Southern California Conferences for Undergraduate Research

Southern California Conferences for Undergraduate Research

RF Source Tracking with Unmanned Autonomous Vehicles


Michael Grino


  • Francois Quitin, Post-doctoral Researcher, University of California Santa Barbara
  • Upamanyu Madhow, Professor of Electrical Engineering, University of California Santa Barbara

In the event of a natural disaster where the cellular infrastructure is wiped out, an Unmanned Autonomous Vehicle (UAV) could be used to track victims by seeking the source of their cell phone signal and providing them with emergency connectivity to the network. In this context, we consider the problem of having a UAV locate an RF source using only Received Signal Strength (RSS) measurements. The UAV is unaware of its initial position relative to the RF source, and uses no additional knowledge of its surroundings other than the RSS to find the source.
The difficulty with using RSS to locate an RF source is that power does not always decrease monotonically with distance, but shows random variations due to multipath propagation of radio waves. Therefore, traditional gradient-based algorithms are not effective.
A new iterative algorithm is proposed that uses the radiation pattern of the UAV’s antenna to find the direction of the RF source. Using this algorithm, the UAV is consistently able to locate the source, with a success rate of 80% for an initial UAV-source distance of 25m, and an error in locating the source direction of only 2.93° at a distance of 15m.
These results indicate that it is possible for a UAV to locate a randomly placed RF source, utilizing only knowledge of RSS. Future work will include using RSS to have a UAV follow an RF source, or to use UAVs to provide connectivity between distant source and destination nodes.

Presented by:

Michael Grino


Saturday, November 17, 2012




Broome Library

Presentation Type:

Poster Presentation