13:30
Auralization and Acoustic Simulation
13:30
20 mins
|
Auralization of UAV Noise based on Sound Sphere Measurements
Leon Steinhoff, Dominik Gebensleben, Luljeta Sinani, Maria Stolz, Julian Benz, Stefan Becker, Carsten Spehr
Abstract: The impact of societal acceptance of unmanned aerial vehicles (UAVs) is continuously growing with the rising number of UAVs operating in various domains such as search and rescue, recreational use, logistics and urban air mobility. A highly important factor of public acceptance is the annoyance caused by the acoustic emissions of UAVs. To study the effects of different vehicle types, flight manoeuvres and environmental or operational influences on the annoyance, it is necessary to model the sound of flying UAVs in a realistic and repeatable manner.
Previous measurements analysed the sound directivity of flying UAVs. These measurements serve as the basis of the auralization techniques proposed in this study. Two basic approaches are compared: The generation of artificial sounds based on the results from the measurements (using methodology developed by Dreier and Vorländer [1]) and the use of real sounds recorded during the measurements. Effects like the doppler shift and ground reflections are also included in the auralization pipeline, following the method of Herold [2]. For validation, the rendered sound experience is compared to a real fly-by recording of different manoeuvres and UAVs by using psychoacoustic measures.
The study also shows a possible integration of the sound into the Unreal Engine environment incorporating frequency and speed dependent directivities and volumes. This method will later be used for acceptance studies which measure the human reaction and interaction with UAV traffic in a virtual environment. It is also shown how the auralization can be used in a virtual U-Space environment and influence the user experience. By incorporating the auralization into virtual environment, this work opens new possibilities for studying human perception, acceptance, and interaction with UAV traffic under varying flight behaviours and environmental conditions.
|
13:50
20 mins
|
Preliminary assessment of active noise cancelling techniques for drone acoustic emissions
Jan Adam Wanatowicz, Roberto Merino-Martinez
Abstract: Unmanned Aerial Vehicles (UAVs) are among the fastest-growing sectors of the aerospace industry. As drones become prevalent in densely populated urban areas, they pose a significant risk of increasing noise pollution. Despite having lower sound power than conventional aircraft, UAV noise is often perceived as more annoying due to its prominent high-frequency content and tonal components.
While traditional noise mitigation relies on hardware design, such as rotor synchronization or blade geometry, recent research highlights the potential of Active Noise Control (ANC) techniques. Because global control of the sound field is physically limited, especially at higher frequencies, this study proposes an onboard ANC system designed to create a localized "quiet zone" beneath the vehicle. In practice, this quiet zone could be selected depending on the application of interest. To evaluate this, an adaptive real-time control algorithm was developed for integration with a DJI Mavic 3 Enterprise drone.
Experimental validation will be conducted in the anechoic chamber at Delft University of Technology, using microphone and speaker arrays mounted to the UAV fuselage. Samples will be gather at the intended "quiet zone" and outside of it to examine possible interference. Acoustic performance of different ANC techniques will be quantified in terms of conventional and sound quality metrics, using the Sound Quality Analysis Toolbox (SQAT) to evaluate loudness, sharpness, tonality, roughness, and fluctuation strength. Furthermore, subjective annoyance ratings will be gathered through psychoacoustic listening tests. The results are expected to serve as proof of concept and evaluate the efficacy of localized ANC in improving the psychoacoustic profile of UAV operations.
|
14:10
20 mins
|
On the psychoacoustic impact of synchrophasing for distributed propellers and ducted fans
Stephen Schade, Roberto Merino-Martinez, Maximilian Marggraf, Karl-Stéphane Rossignol, Sébastien Guérin
Abstract: Synchrophasing has emerged as a possible strategy to mitigate tonal noise by specifically adjusting the relative blade-phase angle Δϕ between adjacent propellers in distributed propulsion systems. Recent studies have demonstrated that tonal noise radiation can be reduced when Δϕ approximates half the propeller blade-to-blade angle (e.g. Δϕ=30 degrees for a six-bladed propeller). These findings motivate the application of synchronized distributed propellers for community noise reduction in the context of urban air mobility. Despite acoustic and psychoacoustic investigations on the benefits of synchrophasing for basic propeller arrays, its psychoacoustic effect has not yet been demonstrated for realistic distributed propulsion systems.
