One of the most important factors in the successful application of unmanned vehicles is the ability to determine where a vehicle is in relation to its local environment. While GPS technology works well in most locations, the satellite coverage necessary for this technology to function is not available in cluttered environments and inside buildings. Current technological workarounds involve high resolution image processing, which is expensive in terms of both monetary and computational cost. Our project team is working to develop an alternative localization solution for drones that uses low cost lasers and off-the-shelf components. The technique involves setting up two or more "lighthouse" beacons in a controlled environment. The beacons will pulse invisible infrared light from LEDs and sweep vertical sheets of infrared lasers around the environment. Photodiodes installed on the UGV will respond to these light sources and pass this information to a processor. Based on the timing of the diode response, the angle between the lighthouse and the diodes can be determined. When combined with ultrasonic altitude data, these angles can be processed to determine the location of the diodes and subsequently the location and orientation of the UGV. The accuracy and precision of the final lighthouse system will be assessed to determine its usefulness for localizing drones in a controlled environment. The end goal of the technical project is to design a system that will facilitate the automated driving of a UGV based on localization data from the lighthouse system while being scalable for future implementation in 3D applications.
[1]
Jeong-Oog Lee,et al.
Vision-Based SLAM System for Small UAVs in GPS-Denied Environments
,
2012
.
[2]
Roman Barták,et al.
Camera-Based Localization and Stabilization of a Flying Drone
,
2015,
FLAIRS Conference.
[3]
Miguel A. Olivares-Méndez,et al.
Visual 3-D SLAM from UAVs
,
2009,
J. Intell. Robotic Syst..
[4]
A Kleusberg,et al.
THE LIMITATIONS OF GPS
,
1990
.