Autonomous UAV Navigation in GPS-denied environments using AprilTags
Advisor Information
Prithviraj Dasgupta
Location
Dr. C.C. and Mabel L. Criss Library
Presentation Type
Poster
Start Date
4-3-2016 12:45 PM
End Date
4-3-2016 2:15 PM
Abstract
Unmanned Aerial Vehicles (UAVs) are being increasingly used in many applications such as aerial surveillance in extreme environments like deserts, icy regions, agricultural surveying, ocean monitoring, etc. However, they are technologically restricted to outdoor environments as UAVs use Global Positioning Systems (GPS) for localization and GPS do not work indoors. So, autonomous flight in a GPS-denied environment is a challenging problem for the UAVs. Hence, my research is to investigate solutions to the UAV indoor localization problem using a fiducial marker-based system called AprilTags. An AprilTag is a pattern of dark and light squares similar to a QR code but easier for cameras to detect robustly and at a distance. The idea is to pre-install markers embedded with AprilTags at different locations in an indoor environment. Each tag will encode the current three-dimensional (3D) coordinates of the location of the tag as well as direction that the UAV should move toward, after it detects the tag. Appropriate processing will be done on the Laptop and communicated with the UAV through a port to detect and analyze the direction encoded in the tag and maneuver itself towards the prescribed direction. The advantage of my proposed approach is that it is very economical and does not require additional sensors or expensive indoor positioning systems. Several real-life applications of my research include inspection of high-risk, indoor environments, such as nuclear power plants, unmanned search and rescue, autonomous inspection of health of buildings and other structures following natural disasters such as earthquakes.
Autonomous UAV Navigation in GPS-denied environments using AprilTags
Dr. C.C. and Mabel L. Criss Library
Unmanned Aerial Vehicles (UAVs) are being increasingly used in many applications such as aerial surveillance in extreme environments like deserts, icy regions, agricultural surveying, ocean monitoring, etc. However, they are technologically restricted to outdoor environments as UAVs use Global Positioning Systems (GPS) for localization and GPS do not work indoors. So, autonomous flight in a GPS-denied environment is a challenging problem for the UAVs. Hence, my research is to investigate solutions to the UAV indoor localization problem using a fiducial marker-based system called AprilTags. An AprilTag is a pattern of dark and light squares similar to a QR code but easier for cameras to detect robustly and at a distance. The idea is to pre-install markers embedded with AprilTags at different locations in an indoor environment. Each tag will encode the current three-dimensional (3D) coordinates of the location of the tag as well as direction that the UAV should move toward, after it detects the tag. Appropriate processing will be done on the Laptop and communicated with the UAV through a port to detect and analyze the direction encoded in the tag and maneuver itself towards the prescribed direction. The advantage of my proposed approach is that it is very economical and does not require additional sensors or expensive indoor positioning systems. Several real-life applications of my research include inspection of high-risk, indoor environments, such as nuclear power plants, unmanned search and rescue, autonomous inspection of health of buildings and other structures following natural disasters such as earthquakes.