Altitude data fusion utilising differential measurement and complementary filter

Abstract

In plenty of researches and real systems, barometer and accelerometer which can provide altitude and acceleration measurements, respectively, are utilised in altitude estimation of small unmanned aerial vehicle. These sensors can draw on their merits and improve the accuracy of altitude estimation. The altitude measurement of barometer has the drift problem as one of its disadvantages which could deteriorate the performance of altitude estimation. In addition, barometer and accelerometer both have stochastic error which reduce the accuracy of altitude estimation. A novel scenario is proposed by utilising differential altitude measurement and complementary filter to deal with above issues. It is accurate and suitable for real-time application. In this scenario, altitude drift is reduced by using differential altitude measurement, and the stochastic error of altitude is minimised by using complementary filter algorithm. Stair and flight experiments are implemented to test this scenario. The experimental results indicate that it can provide good dynamic performance and flexibility for altitude estimation.