Influence of Static and Dynamic Downwash Interactions on Multi-Quadrotor Systems

TL;DR

This work tackles the challenge of downwash-induced instability in dense multi-quadrotor systems by performing a data-driven experimental study that combines six-axis force/torque measurements and high-resolution PIV to map near-field and far-field wake structures for single and two-quadrotor configurations. The authors quantify how downwash from an upper quadrotor alters thrust and pitch moments on a lower quadrotor as a function of relative separation, and they show that far-field wakes obey turbulent-jet scaling with self-similar profiles, even at intermediate Reynolds numbers around $Re \approx 9000$. A four-parameter reduced-order model ($l$, $z_0$, $U_i$, $z_m/l$) captures the observed scaling and wake merging, and an open-access dataset accompanies the findings. Dynamic proximity experiments reveal phase-dependent force and moment variations, highlighting when approach vs retreat phases produce stronger disturbances, thereby enabling flow-aware control strategies to extend the operational envelope in dense formations.

Abstract

Flying multiple quadrotors in close proximity presents a significant challenge due to complex aerodynamic interactions, particularly downwash effects that are known to destabilize vehicles and degrade performance. Traditionally, multi-quadrotor systems rely on conservative strategies, such as collision avoidance zones around the robot volume, to circumvent this effect. This restricts their capabilities by requiring a large volume for the operation of a multi-quadrotor system, limiting their applicability in dense environments. This work provides a comprehensive, data-driven analysis of the downwash effect, with a focus on characterizing, analyzing, and understanding forces, moments, and velocities in both single and multi-quadrotor configurations. We use measurements of forces and torques to characterize vehicle interactions, and particle image velocimetry (PIV) to quantify the spatial features of the downwash wake for a single quadrotor and an interacting pair of quadrotors. This data can be used to inform physics-based strategies for coordination, leverage downwash for optimized formations, expand the envelope of operation, and improve the robustness of multi-quadrotor control.

Figures (16)

• Figure 1: Axial downwash velocity, $\bar{u}$, (left) of downwash below a Crazyflie quadrotor. The individual propeller flows observed in the near-field merge to form a turbulent jet in the far-field ($z/l>6.5$). A long-exposure image (right) of dynamic interaction between quadrotors, where two hovering Crazyflies, one accelerated using a linear traverse (CrazyRail) rapidly approaches another mounted on a load cell measuring forces and moments due to the interaction.
• Figure 2: Close proximity quadrotor flight under the influence of downwash; Case I (left) depicts vertical alignment with quadrotors (A) and (B) in a stacked configuration. Case II (right) illustrates quadrotor (B) horizontally offset from quadrotor (A). Black arrows adjacent to the rotors indicate their direction of rotation. In this figure, $F$ represents force, while $M$ represents moment.
• Figure 3: Coordinate system and setup schematic to measure aerodynamic interactions on forces and moments acting on a pair of quadrotors.
• Figure 4: PIV setup for quadrotor downwash velocity data acquisition. The laser sheet illuminates the tracer particles along the plane (side-view), slicing through the centers of the front two rotors of the quadrotors (top-view), capturing the flow field along that plane.
• Figure 5: Mean and standard deviations (unsteadiness) of forces and moments experienced by upper quadrotor generating downwash at a separation (${\Delta}x$, ${\Delta}z$) above the lower quadrotor. Horizontal and vertical separations are normalized by the arm length, $l$, in Fig. \ref{['fig:force_moment_setup']}