Positioning and recovering of fixed-wing UAV using updating final-state control and adaptive nonstationary control

Abstract

Descent deceleration using parachutes is widely used in the aerospace field for scenarios such as weather observation using the descent of high-altitude balloons (radiosondes), re-entry during space sample return, and emergency drone landings. These parachutes are used to protect the aircraft and samples, and to avoid hazards with objects and humans on the ground. However, it is difficult to control the dropping point using parachutes, and there have been concerns regarding accidents caused by insufficient deceleration and adverse environmental effects caused by the disposal of unrecovered objects. To solve these problems, we propose a control method for the aerial recovery of low-speed descent objects using fixed-wing unmanned aerial vehicles (UAVs). Our previous work has shown a flight trajectory generation method that employs updating final-state control (UFSC) to handle variations in the terminal target states until the object is recovered. The frequency-shaped optimal regulator control that enables a return to steady-state flight has also been used after UFSC. However, owing to the characteristics of each control method, the problem of sudden fluctuations in the control input before and after recovery of a descending object remained, and performing aerial recovery using an actual aircraft was predicted to be difficult. In this study, this problem is avoided by switching from UFSC to adaptive nonstationary control (ANC), which has proposed in previous studies. ANC can be applied when the aircraft approaches a descending object, and it can be confirmed that the control input is smooth in a series of trajectory designs until return to steady-state flight.

Keywords

• adaptive nonstationary control
• aerial recovery
• final-state control
• mode switching control
• optimal control
• trajectory generation
• unmanned aerial vehicle (uav)