Air Data Parameters Research Articles

Fault-tolerant FADS system development for a hypersonic vehicle via neural network algorithms

Hypersonic vehicles suffer from extreme aerodynamic heating during flights, especially around the area of leading edge due to its small curvature. Therefore, flush air data sensing (FADS) system has been developed to perform accurate measurement of the air data parameters. In the present study, the method to develop the FADS algorithms with fail-operational capability for a sharp-nosed hypersonic vehicle is provided. To be specific, the FADS system implemented with 16 airframe-integrated pressure ports is used as a case study. Numerical simulations of different freestream conditions have been conducted to generate the database for the FADS targeting in 2≤Ma≤5 and 0km≤H≤30km. Four groups of neural network algorithms have been developed based on four different pressure port configurations, and the accuracy has been validated by 280 groups of simulations. Particularly, the algorithms based on the 16-port configuration show an excellent ability to serve as the main solver of the FADS, where 99.5% of the angle-of-attack estimations are within the error band ±0.2∘. The accuracy of the algorithms is discussed in terms of port configuration. Furthermore, diagnosis of the system health is present in the paper. A fault-tolerant FADS system architecture has been designed, which is capable of continuously sensing the air data in the case that multi-port failure occurs, with a reduction in the system accuracy.

Aerodynamic Errors Mathematical Modeling in Air Data Systems Estimation Technology in Flight Tests Using Satellite Navigation Systems

Problems of mathematical modeling of onboard air data systems errors are of paramount importance in pitot-static sources errors determination, air data systems evaluation in flight tests. The problems of development, identification and assessment of the mathematical models of errors adequacy acquire main importance in the modern technology of the air parameters true values determination, air data systems evaluation using satellite navigation systems, developed and applied in the practice of flight tests at JSC "FRI n.a. M. M. Gromov". This paper gives a general description of an air data systems estimation technology using satellite navigation systems. The principles of solving problems of aircraft data systems aerodynamic errors mathematical modeling are stated. The structure of mathematical models, factors of the aerodynamic errors, relationship of the solving problems of errors modeling within the framework of technology with a flight experiment plan are shown. Mathematical models parameters identification are based on a complex solving of the problems of a true air data parameters values and aerodynamic errors determination in flight tests. New results of mathematical modeling of errors in tests at high angles of attack in 2018 year of medium-range and short-range aircraft are presented. The results illustrate the technology effectiveness in solving the problems of flight tests at high angles of attack information support, aerodynamic errors modeling, air data systems estimation. The applied modeling methods make it possible to allocate in the mathematical models of pitot-static sources aerodynamic errors even the factors of very weak aerodynamic influence, comparable with the minimum pressure sensors instrumental errors.

A Machine Learning Approach for Estimating Air Data Parameters of Small Fixed-Wing UAVs Using Distributed Pressure Sensors

This paper presents a method for estimating the air data parameters for a small fixed-wing, unmanned aerial vehicle (UAV) using an arrangement of low-cost Micro-electromechinal systems (MEMS)-based pressure sensors embedded in the surface of the UAV. The pressure measurements are used in a machine learning (ML) model to estimate the angle of attack, sideslip angle, and airspeed. Two ML algorithms based on artificial neural networks (NNs) and linear regression (LR) are implemented, tested, and assessed using data collected from wind tunnel experiments and a flight test, and the results are compared to a benchmark flight test. Training the ML algorithms using wind tunnel data was found to introduce several potential error sources that need to be addressed in order to provide accurate estimation on the benchmark flight test, whereas training the algorithms using flight data provides lower estimation RMSE values. The performance of the NN structures has been found to slightly outperform the LR algorithms in estimation accuracy. Finally, results from using different sensor configurations and a pseudo-Reynolds number are presented in an effort to evaluate the influence of sensor number and placement on the accuracy of the method.

Numerical Calibration of a Low-Speed sUAS Flush Air Data System

AbstractA method using computational fluid dynamics to numerically calibrate a flush air data system is presented. A small unmanned aircraft system (sUAS) has been equipped with a flush air data system and experimentally tested. The flush air data system uses computational fluid dynamics to train neural networks and is validated using the in-flight data that were previously collected. Results of the flight validation are presented, along with ways to improve the accuracy of the system. Several different calibration approaches are presented and compared with each other. The best-case results with the in-flight calibration are 0.59° and 0.66° for angle of attack and sideslip, respectively, whereas the best-case results when calibrated with computational fluid dynamics data are 0.78° and 0.90°. It is also possible to estimate other air data parameters, such as dynamic pressure, static pressure, and density, with neural networks, but the direct calculation is more accurate. Calibrating the system numerically, such as with the use of computational fluid dynamics, removes the need for any calibration flights. Although not as accurate as the in-flight calibration, numerical calibration is possible and can save the user time and expense.

