Multi-sensor Fusion Height Prediction Algorithm Based on Kalman Filtering

Unmanned aerial vehicle (UAV) is more and more widely used in several situations, such as geography experiments or even military application. Height estimation is of great importance to control a UAV. However, estimating UAV flying height by a single sensor is always of low accuracy. In order to solve this problem, we will propose a multi-sensor fusion algorithm based on Kalman filtering for UAVs flying height estimation. Then main principle is to fuse the data collected from accelerator, barometer and laser ranging module and then to predict the height of a UAV. This study will verify the validation of the proposed fusion algorithm on the flying control platform based on Pixhawk. According to the comparative experiment, the height values predicted by the proposed algorithm are convergent. The flying experiment indicates that applying the proposed algorithm to positioning estimation problem can achieve a stable position control of UAVs.