This work verifies if the sound quality improves using synchrophasing for two different distributed propulsion systems. The first system consists of eight electrically driven, propellers distributed along the wings with a propeller diameter of 1.85 m. The second system is powered by 26 ducted, tiltable fans distributed above the wings with a fan diameter of 0.45 m. We address two research questions: (1) What tonal noise reduction potential and sound quality improvement can be achieved for these propulsion systems under realistic flight conditions? (2) Are the results equally valid for both propulsion systems?
To address these questions, an auralization framework based on measured noise emission data is employed. Measurements were performed in the Co-/Counter-Rotating Acoustic Fan Test rig (CRAFT) and in the Acoustic Wind tunnel Braunschweig (AWB) under realistic conditions for urban air mobility propulsion systems. Several measurements are combined to represent the distributed propulsion systems. Virtual flyover simulations and binaural noise syntheses are conducted to determine their noise signatures. As a post-processing step, sound quality metrics are calculated to examine the psychoacoustic characteristics of the obtained noise signatures.
For distributed propellers, different Δϕ values are considered. For the distributed fan stages, the hypothesis is formulated that, with low-count OGV fan engines, optimal stator clocking positions can efficiently reduce the tonal noise radiation. Therefore, an evolution-based optimization routine is employed to adjust the stator clocking angles of the 26 ducted fan engines.
|
14:30
20 mins
|
Acoustic Signatures of UAVs and How We Test Them: A Methodological Review for Human-Response Research
Tianjing Feng, Jian Kang
Abstract: Unmanned aerial vehicle (UAV) noise is a growing concern for the low-altitude economy and urban air mobility, because public acceptance will depend strongly on acoustic impact. Existing reviews largely summarise emission, propagation and annoyance results, but do not systematically examine how perception experiments represent UAV-specific acoustics. This review therefore asks: (i) how UAV acoustic signatures differ from more conventional environmental noise and through which psychoacoustic mechanisms they are expected to influence human responses; and (ii) how these signatures and key contextual factors are currently represented and reported in UAV-noise perception experiments. Across the reviewed work, UAV noise emerges as a strongly time-varying, operation-dependent and directionally structured signal: hover, climb and translation phases, RPM and payload changes, vehicle attitude, proximity to façades or ground, rotor configuration, dominant BPF/harmonic bands and signal-to-noise ratio all systematically shape what reaches the listener. In contrast, most perception studies operationalise UAVs in a reduced form, focusing on vehicle type and overall SPL, with comparatively little systematic control of flight mode and attitude, payload state, source–receiver geometry, reflections, background masking or visual scene. SPL strategies are usually tailored to the task—wide ranges with 5–10 dB steps for annoyance–level or detection functions, fixed or equalised level/loudness when isolating spectral or sound-quality differences, and SNR-based designs when studying masking—while auralisation practices range from mono or stereo headphone replay to binaural rendering and multi-channel or Ambisonics reproduction. Building on this contrast, the review formulates design-oriented guidance that links UAV acoustic properties to experimental choices. Stimuli should be built around clearly defined operations (e.g. take-off, hover, translation, landing or repeated service events) with sufficient temporal evolution and repeated flights per condition; SPL and SNR schemes should be chosen explicitly according to whether the aim is annoyance, detection, masking or sound quality; and spatial layout, reflections, background sound and visuals should be planned and documented so that listener position, masking and visual salience match the intended scenario, with spatial audio preferred over purely mono playback whenever directional or movement cues are part of the research question.
|
|