Vibration Testing of Absolute Pressure Sensor for a Flush Air Data System (FADS)

ABSTRACT Flush Air Data Sensing System (FADS) is an important subsystem required for a winged re-entry vehicle. It essentially makes use of surface pressure measurements located at a suitable position in the vehicle for deriving the air data parameters of the vehicle such as angle of attack, angle of sideslip, Mach number, etc. These air data parameters are used by the control and guidance system for carrying out gain scheduling, trim scheduling of control surfaces and overall flight management. The accuracy of the pressure sensors used in FADS is critical in achieving the targeted accuracy of the air data parameters. In the FADS system considered in this paper, piezoresistive MEMS-based pressure sensors are used. The FADS system is located in the nosecone of the vehicle which is a region prone to vibration during transportation as well as during flight. It is essential that the pressure sensors perform with the desired accuracies in spite of the vibration environment. Sine and random vibration testing are carried out as part of qualification testing of the vibration sensors of FADS. Analysis of the test results shows that the accuracies demanded by the system are provided by the pressure sensors under the vibration environment.

Leading-Edge Flow Sensing for Aerodynamic Parameter Estimation

The identification of inflow air-data quantities such as the airspeed, angle of attack, and local lift coefficient on various sections of a wing or rotor blade is beneficial for load monitoring, aerodynamic diagnostics, and control on devices ranging from air vehicles to wind turbines. Real-time measurement of aerodynamic parameters during flight provides the ability to enhance aircraft operating capabilities while preventing dangerous stall situations. This Paper presents a novel leading-edge flow sensing algorithm for the determination of the air-data parameters using discrete surface pressures measured at a few ports in the vicinity of the leading edge of a wing or blade section. The approach approximates the leading-edge region of the airfoil as a parabola and uses pressure distribution from the exact potential-flow solution for the parabola to fit the pressures measured from the ports. Pressures sensed at five discrete locations near the leading edge of an airfoil are given as input to the algorithm to solve for the model parabola-flow problem using a simple nonlinear regression. The algorithm directly computes the inflow velocity and the stagnation-point location. A look-up table approach is adopted to deduce the section angle of attack and lift coefficient. The performance of the algorithm is assessed using experimental data in the literature for airfoils under different flow conditions. The results show good correlation between the actual and predicted aerodynamic quantities within the prestall regime, even for rotating blade sections.

Attitude, Angle of Attack and Angle of Sideslip Determination Using Carrier Phase Integer Ambiguity Estimation

The Global Navigation Satellite System (GNSS) positioning solution may be computed to a cm level of error, if the signal carrier phase is properly considered during the calculations. With such a precision, even attitude angles can be precisely estimated if more than one antennae is available. However, using the carrier phase information depends on solving the carrier phase integer ambiguity problem which is intrinsic to this approach. Since only a fraction of the carrier phase is observed by the GNSS receivers, the solution for the integer ambiguity is usually obtained by means of search algorithms, based on least squares. This contribution investigates the performance of the integer ambiguity solution for the attitude determination problem and also proposes a novel application: angle of attack and angle of sideslip estimation for specific aeronautical application. This novel application benefits aeronautical industry by introducing a new alternative for measuring reliable airdata parameters, namely angle of attack and angle of sideslip, by using a nonsusceptible to icing conditions data. The results presented here are based on the constrained LAMBDA method. Dynamic real data are used for evaluation of the proposed novel application.

Flush Air Data Sensing System

Flush air data sensing system (FADS) forms a mission-critical subsystem in re-entry vehicles. It makes use of surface pressure measurements from the nose cap of the vehicle for deriving air data parameters such as angle of attack, angle of sideslip, Mach number, etc. of the vehicle. These parameters are used by the flight control and guidance systems, and also assist in the overall mission management. The overall system engineering of FADS, including selection of pressure transducers, tubing size, port geometry, FADS algorithm and associated processing electronics along with the integration scheme is addressed in this article. Details of the qualification tests carried out in wind tunnel for end-to-end verification of the entire FADS system are covered in brief. Majority of the tests were carried out in a low-speed wind tunnel at a wind speed of 65 m/s (Mach number 0.2). The flight performance of FADS is also discussed in this article.

Inverse Flush Air Data System (FADS) for Real Time Simulations

Flush Air Data Sensing System (FADS) forms a mission critical sub system in future reentry vehicles. FADS makes use of surface pressure measurements from the nose cap of the vehicle for deriving the air data parameters of the vehicle such as angle of attack, angle of sideslip, Mach number, etc. These parameters find use in the flight control and guidance systems, and also assist in the overall mission management. The FADS under consideration in this paper makes use of nine pressure ports located in the nose cap of a technology demonstrator vehicle. In flight, the air data parameters are obtained from the FADS estimation algorithm using the pressure data at the nine pressure ports. But, these pressure data will not be available, for testing the FADS package during ground simulation. So, an inverse software to FADS which estimates the pressure data at the pressure ports for a given flight condition is developed. These pressure data at the nine ports will go as input to the FADS package during ground simulation. The software is run to generate the pressure data for the descent phase trajectory of the technology demonstrator. This data is used again to generate the air data parameters from FADS algorithm. The computed results from FADS algorithm match well with the trajectory data.

