An Adaptive Flare Scheme for Autonomous Landing of a Fixed-Wing UAV

Abstract

Despite significant improvements in autonomous flight control systems, the automatic landing of a UAV still possess significant technical challenges, especially during the flare mode, which is usually considered as a crucial part of landing. Flare starts right after the glide slope phase at about a few tenths of feet above ground and lasts till touchdown. Traditionally, in flare a predefined fixed profile of altitude descent is followed, which is usually an exponential function. However, a downside of this scheme is that the altitude at which flare is initiated is preset. This requires the UAV to reach this altitude at the end of the glide slope before flare is commenced. In case of any uncertainties or disturbances, if the UAV reaches the end of the glide slope phase at a slightly higher altitude, starting of flare mode is delayed, which may lead to go-around as the only viable option, especially on shorter runways. This paper presents an adaptive flare scheme to address the shortcomings of the traditional flare scheme. The proposed scheme is designed to be adaptive towards bounded variations in flight parameters such as velocity, altitude and flight path angle etc., at the beginning of the flare. Efficacy of the proposed scheme is demonstrated by the results based on simulations incorporating bounded variations in flight parameters.