2019/12/10 by Francesca Baldini, Baldini, Francesca, Animashree Anandkumar +3
Computer Science · Engineering · #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Robotic Path Planning Algorithms #Robotics (cs.RO) #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.1912.04527
openalex publication_date 2019/12/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this work, we propose a new learning approach for autonomous navigation\nand landing of an Unmanned-Aerial-Vehicle (UAV). We develop a multimodal fusion\nof deep neural architectures for visual-inertial odometry. We train the model\nin an end-to-end fashion to estimate the current vehicle pose from streams of\nvisual and inertial measurements. We first evaluate the accuracy of our\nestimation by comparing the prediction of the model to traditional algorithms\non the publicly available EuRoC MAV dataset. The results illustrate a 25 %\nimprovement in estimation accuracy over the baseline. Finally, we integrate the\narchitecture in the closed-loop flight control system of Airsim - a plugin\nsimulator for Unreal Engine - and we provide simulation results for autonomous\nnavigation and landing.\n