Quadrotor aircraft attitude estimation and control based on Kalman filter

Quadrotor, as one type of unmanned aircraft vehicles, has gained increasing interests in control area, partially due to its simple aerodynamics and complex dynamics. In this work, a quadrotor system has been constructed with commercial off-the-shelf products. The sensors of inertial measurement unit are micro-electro-mechanical system, whose errors can be analyzed in an auto regressive model. A new attitude estimation scheme based on Kalman filter is proposed, which conducts separate data fusion tasks in a short period and a long period. The proposed attitude sensing method has been validated using the experimental system. In addition, a double-gain proportional differential controller has been designed to regulate attitude dynamics. A satisfactory control performance has been achieved in some test cases.