver.Design/methodology/approachMARG (magnetic, acceleration and rate gyros) sensors constitute the basic hardware, which are integrated by the Kalman filter. The error dynamics for attitude and gyro biases is obtained in the navigation frame, providing a much simpler approach than usually taken in the literature, since it relies on direct quaternion differentiation. The state vector associated to the error dynamics possesses six components: three are associated to the quaternion error and three concern gyro bias estimates.FindingsThe AHRS is implemented in an ARM (Advanced RISC Machine) processor and tested with experimental data. The accelerated case is treated by two complementary approaches: by changing the noise variance in the Kalman filter, and by obtaining an acceleration information from GPS (global positioning system) velocity measurements. Experimental results are presented and the performance is compared with commercial ARHS systems.Practical implicationsThe proposed AHRS can be implemented with low cost MARG sensors, and GPS aiding, with use for instance in UAV (unmanned aerial vehicle) and small aircrafts' attitude estimation, for navigation and control applications.Originality/valueUsually the AHRS designs employ as states total gyro bias and Euler angles, or quaternion, and do not consider the accelerated case. Here the state is comprised by gyro bias and quaternion error variables, which attenuates the effect of nonlinearities, and two complementary procedures tackle the accelerated case: acceleration correction by using a GPS derived acceleration signal and change in the output noise covariance used by the Kalman filter.

This work presents a systematic comparison between several detailed chemical kinetic models recently developed and available experimental data. The aim is to assess the predictive capabilities of the combustion with air of the following fuels: hydrogen, methane, ethanol and liquefied petroleum gas, in a large range of equivalence ratio. The prevailing thermodynamical conditions range from ambient to the more stringent ones, such as high pressure combustion. In order to assess the predictive performance of the twelve chosen chemical kinetics models, the results of numerical simulations are compared with existing experimental data of the combustion process in two simplified physical systems: the perfectly stirred reactor and the freely propagating premixed laminar flame. When ambient conditions are considered, the comparisons reveal a good agreement among most of detailed kinetic mechanisms, on the prediction of thermochemical properties of practical interest with respect to the corresponding experimental data only, as far as lighter fuels such as hydrogen, methane and ethanol are considered. The chosen mechanisms are shown to meet with di fficulties when mixtures of liquefied petroleum gas and air are considered, even in ambient conditions. The obtained results highlight the necessity for (i) updating the existing mechanisms with the use of recent experimental results and (ii) the development of new comprehensive models. Keywords: detailed kinetic mechanisms, hydrogen, natural gas, ethanol, liquefied petroleum gas

This work deals with the optimization of ∞ight test maneuvers for aerodynamic parameter estimation considering that the measurements are contaminated with colored residuals. The colored residuals consideration is important to give the direct and realistic assessment of the parameter estimation uncertainty levels prior to ∞ight tests. The optimization technique is based on the concept of ∞ight test data information content and Cramer-Rao lower bound. The discrete autocorrelation matrix of the measurement noise is used in order to compose the optimization criteria considering colored residuals. Some results of a ∞ight test campaign of the CEA-205 CB.9 Curumim aircraft are discussed. The advantages of the proposed maneuvers optimization technique are presented, stressing the easiness of implementation of the signals and the strong improvement in the estimation procedures made possible with the application of the optimized maneuver signals.

State and parameter estimation using flight test data is highly affected by process and measurement noises, especially with noises displaying time varying statistical properties. Hence, if an estimation problem is to be solved, an adaptive filtering approach is recommended. It is also desirable to obtain the estimates online, simultaneously with flight execution, aiming at a maneuver validation before concluding the flight. Indeed, it is more expensive to put the aircraft back in the air than to extend a little the flight and repeat a test point. Flight path reconstruction is a technique which produces a consistent flight test data set from noisy measurements as a preprocessing scheme to a parameter identification routine. Air data can also be calibrated simultaneously if the problem is formulated properly. This work proposes a methodology to deal with time varying noise statistical properties using a new approach for an adaptive extended Kalman filter. Besides the main filter, two other Kalman filters are proposed to run in parallel, to estimate the process and measurement noise statistics based on the main filter residuals. The proposed adaptive method is derived from the covariance matching technique, by employing filter residuals to adjust the noise statistical properties. Because the method has a low computational cost and is recursive, it is suitable for online applications. The method is validated in a flight path reconstruction application, with simultaneous air data calibration for angle of attack, angle of sideslip, and static pressure sensors. A 100 samples Monte Carlo simulation and real flight test data analysis are used for performance evaluation. Because the proposed approach adequately estimates the statistical noise properties, improved performance is obtained.

Autonomous learning techniques are based on experience acquisition. In most realistic applications, experience is time-consuming: it implies sensor reading, actuator control and algorithmic update, constrained by the learning system dynamics. The information crudeness upon which classical learning algorithms operate make such problems too difficult and unrealistic. Nonetheless, additional information for facilitating the learning process ideally should be embedded in such a way that the structural, well-studied characteristics of these fundamental algorithms are maintained. We investigate in this article a more general formulation of the Q-learning method that allows for a spreading of information derived from single updates towards a neighbourhood of the instantly visited state and converges to optimality. We show how this new formulation can be used as a mechanism to safely embed prior knowledge about the structure of the state space, and demonstrate it in a modified implementation of a reinforcement learning algorithm in a real robot navigation task.

Controllers for mobile robots are often designed by taking into account only the kinematics of the robot. The difficulties associated with the dynamic effects such as mass and inertia moments arise from the nonlinearities of the robot model and from the uncertain about these parameters. This paper proposes a controller based on the certainty equivalence principle that considers the dynamic effects on the robot. The proposed controller deals with the nonlinearities by means of feedback linearization, while the unknown parameters are estimated by a modified least squares algorithm. Simulated and experimental results are presented.

This work improves recent results concerning the adaptive control of mobile robots via neural and wavelet networks, in the sense that the stability proof, based on the second method of Lyapunov, encompasses (1) unmodeled dynamics and disturbances in the robot model; (2) adaptation of all parameters in the wavelet networks; and (3) a flexible procedure for automatically adjusting the wavelet architecture. Prior knowledge of dynamic of the mobile robot and network training is not necessary because the controller learns the dynamics online. The wavelet network's parameters and structure are also adapted online. Simulation results are presented by using parameters of the Magellan mobile robot from IS Robotics, Inc.

This work describes the application of the output-error method using the Levenberg-Marquardt optimization algorithm to the Flight Path Reconstruction (FPR) problem, which constitutes an important preliminary step towards the aircraft parameter identification. This method is also applied to obtain the aerodynamic and control derivatives of a regional jet aircraft from flight test data with measurement noise and bias. Experimental results are reported, employing a real jet aircraft, with flight test data acquired by smart probes, inertial sensors (gyrometers and accelerometers) and Global Positioning Systems (GPS) receivers.

This paper investigates the impact that errors in the innovation correlation calculations has upon the steady-state Kalman filter gain identification. This issue arises in all real time applications, where the correlations must he calculated from experimental data. The algorithm proposed by [L. Hong (1991)] is considered and equations describing the impact are established. Simulation results are presented and discussed. Finally, experimental results for the algorithm in [L. Hong (1991)], applied to estimate the states of a servo system, are presented.

A supervised neural classifier based on Fisher criterion is implemented to classify two regions in a real speckled SAR image. Regions around pre-classified pixels are presented to train the neural network that learns a sub-optimal set of masks via back-propagation algorithm. Classification performance is evaluated by using the ground truth. Results with higher than 90% of correct classification are obtained. The results are also compared with a statistical classifier based on Kullback-Liebler distance via the Kappa coefficient.