Modeling Aggregate Downwash Forces for Dense Multirotor Flight (2312.03488v1)

Abstract: Dense formation flight with multirotor swarms is a powerful, nature-inspired flight regime with numerous applications in the realworld. However, when multirotors fly in close vertical proximity to each other, the propeller downwash from the vehicles can have a destabilising effect on each other. Unfortunately, even in a homogeneous team, an accurate model of downwash forces from one vehicle is unlikely to be sufficient for predicting aggregate forces from multiple vehicles in formation. In this work, we model the interaction patterns produced by one or more vehicles flying in close proximity to an ego-vehicle. We first present an experimental test rig designed to capture 6-DOF exogenic forces acting on a multirotor frame. We then study and characterize these measured forces as a function of the relative states of two multirotors flying various patterns in its vicinity. Our analysis captures strong non-linearities present in the aggregation of these interactions. Then, by modeling the formation as a graph, we present a novel approach for learning the force aggregation function, and contrast it against simpler linear models. Finally, we explore how our proposed models generalize when a fourth vehicle is added to the formation.

• The paper introduces a novel non-linear modeling approach using Deep Sets to accurately predict aggregate downwash forces in dense multirotor formations.
• The experimental setup measured six degrees of freedom for various formations (Side by Side, Leader Follower, and Stack) to capture complex aerodynamic interactions.
• The findings reveal that simple linear summation of drone downwash data is insufficient for flight stability, emphasizing the need for advanced modeling techniques.

Introduction

The research explores the challenges and intricacies of dense formation flight among multirotor drones, which involves multiple drones flying in close vertical proximity. Precise modeling of aerodynamic downwash forces is essential but understudied in such scenarios, where individual drones affect each other's airflow. The paper introduces an innovative approach for accurately capturing and modeling aggregate downwash forces emanating from several drones in close formation, providing significant insights for improving drone swarm flight stability and performance.

Experimental Methodology

The research team developed a sophisticated experimental rig capable of measuring six degrees of freedom (6-DOF) exogenic forces on a drone frame. The setup was carefully designed to capture transient aerodynamic forces realistically, keeping biases to a minimum. The core of the experimental design included flight patterns to analyze how multiple nearby drones affect one another. Three primary formations were studied: Side by Side, Leader Follower, and Stack. These patterns were chosen due to their likelihood of creating complex interactions between the airflow of the drones.

Modeling Approach

Modeling the aggregate downwash forces was done using both linear and non-linear approaches. In the linear model, forces from individual drones are simply summed up, an assumption that may work for certain formation configurations. However, due to the non-linear nature of aerodynamics, a more complex approach was needed for accurate force prediction across varying flight conditions. The research proposes a novel non-linear model using techniques from the domain of Deep Sets, which allows modeling the interactions of the entire formation as opposed to only pairs of drones. This method was evaluated for its capacity to predict aerodynamic forces in different formation scenarios.

Findings and Conclusion

The paper's findings indicate that the linear model might provide reasonable estimations for formations with one drone significantly closer to another, but fails to accurately predict the downwash impacts in formations with multiple drones at different distances. The developed non-linear model, by contrast, more effectively captures the complex aerodynamic interactions, particularly in scenarios with higher non-linearity due to increased drone formation altitude.

The research illustrates that the downwash forces from drones do not expand significantly laterally with distance and that the aerodynamic influence of one drone is minimal on another outside of the downwash field itself. The results highlight that straightforward summation of single-drone models does not suffice when multiple drones are subjected to overlapping downwash zones, which can cause considerable predictive errors.

As a part of acknowledging future work, the paper suggests further investigation into the effects of varying velocities on downwash forces and exploring ways to generalize models to formations with more than three drones. The contribution of this research is a step forward in enabling precise control and planning for dense multirotor formations, with potential applications across defense, disaster response, and large-scale aerial sensing.