Predictor–Corrector Method for Scramjet Inlet Air Mass Flow Rate Measurement

The accurate and real-time measurement of the scramjet inlet air mass flow rate is critical to flight control. To measure the scramjet inlet air mass flow rate, air data parameters are estimated. The inertial-navigation-system-computed air data are real time but not accurate enough. The flush air data sensing system can provide sufficiently accurate estimations of air data parameters, but it is not employed during the flight. This paper proposes a predictor–corrector method. Accurate air data estimations are not required. The scramjet inlet air mass flow rate estimated by the inertial-navigation-system-computed air data is taken as the prediction value. Surface pressure measurements combined with the inertial navigation system provide the correction value. The algorithm is simple enough and can be implemented in real time. Comparing the results of the scramjet inlet air mass flow rate estimated by the inertial navigation system and predictor–corrector method, it is shown that the predictor–corrector method can greatly improve the accuracy of the scramjet inlet air mass flow rate measurement. The scramjet inlet air mass flow rate measurement sensitivity to the freestream static pressure, static temperature, and angle of attack is reduced significantly by the predictor–corrector method.

A Comprehensive Probabilistic Framework to Learn Air Data from Surface Pressure Measurements

Use of probabilistic techniques has been demonstrated to learn air data parameters from surface pressure measurements. Integration of numerical models with wind tunnel data and sequential experiment design of wind tunnel runs has been demonstrated in the calibration of a flush air data sensing anemometer system. Development and implementation of a metamodeling method, Sequential Function Approximation (SFA), are presented which lies at the core of the discussed probabilistic framework. SFA is presented as a tool capable of nonlinear statistical inference, uncertainty reduction by fusion of data with physical models of variable fidelity, and sequential experiment design. This work presents the development and application of these tools in the calibration of FADS for a Runway Assisted Landing Site (RALS) control tower. However, the multidisciplinary nature of this work is general in nature and is potentially applicable to a variety of mechanical and aerospace engineering problems.

Learning Air-Data Parameters for Flush Air Data Sensing Systems

An adaptive scattered data approximation scheme was developed to calibrate the Flush Air Data System (FADS) of a surface vessel. An array of pressure sensors were mounted flush with the deckhouse periphery and the airdata parameters were extracted from the pressure measurements. The developed Galerkin derived selfadaptive greedy function approximation scheme gave reliable and robust surrogates for predicting wind speed and direction. The resulting surrogates were also used to evaluate the sensitivity to each of the flush mounted pressure sensor. Fault tolerance of the proposed surrogates were also studied with respect to pressure sensor failure.

Recalibration of Flush Air Data System (FADS) on Pressure Port/Sensor Failures

Flush Air Data sensing System (FADS) forms a mission critical sub system in future reusable space transportation systems. FADS makes use of surface pressure measurements from the nose cap of the vehicle for deriving the air data parameters of the vehicle like angle of attack, angle of sideslip, Mach number, etc. These parameters are used by the control and guidance system of the vehicle for reentry flight management as well as for subsequent mission phases upto landing. The on board algorithm makes use of calibration coefficients derived from wind tunnel data to achieve better accuracy in the estimated air data parameters. In the event of faults in FADS due to port blockages or pressure sensor failures, recalibration is essential for achieving the desired accuracy of air data estimates. This paper describes recalibration of FADS using surface fit method in the event of pressure port blockage or sensor failures. The method is based on the storage of coefficients of a 4th order surface fit corresponding to the wind tunnel calibration data and recalibration using these polynomial coefficients.

Unmanned air vehicle air data estimation using a matrix of pressure sensors: a comparison of neural networks and look-up tables

Flush airdata sensing (FADS) systems are cost- and weight- effective alternatives to current air data booms for measuring important air data parameters such as airspeed, angle of attack, sideslip, etc. Most applications consider large manned/unmanned air vehicles where the Pitot-static tube is located at the nose tip. However, traditional air data booms can be physically impractical for micro- (unmanned) air vehicles (MAVs) and, in this article, a FADS system mounted on the wing leading edge of a MAV flown at low speeds of Mach 0.07 (wind tunnel experiments under corresponding conditions) is designed. Moreover, two approaches for converting the FADS system pressure to meaningful air data are compared: a neural network (NN) approach and a look-up table (LUT). Results have shown that instrumentation weight and cost were reduced by 80 per cent and 97 per cent, respectively, in comparison to a traditional air data boom. Overall, the NN estimation accuracies were 0.51°, 0.44 lb/ft2, and 0.62 m/s and the LUT estimation accuracies 1.32°, 0.11 lb/ft2, and 0.88 m/s for the angle of attack, static pressure, and airspeed, respectively. It was also found that the LUT has faster execution times while the NN was in most cases more robust to sensor faults. However, while the LUT requires high memory usage, especially for higher dimensions, the NN can be executed in a few lines of code.

