Path-Tracking Control of Stochastic Quadrotor Aircraft in Three-Dimensional Space

Abstract

Despite the fact that environmental loads (forces and moments) induced by wind on quadrotor vertical take-off and landing (VTOL) aircraft consist of both deterministic and stochastic components, all existing works on controlling the aircraft either ignore these loads or treat them as deterministic. This ignorance or treatment deteriorates the control performance in a practical implementation. This paper presents a constructive design of controllers for a quadrotor aircraft to track a reference path in three-dimensional (3D) space under both deterministic and stochastic disturbances. A combination of Euler angles and unit-quaternion for the attitude representation of the aircraft is used to result in an effective control design, and to achieve path-tracking control results. Weak and strong nonlinear Lyapunov functions are introduced to overcome difficulties caused by underactuation and Hessian terms induced by stochastic differentiation rule. To overcome the inherent underactuation of the aircraft, the roll and pitch angles of the aircraft are considered as immediate controls. Potential projection functions are introduced to design estimates of the deterministic components and covariances of the stochastic components. Simulations illustrate the results.