ROBUST DRONE NAVIGATION
At UASys we aim to bring advanced visual perception and environmental awareness to drones. Our industry-leading visual navigation stack running on our aerial robots gives them that extra edge, allowing them to fly fully autonomously indoors and outdoors without relying on external sources like GPS. Our UAVs can navigate in complex, unstructured real-world environemnts, thanks to multiple exteroceptive sensors and advanced visual perception pipeline.
With our high resolution mapping system and autonomous inspection planning module, industrial inspection has never been this easy.
Cover large tracts of area in a single flight, with flight times of over 40 minutes with a high-resolution mapping payload. 2D and 3D maps of surveyed area available immediately after landing
Optimize crop yield by analyzing soil profile and vegetation data. Our multispectral payloads and long flight times allow generation of complete NDVI maps for use in precision agriculture.
GPS Denied navigation
Our Visual-Inertial Navigation System (VINS) allows drone flight in challenging scenarios where GPS reception is patchy or non-existent. Image data from multiple cameras and an Inertial Measurement Unit (IMU) is fused together to obtain a state estimate for the vehicle when flying in urban/indoor environments with sub-optimal GPS reception.
Using stereo imaging techniques on image data from our multi-camera system, we perform realtime depth extraction and mapping. This local map is used to continuously plan an optimal trajectory for the vehicle to get to its goal.
With all the talk of integrating drones into civilian airspace, there is need for better safety and GPS-agnostic navigation methods – visual navigation and obstacle avoidance is paramount to integrating drones (micro-aerial-vehicles, MAVs) into our cities and elsewhere, because external navigation aids cannot be relied on in every situation, and neither can pilot experience. Visual[…]
Kabir will be giving a talk on Project Artemis at the Embedded Linux Conference 2016, San Diego, CA! This talk is for you if you’re interested in robotics of any kind, and especially for those who are currently working with drones, including developers, systems integrators, pilots and students who wish to take up research in this field in the[…]
MAV2 was flown aggressively to test our visual-inertial estimator setup, and it performed beautifully! Video below. Synchronised IMU and camera frames from the two forward cameras (stereo configuration) are used for state estimation. No GPS!