2018/09/15 by Michael Warren, Melissa Greeff, Warren, Michael +10 · 2 citations
Engineering · Computer Science · #Robotics and Sensor-Based Localization #Robotic Path Planning Algorithms #Advanced Vision and Imaging
paper · pdf · doi:10.48550/arxiv.1809.05757
Redundant navigation systems are critical for safe operation of UAVs in\nhigh-risk environments. Since most commercial UAVs almost wholly rely on GPS,\njamming, interference and multi-pathing are real concerns that usually limit\ntheir operations to low-risk environments and Visual Line-Of-Sight. This paper\npresents a vision-based route-following system for the autonomous, safe return\nof UAVs under primary navigation failure such as GPS jamming. Using a Visual\nTeach & Repeat framework to build a visual map of the environment during an\noutbound flight, we show the autonomous return of the UAV by visually\nlocalising the live view to this map when a simulated GPS failure occurs,\ncontrolling the vehicle to follow the safe outbound path back to the launch\npoint. Using gimbal-stabilised stereo vision alone, without reliance on\nexternal infrastructure or inertial sensing, visual odometry and localisation\nare achieved at altitudes of 5-25 m and flight speeds up to 55 km/h. We examine\nthe performance of the visual localisation algorithm under a variety of\nconditions and also demonstrate closed-loop autonomy along a complicated 450 m\npath.\n