Aerodynamic Force Modeling of Multirotor Unmanned Aerial Vehicles

Abstract

This paper describes a novel multirotor unmanned aerial vehicle aerodynamic force model with the aim of creating a multirotor simulation that is suitable for the evaluation of station-keeping performance in turbulent wind. It fills the gap between simple models that ignore important aerodynamic effects and other more comprehensive but computationally expensive models. The model is synthesized from static wind tunnel test data and accounts for variations in angular rotor speed, apparent wind speed, and angle of attack on the vehicle's rotors. Experimental validation is performed through indoor free-flight pitch step responses. On average, the simulated root mean square pitch tracking error is found to be within 31% of the experiment. The usefulness of the whole system to determine station-keeping performance is assessed by free-flight station-keeping experiments in a wind tunnel environment. The tracking error along all three axes is found to increase with increasing mean wind speed. The simulated standard deviation of errors along the wind direction is on average within 7% of the experiment, whereas those of the errors along axes perpendicular to the wind direction are found to be within 56%. Overall trends are similar in all cases, making the model suitable for performance comparisons of flight controllers.