Unscented Kalman Filter for an orientation module of a quadrotor mathematical model

Abstract

The article describes the Unscented Kalman Filter, used in an orientation module of the quadrotor. Control of the quadrocopter is an important issue. Since it belongs to the group of the UAVs (Unmanned Aerial Vehicle), quadrotor must pass safety requirements in order to be used in urban spaces. Here, authors are solving one of the most important problems: estimation of the state vector of the quadrotor. Multirotors like quadrotor can work in semi autonomous or full autonomous mode. Regardless of the used mode, at least one control loop must be always on: an orientation loop. Normally a regulator responsible for an orientation stabilization is based on measurements from IMU (Inertial Measurement Unit). Here, an extended version of the state estimator, based on IMU readings and prediction from the model is introduced. Authors have implemented the Unscented Kalman Filter (firstly described in [1]) and proved that, this estimator is suitable in case of UAV.