2025/04/21 by Emre Saldiran, Mehmet Hasanzade, Aykut Cetin +3 · 2 citations
Computer Science · Engineering · #Aeronautics #Air Traffic Management and Optimization #Artificial intelligence #Automotive engineering #Autonomous Vehicle Technology and Safety #Computer science #Engineering #Mobile robot #Radar #Radar tracker #Real-time computing #Remotely operated underwater vehicle #Robot #Robotic Path Planning Algorithms #Simulation #Telecommunications #Terrain #Vehicle dynamics #Vehicle safety
paper · doi:10.1109/taes.2025.3559899
openalex publication_date 2025/04/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/06/26
The integration of unmanned aerial vehicles (UAVs) with vertical takeoff and landing (VTOL) capability into commercial and military applications marks a significant advancement in aerial technology, necessitating robust systems for safe operation. Performing autonomous safe landing in emergencies remains a critical concern among other challenges. Emergencies in UAVs can arise from various factors such as system failures, adverse weather conditions, or mission-critical situations requiring immediate landing. Addressing this challenge, this paper presents an autonomous safe landing system designed to provide operation time assurance for UAVs flying over unknown terrains. To achieve this, we partition the point cloud data generated by the LIDAR sensor into grids for safe landing site identification and selection. Our approach requires a low computational load and is validated through tests on various terrain types under real-world conditions while utilizing only the on-board sensing and computational capability of the UAV.