Modeling and Adaptive Control for Multirotor Subject to Thruster Dynamics

Abstract

Research on high-speed flight dynamics of multirotors is in increasing demand for “quick-reach” missions such as short-distance delivery and disaster assessment. The conventional works have integrated a linear thruster model in the dynamics modeling stage and developed a corresponding control algorithm; however, the linear thruster assumption has been shown to be accurate only in a near-hover state through wind tunnel tests. This paper proposes a novel multirotor dynamics model incorporating blade element momentum theory (BEM), which can precisely predict transient thruster output in various flight states (e.g., no flow, oblique flow, and pure side flow). With the proposed dynamics model, degradation of control performance caused by disturbances and variations is noticed under typical flight state (e.g., high-speed forward flight and drastic vertical rising). Finally, an adaptive robust backstepping control algorithm is proposed to achieve guaranteed performance under the aforementioned variations and disturbances. The simulations are carried out on a small quadrotor to verify the proposed approach.