Failure management scheme for use in a flush air data system

This paper concerns the development of a failure management software for use in a Flush Air Data System (FADS). FADS is used for the online computation of air data parameters namely, Mach number, angle of attack and angle of side slip. Failure management, an essential requirement for FADS, especially in the context of aircraft flight control applications, has been addressed using a novel concept called failure indicator vector. This new methodology ensures the selection of correct values out of a number of redundant values of a computed air data parameter. This method leads to symbolic processing techniques, which is found to be very effective in terms of programming efficiency and simple procedure for logical reasoning. The method and the software have been successfully tested using the wind tunnel data generated at NAL.

Fault-Tolerant Neural Network Algorithm for Flush Air Data Sensing

A fault-tolerant neural network algorithm was successfully developed for use with flush air data sensing systems. This algorithm is composed of a combination of aerodynamic and neural network models used to translate a discrete pressure distribution from the nose of an aircraft into a set of air data parameters, including static pressure, dynamic pressure, Mach number, angle of attack, and angle of sideslip. Techniques were developed to detect and eliminate the effect of a lost signal from the measured pressure distribution. This system was evaluated with archived data, and its performance was compared with a signal processing system based completely on aerodynamic models

Flush Air Data System Calibration Using Numerical Simulation

The use of computational fluid dynamics simulations for calibrating a flush air data system is described, In particular, the flush air data system of the HYFLEX hypersonic vehicle is used as a case study. The HYFLEX air data system consists of nine pressure ports located flush with the vehicle nose surface, connected to onboard pressure transducers, After appropriate processing, surface pressure measurements can he converted into useful air data parameters. The processing algorithm requires an accurate pressure model, which relates air data parameters to the measured pressures. In the past, such pressure models have been calibrated using combinations of flight data, ground-based experimental results, and numerical simulation. We perform a calibration of the HYFLEX flush air data system using computational fluid dynamics simulations exclusively, The simulations are used to build an empirical pressure model that accurately describes the HYFLEX nose pressure distribution ol cr a range of flight conditions. We believe that computational fluid dynamics provides a quick and inexpensive way to calibrate the air data system and is applicable to a broad range of flight conditions, When tested with HYFLEX flight data, the calibrated system is found to work well. It predicts vehicle angle of attack and angle of sideslip to accuracy levels that generally satisfy flight control requirements. Dynamic pressure is predicted to within the resolution of the onboard inertial measurement unit. We find that wind-tunnel experiments and flight data are not necessary to accurately calibrate the HYFLEX flush air data system for hypersonic flight.

In-flight demonstration of a real-time flush airdata sensing system

A prototype real-time flush airdata sensing system has been developed and flight tested at the NASA Dryden Flight Research Center. This system uses a matrix of pressure orifices on the vehicle nose to estimate airdata parameters in real time using nonlinear regression. The algorithm is robust to sensor failures and noise in the measured pressures. The real-time flush airdata sensing (FADS) system has been calibrated using inertial trajectory measurements that were bootstrapped for atmospheric conditions using meteorological data. Mach numbers as high as 1.6 and angles of attack greater than 45 deg have been tested. The system performance has been evaluated by comparing the real-time FADS to the ship system airdata computer measurements to give a quantitative evaluation relative to an accepted measurement standard. Nominal agreements of approximately 0.003 in Mach number and 0.20 deg in angle of attack and angle of sideslip have been achieved.

Air data prediction from surface pressure measurements on guided munitions

The control of guided munitions over wide ranges of flight conditions has posed new requirements for estimating air data parameters. A common approach to meet the control stability margins and time response requirements is varying autopilot gains as a function of estimated air data parameters. A feasibility study to estimate air data parameters using flush-orifice pressure sensors has been completed. Wind-tunnel measurements of static pressure on sensors mounted on a representative missile nose were used. The primary objective of this study was to explore options for in-flight estimation of air data parameters using only the static pressure data base. The air data parameters of interest include Mach number, angle of attack, aerodynamic roll angle, and freestream pressure. An iterative process based on the differential corrections method was used to determine the air data parameters and expected error in the final estimates. Examples are given that provide sensitivity to pressure measurement noise and different sensor configurations. The air data estimates obtained with this method are suitable for autopilot gain scheduling over a wide range of flight conditions.