Neural-Network-Based Flush Air Data Sensing System Demonstrated on a Mini Air Vehicle

Samy, Ihab; Postlethwaite, Ian; Gu, Da‐Wei; Green, John

doi:10.2514/1.44157

Cited by 64 publications

(44 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A variety of different man-made sensors have also been used to measure the distribution of flow properties on the wings of small UAS. The most common are pressure sensors [6][7][8][9][10][11][12], but more novel sensors such as hotfilms [13,14] and artificial hair sensors [15,16] have also been used to measure flow velocity.…”

Section: Introductionmentioning

confidence: 99%

Distributed Pressure Sensing–Based Flight Control for Small Fixed-Wing Unmanned Aerial Systems

Wood

Araujo-Estrada

Richardson

et al. 2019

Journal of Aircraft

View full text Add to dashboard Cite

Small fixed-wing Unmanned Aerial Systems (UAS) might require increased agility when operating in turbulent wind fields. In these conditions, conventional sensor suites could be augmented with additional flow-sensing to extend the aircraft's usable flight envelope. Inspired by distributed sensor arrays in biological systems, a UAS with a chord-wise array of pressure sensors was developed. Wind-tunnel testing characterised these sensors alongside a conventional airspeed sensor and an angle-of-attack (AoA) vane, and showed a single pressure measurement gave a linear response to AoA pre-stall. Flight tests initially manually piloted the vehicle through pitching manoeuvres, then in a series of automated manoeuvres based on closed-loop feedback using an estimate of AoA from the single pressure port. The AoA estimate was successfully used to control the attitude of the aircraft. An Artificial Neural Network (ANN) was trained to estimate the AoA and airspeed using all pressure ports in the array, and validated using flight-trial data. The ANN more accurately estimated the AoA over a single port method with good robustness to stall and unsteady flow. Distributed flow sensors could be used to supplement conventional flight control systems, providing enhanced information about wing flow conditions with application to systems with highly flexible or morphing wings.

show abstract

Section: Introductionmentioning

confidence: 99%

Distributed Pressure Sensing–Based Flight Control for Small Fixed-Wing Unmanned Aerial Systems

Wood

Araujo-Estrada

Richardson

et al. 2019

Journal of Aircraft

View full text Add to dashboard Cite

show abstract

“…[8][9][10] While in other cases the sensors comprise part of a flush air data system used to give estimates of aerodynamic state variables such as air speed, angle of attack and sideslip. 11,12 There have also been investigations into the potential advantages offered by unconventional flow sensors such as artificial hair based flow velocity sensors. 13,14 The use of force or strain sensing has also been explored, but not to the same extent, with studies showing that a small number of strain sensors could potentially offer advantages for attitude control based on total moment measurement, both theoretically 15 and experimentally.…”

Section: Introductionmentioning

confidence: 99%

Bio-inspired Distributed Strain and Airflow Sensing for Small Unmanned Air Vehicle Flight Control

Araujo-Estrada¹,

Salama²,

Greatwood³

et al. 2017

AIAA Guidance, Navigation, and Control Conference

View full text Add to dashboard Cite

“…They have been successfully designed and tested on a variety of engineering systems [4,[11][12][13][14][15][16]. In fact, authors in [17,18] have shown that NNs with at least one hidden layer are capable of approximating any nonlinear function when provided with sufficient training data and time.…”

Section: Introductionmentioning

confidence: 99%

Detection and accommodation of sensor faults in UAVs- a comparison of NN and EKF based approaches

Samy

Postlethwaite

2010

49th IEEE Conference on Decision and Control (CDC)

Self Cite

View full text Add to dashboard Cite

In this paper we propose two schemes for sensor fault detection and accommodation (SFDA); one based on a neural network (NN) and the other an extended Kalman filter (EKF). The objective is to compare both approaches in terms of execution time, robustness to poorly modelled dynamics and sensitivity to different fault types. The schemes are tested on an unmanned air vehicle (UAV) application where traditional sensor redundancy methods can be too heavy and/or costly. In an attempt to reduce the false alarm rates and the number of undetected faults, a modified residual generator, originally proposed in [11], is implemented. Simulation work is presented for use on a UA V demonstrator under construction with support from BAE Systems and EPSRC. Results have shown that the NN-SFDA scheme outperforms the EKF-SFDA scheme with only 1 missed fault, zero false alarms and an average estimation error ofO.31deg/s for 112 different test conditions.

show abstract

Neural-Network-Based Flush Air Data Sensing System Demonstrated on a Mini Air Vehicle

Cited by 64 publications

References 18 publications

Distributed Pressure Sensing–Based Flight Control for Small Fixed-Wing Unmanned Aerial Systems

Distributed Pressure Sensing–Based Flight Control for Small Fixed-Wing Unmanned Aerial Systems

Bio-inspired Distributed Strain and Airflow Sensing for Small Unmanned Air Vehicle Flight Control

Detection and accommodation of sensor faults in UAVs- a comparison of NN and EKF based approaches

Contact Info

Product

Resources

About