A K Nearest Neighborhood-Based Wind Estimation for Rotary-Wing VTOL UAVs

Wang, Liyang; Misra, Gaurav; Bai, Xiaoli

doi:10.3390/drones3020031

Cited by 19 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ultimately, while our methods build upon existing theoretical frameworks proposed by [ 9 , 10 ], to the best of our knowledge this work presents the first scientific investigation of a real-world application using machine learning methods to accurately predict windspeeds using onboard drone attitude and inertial data. During placid periods of wind activity, we found that the linear model performed on par with the KNN algorithm and LSTM neural network.…”

Section: Discussionmentioning

confidence: 99%

“…Such methods, while shown to be effective, are less extendable to other drone models than our proposed methodology, which eliminates the need for complex mathematical representations by instead learning the wind-attitude relationship directly from available attitude data. Work in [ 9 ] established the feasibility of using k-nearest neighbors (KNN) for rotary-wing-based windspeed estimation. However, speeds in their indoor experimentation did not exceed 3.07 m·s −1 , which constrains the applicability and representativeness of their findings.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Two Supervised Machine Learning Approaches for Wind Velocity Estimation Using Multi-Rotor Copter Attitude Measurements

Crowe

Pamula

Cheung

et al. 2020

Sensors

View full text Add to dashboard Cite

In this work we address the adequacy of two machine learning methods to tackle the problem of wind velocity estimation in the lowermost region of the atmosphere using on-board inertial drone data within an outdoor setting. We fed these data, and accompanying wind tower measurements, into a K-nearest neighbor (KNN) algorithm and a long short-term memory (LSTM) neural network to predict future windspeeds, by exploiting the stabilization response of two hovering drones in a wind field. Of the two approaches, we found that LSTM proved to be the most capable supervised learning model during more capricious wind conditions, and made competent windspeed predictions with an average root mean square error of 0.61 m·s−1 averaged across two drones, when trained on at least 20 min of flight data. During calmer conditions, a linear regression model demonstrated acceptable performance, but under more variable wind regimes the LSTM performed considerably better than the linear model, and generally comparable to more sophisticated methods. Our approach departs from other multi-rotor-based windspeed estimation schemes by circumventing the use of complex and specific dynamic models, to instead directly learn the relationship between drone attitude and fluctuating windspeeds. This exhibits utility in a range of otherwise prohibitive environments, like mountainous terrain or off-shore sites.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Two Supervised Machine Learning Approaches for Wind Velocity Estimation Using Multi-Rotor Copter Attitude Measurements

Crowe

Pamula

Cheung

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…As is well known, the ideal hover is possible when = 0 and all forces which act on a quadcopter are compensated. The roll, pitch, and yaw angles in a turbulent atmosphere are sums of the average and fluctuation components: (19) can be transformed to the case of ideal hover by means of their linearization. In the small-angle approximation , ≪ π, at ′ ≪ and under the conditions ̈=̈=̈= 0 and = 0, the equations for estimates of the horizontal components of the wind field � = and � = take the forms…”

Section: Estimates Of the Wind Velocity Componentsmentioning

confidence: 99%

“…The K Nearest Neighborhood learning-based method is suggested in [19] for rotary-wing vertical take-off and landing UAVs. Experimental results obtained with the use of a Parrot AR.Drone demonstrate the accuracy and robustness of the developed wind estimation algorithms under hovering conditions.…”

Section: Introductionmentioning

confidence: 99%

Low-Altitude Atmospheric Turbulence Sounding on the Basis of Unmanned Aerial Vehicle

Shelekhov¹,

Afanasiev²,

Kobzev³

et al. 2020

Preprint

View full text Add to dashboard Cite

Based on the theory of turbulence, equations are derived for estimations of turbulent fluctuations of the longitudinal and transverse components of the wind velocity during ideal hovering of a quadcopter in a turbulent atmosphere. We present the results of experiments which were carried out on the territory of the Geophysical Observatory of Institute of Monitoring of Climatic and Ecological Systems, Siberian Branch, Russian Academy of Sciences, located in Tomsk Akademgorodok on the territory with complex orography, in a parkland zone with buildings of research institutes and motorways. Time series of turbulent fluctuations of the longitudinal and transverse components of wind velocity fluctuations were received with the use of an automated weather station, and time series of estimates of these components, from data of a DJI Phantom 4 Pro quadcopter during hovering. According to the automated weather station data, anisotropy was observed in one experiment during measurements in the atmosphere, but this phenomenon was not observed in the other experiment: the fluctuation spectra of all components of wind speed fluctuations coincide. The spectra of fluctuations of the longitudinal and transverse wind velocity components based on the automated weather station data and UAV telemetry are presented. The fluctuation spectra of these components for the automated weather station data and quadcopter generally coincide. The behavior of the spectra coincides with the spectrum which corresponds to Kolmogorov–Obukhov “–5/3” law within the inertial range. The turbulent spectra of the wind velocity fluctuations obtained with the use of the automatic weather station and with the unmanned aerial vehicle differ in the high-frequency spectral region.

show abstract

“…to heights of about 500 m, is the development of methods for diagnostics of the turbulent atmosphere with UAVs. The results of diagnostics of the speed of air mass flows with UAV are reported in [18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35]. In [36][37][38][39][40][41][42][43][44][45][46], the fundamental possibility of measuring the turbulence spectra with fixed-wing UAVs of various sizes and weights was shown.…”

Section: Introductionmentioning

confidence: 99%

Low-altitude Sensing of Urban Atmospheric Turbulence with UAV

Shelekhov¹,

Afanasiev²,

Shelekhova³

et al. 2022

Preprint

View full text Add to dashboard Cite

The capabilities of a quadcopter in the hover mode for low-altitude sensing of atmospheric turbulence with high spatial resolution in urban areas characterized by complex orography are investigated. The studies were carried out in different seasons (winter, spring, summer, and fall), and the quadcopter hovered in the immediate vicinity of ultrasonic weather stations. The DJI Phantom 4 Pro quadcopter and AMK-03 ultrasonic weather stations installed in different places of the studied territory were used in the experiment. The smoothing procedure was used to main regularities in the behavior of the longitudinal and lateral spectra of turbulence in the inertial and energy production ranges. The longitudinal and lateral turbulence scales were estimated by the least-square fit method with the von Karman model as a regression curve. It is shown that the turbulence spectra obtained with DJI Phantom 4 Pro and AMK-03 generally coincide with minor differences observed in the high-frequency region of the spectrum. In the inertial range, the behavior of the turbulence spectra shows that they obey the Kolmogorov-Obukhov “5/3” law. In the energy production range, the longitudinal and lateral turbulence scales and their ratio measured by DJI Phantom 4 Pro and AMK-03 agree to a good accuracy. Discrepancies in the data obtained with the quadcopter and the ultrasonic weather stations at the territory with complex orography are explained by the partial correlation of the wind velocity series at different measurement points and the influence of the inhomogeneous surface.

show abstract

A K Nearest Neighborhood-Based Wind Estimation for Rotary-Wing VTOL UAVs

Cited by 19 publications

References 15 publications

Two Supervised Machine Learning Approaches for Wind Velocity Estimation Using Multi-Rotor Copter Attitude Measurements

Two Supervised Machine Learning Approaches for Wind Velocity Estimation Using Multi-Rotor Copter Attitude Measurements

Low-Altitude Atmospheric Turbulence Sounding on the Basis of Unmanned Aerial Vehicle

Low-altitude Sensing of Urban Atmospheric Turbulence with UAV

Contact Info

Product

Resources

About