Using Alpha-beta filtration for robustness improvement of a quadrocopter positioning system

Quadrocopter is an unmanned aerial vehicle (UAV) platform. The position of the robot is determined based on readings from an accelerometer and a gyroscope, but the measurement signals contain broadband noise. This article describes a solution for filtering out the noise based on an Alpha - beta filter. It also presents the methodology of designing and implementing such a filter for noise cancellation in measurement signals from accelerometer and gyroscope. The computational complexity of the Alpha - beta filter is similar to a second order low pass filter, but there are no noticeable time delays and filtration gives similar results to that of a Kalman filter. The described filter was designed and implemented in a real vehicle, tested and validated.