A survey of safe landing zone detection techniques for autonomous unmanned aerial vehicles (UAVs)

•Different types of landing zones for autonomous Unmanned Aerial Vehicle.•Classification of landing zone detection techniques.•Existing methods for safe landing zone detection techniques. The age of automation is upon us. Few decades earlier, nearly all the flying vehicles were human-controlled. Nowadays, almost every air vehicle is partially automated or getting closer to full automation. This race towards full automation has led to the introduction of features like autopilot. Unmanned aerial vehicles (UAVs) are the tiniest version of all types of air vehicles. The widespread usage of autonomous UAVs has spawned the need for safe landing zone (SLZ) detection techniques for UAV landing. A SLZ detection becomes an important face of a mission when the UAV needs emergency landing due to the technical difficulties or adverse weather conditions on the way of its operation. Before directly proceeding for landing, a UAV has to decide whether the landing zones are safe or not. On-board visual sensors provide potential information of the ground surface in the form of image or signal. Different image processing and safe landing area detection (SLAD) algorithms are then used to identify the best possible landing sites from the input data. In this survey, we discuss indoor and outdoor landing zone detection techniques. We further classify outdoor landing zones as either static or dynamic and discuss existing literature in the specific categories. We critique the shortcomings of existing SLZ detection techniques while also acknowledging their contributions. Further, we point to potential areas of improvement and future directions of the safe landing zone detection algorithms we surveyed. This survey paper may be a useful tutorial for understanding the types of landing zones and landing zone detection techniques for the UAVs, the strengths of zone detection algorithms, and the open areas for future improvement and research.