2021/09/30 by Pranay Mathur, Yash Jangir, Mathur, Pranay +3
Computer Science · Engineering · #FOS: Computer and information sciences #Robotic Path Planning Algorithms #Robotics (cs.RO) #Robotics and Sensor-Based Localization #Underwater Vehicles and Communication Systems
paper · pdf · doi:10.48550/arxiv.2109.15114
openalex publication_date 2021/09/30 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Autonomous landing systems for Micro Aerial Vehicles (MAV) have been proposed\nusing various combinations of GPS-based, vision, and fiducial tag-based\nschemes. Landing is a critical activity that a MAV performs and poor resolution\nof GPS, degraded camera images, fiducial tags not meeting required\nspecifications and environmental factors pose challenges. An ideal solution to\nMAV landing should account for these challenges and for operational challenges\nwhich could cause unplanned movements and landings. Most approaches do not\nattempt to solve this general problem but look at restricted sub-problems with\nat least one well-defined parameter. In this work, we propose a generalized\nend-to-end landing site detection system using a two-stage training mechanism,\nwhich makes no pre-assumption about the landing site. Experimental results show\nthat we achieve comparable accuracy and outperform existing methods for the\ntime required for landing.\n