2021/12/02 by Nicholas Polosky, Tyler Gwin, Polosky, Nicholas +7
Computer Science · Engineering · #Advanced Neural Network Applications #Air Traffic Management and Optimization #FOS: Computer and information sciences #Machine Learning (cs.LG) #Robotics (cs.RO) #UAV Applications and Optimization
paper · pdf · doi:10.48550/arxiv.2112.01688
openalex publication_date 2021/12/02 · openalex created_date 2022/11/12 · openalex updated_date 2026/07/28
Interest in unmanned aerial system (UAS) powered solutions for 6G\ncommunication networks has grown immensely with the widespread availability of\nmachine learning based autonomy modules and embedded graphical processing units\n(GPUs). While these technologies have revolutionized the possibilities of UAS\nsolutions, designing an operable, robust autonomy framework for UAS remains a\nmulti-faceted and difficult problem. In this work, we present our novel,\nmodular framework for UAS autonomy, entitled MR-iFLY, and discuss how it may be\nextended to enable 6G swarm solutions. We begin by detailing the challenges\nassociated with machine learning based UAS autonomy on resource constrained\ndevices. Next, we describe in depth, how MR-iFLY's novel depth estimation and\ncollision avoidance technology meets these challenges. Lastly, we describe the\nvarious evaluation criteria we have used to measure performance, show how our\noptimized machine vision components provide up to 15X speedup over baseline\nmodels and present a flight demonstration video of MR-iFLY's vision-based\ncollision avoidance technology. We argue that these empirical results\nsubstantiate MR-iFLY as a candidate for use in reducing communication overhead\nbetween nodes in 6G communication swarms by providing standalone collision\navoidance and navigation capabilities.\n