skip to main content

Title: Wind Profiling in the Lower Atmosphere from Wind-Induced Perturbations to Multirotor UAS
We present a model-based approach to estimate the vertical profile of horizontal wind velocity components using motion perturbations of a multirotor unmanned aircraft system (UAS) in both hovering and steady ascending flight. The state estimation framework employed for wind estimation was adapted to a set of closed-loop rigid body models identified for an off-the-shelf quadrotor. The quadrotor models used for wind estimation were characterized for hovering and steady ascending flight conditions ranging between 0 and 2 m/s. The closed-loop models were obtained using system identification algorithms to determine model structures and estimate model parameters. The wind measurement method was validated experimentally above the Virginia Tech Kentland Experimental Aircraft Systems Laboratory by comparing quadrotor and independent sensor measurements from a sonic anemometer and two SoDAR instruments. Comparison results demonstrated quadrotor wind estimation in close agreement with the independent wind velocity measurements. However, horizontal wind velocity profiles were difficult to validate using time-synchronized SoDAR measurements. Analysis of the noise intensity and signal-to-noise ratio of the SoDARs proved that close-proximity quadrotor operations can corrupt wind measurement from SoDARs, which has not previously been reported.  more » « less
Award ID(s):
1520825 1821145
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Page Range / eLocation ID:
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    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. 
    more » « less
  2. Abstract. A tethered-balloon system (TBS) has been developed and is beingoperated by Sandia National Laboratories (SNL) on behalf of the U.S.Department of Energy's (DOE) Atmospheric Radiation Measurement (ARM) UserFacility in order to collect in situ atmospheric measurements withinmixed-phase Arctic clouds. Periodic tethered-balloon flights have beenconducted since 2015 within restricted airspace at ARM's Advanced MobileFacility 3 (AMF3) in Oliktok Point, Alaska, as part of the AALCO (AerialAssessment of Liquid in Clouds at Oliktok), ERASMUS (Evaluation of RoutineAtmospheric Sounding Measurements using Unmanned Systems), and POPEYE(Profiling at Oliktok Point to Enhance YOPP Experiments) field campaigns. Thetethered-balloon system uses helium-filled 34 m3 helikites and 79 and104 m3 aerostats to suspend instrumentation that is used to measureaerosol particle size distributions, temperature, horizontal wind, pressure,relative humidity, turbulence, and cloud particle properties and tocalibrate ground-based remote sensing instruments. Supercooled liquid water content (SLWC) sondes using the vibrating-wireprinciple, developed by Anasphere Inc., were operated at Oliktok Point atmultiple altitudes on the TBS within mixed-phase clouds for over 200 h.Sonde-collected SLWC data were compared with liquid water content derivedfrom a microwave radiometer, Ka-band ARM zenith radar, and ceilometer at the AMF3, as well as liquid water content derived from AMF3 radiosonde flights. The in situ data collected by the Anasphere sensors were also compared with data collected simultaneously by an alternative SLWC sensor developed at the University of Reading, UK; both vibrating-wire instruments were typically observed to shed their ice quickly upon exiting the cloud or reaching maximum ice loading. Temperature sensing measurements distributed with fiber optic tethered balloons were also compared with AMF3 radiosonde temperature measurements. Combined, the results indicate that TBS-distributedtemperature sensing and supercooled liquid water measurements are inreasonably good agreement with remote sensing and radiosonde-basedmeasurements of both properties. From these measurements and sensorevaluations, tethered-balloon flights are shown to offer an effective methodof collecting data to inform and constrain numerical models, calibrate andvalidate remote sensing instruments, and characterize the flight environmentof unmanned aircraft, circumventing the difficulties of in-cloud unmanned aircraft flights such as limited flight time and in-flight icing. 
