Research on relative height measurement based on multi-sensor fusion technology

Real-time relative height measurement is very important for control system of agricultural plant protection UAV. In the paper we research a multi-sensor relative height fusion algorithm which combined radio altimeter, barometric altimeter, GPS satellite navigation receiver and inertial measurement unit. According to association with GPS speed value, a method which base variable coefficient is proposed to eliminate outlier errors of radio altimeter and an algorithm based half-step interpolation is used to reconstruct the radio altimeter data. A Kalman filter is used for fusion the multisensors measurement value and give the optimal estimation of relative height of UAV. Through the experiment data, the method can eliminate 95% outlier error, and the height measurement error is reduced to 0.15m. The method can be used in super low-altitude plant protection UAV flight control system.