Vision‐assisted deep stall landing for a fixed‐wing UAV

• Published in: Journal of field robotics 2022-10, Vol.39 (7), p.1138-1152
• Main Authors: Kim, Doyoung, Park, Sanghyuk
• Format: Article
• Language: English
• Subjects: Airspeed; Algorithms; Confined spaces; Control methods; deep stall landing; fixed‐wing UAV; Flight tests; Landing; Maneuvers; Mouse devices; Stalling; Tracking; Unmanned aerial vehicles; vision‐assisted guidance
• ISSN: 1556-4959; 1556-4967
• DOI: 10.1002/rob.22100

This paper reports a guidance and control method for vision‐assisted deep stall landing for a fixed‐wing unmanned aerial vehicle (UAV). With a deep stall maneuver, a fixed‐wing UAV can land on a confined area with a deep flight path angle. This paper combines vision‐assisted guidance for approach and deep stall maneuvers for intuitive fixed‐wing UAV operation. A target landing area is chosen and tracked by a human operator with a computer mouse on the aerial image. Then, a line‐of‐sight vector is created to generate a guidance command. The proposed method is composed of two phases: a vision‐assisted approach (Phase I) and vision‐assisted deep stall landing (Phase II). In Phase I, the UAV is guided toward a mouse‐tracked landing area while maintaining the flight speed and altitude. In Phase II, the deep stall is initiated by a deep stall start algorithm, and the UAV is guided to the mouse‐tracked target landing area. The feasibility of the suggested method is verified by a series of flight tests, where the testbed UAV successfully landed on a 10 m × 10 m target.