    more » « less
  3. Abstract Daytime atmospheric boundary layer (ABL) dynamics—including potential temperature budgets, water vapour budgets, and entrainment rates—are presented from in situ flight data taken on six afternoons near Fresno in the San Joaquin Valley (SJV) of California during July/August 2016. The flights took place as a part of the California Baseline Ozone Transport Study aimed at investigating transport pathways of air entering the Central Valley from offshore and mixing down to the surface. Midday entrainment velocity estimates ranged from 0.8 to 5.4 cm s −1 and were derived from a combination of continuously determined ABL heights during each flight and model-derived subsidence rates, which averaged -2.0 cm s −1 in the flight region. A strong correlation was found between entrainment velocity (normalized by the convective velocity scale) and an inverse bulk ABL Richardson number, suggesting that wind shear at the ABL top plays a significant role in driving entrainment. Similarly, we found a strong correlation between the entrainment efficiency (the ratio of entrainment to surface heat fluxes with an average of 0.23 ± 0.15) and the wind speed at the ABL top. We explore the synoptic conditions that generate higher winds near the ABL top and propose that warm anomalies in the southern Sierra Nevada mountains promote increased entrainment. Additionally, a method is outlined to estimate turbulence kinetic energy, convective velocity scale ( w * ), and the surface sensible heat flux in the ABL from a slow, airborne wind measurement system using mixed-layer similarity theory. 
    more » « less
  4. The dataset is derived from HELiX Uncrewed Aircraft System flights that were conducted in the Central Arctic Ocean over sea ice during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition. The data include Universal Coordinated Time (UTC), downwelling and upwelling shortwave radiation measurements, and position and attitude from the Uncrewed Aircraft System (UAS). Temperature, relative humidity and pressure from two different sensors are also provided. A quality control flag is associated with each scientific measurement. A flight flag is also included to indicate the different phases of the flight - on the ground, take-off/landing phases, and in flight. All the data have been synchronized and interpolated at 10 hertz (Hz). The purpose of this dataset is to provide information on albedo over different features of the sea ice (snow, melt pond, ocean). Three flight patterns were implemented during the campaign with the HELiX, a grid pattern at constant altitude (15 meters or 7 meters above ground level), hovering flights ( 2-5 minutes hovering over identified sea ice features at low altitude ~ 3 meters above ground level), and profiles up to 400 meters above ground level. Displaying latitude, longitude and altitude will help users to identify the flight pattern. Albedo measurements have been validated with surface-based measurements and details can be found in de Boer, G. R. Calmer, G. Jozef, J. Cassano, J. Hamilton, D. Lawrence, S. Borenstein, A. Doddi, C. Cox, J. Schmale, A. Preußer and B. Argrow (2021): Observing the Central Arctic Atmosphere and Surface with University of Colorado Uncrewed Aircraft Systems, Nature Scientific Data, in prep. 
    more » « less
  5. Rapp, Anita (Ed.)
    Vertical motions over the complex terrain of Idaho’s Payette River Basin were observed by the Wyoming Cloud Radar (WCR) during 23 flights of the Wyoming King Air during the SNOWIE field campaign. The WCR measured radial velocity, V_r, which includes the reflectivity-weighted terminal velocity of hydrometeors (V_t), vertical air velocity (w), horizontal wind contributions as a result of aircraft attitude deviations, and aircraft motion. Aircraft motion was removed through standard processing. To retrieve vertical radial velocity (W), V_r was corrected using rawinsonde data and aircraft attitude measurements. w was then calculated by subtracting the mean W, (W ̅), at a given height along a flight leg long enough for W ̅ to equal the mean reflectivity weighted terminal velocity, (V_t ) ̅, at that height. The accuracy of the w and (V_t ) ̅ retrievals were dependent on satisfying assumptions along a given flight leg that the winds at a given altitude above/below the aircraft did not vary, the vertical air motions at a given altitude sum to 0 m s-1, and (V_t ) ̅ at a given altitude did not vary. The uncertainty in the w retrieval associated with each assumption is evaluated. Case studies and a project wide summary show that this methodology can provide estimates of w that closely match gust probe measurements of w at the aircraft level. Flight legs with little variation in equivalent reflectivity factor at a given height and large horizontal echo extent were associated with the least retrieval uncertainty. The greatest uncertainty occurred in regions with isolated convective turrets or at altitudes where split cloud layers were present. 
    more » « less