The current research represents a first steps towards developing a decentralized network

of autonomous, intelligent and inexpensive unmanned aerial vehicles (UAV), which could

be used for a variety of scientific missions where measurements from a distributed net-

work of nodes could significantly improve the prediction. First we present a ground robot

and a UAV implementation of Payload Directed Flight showing how a camera sensor

can be used for guidance and navigation. Secondly we present an adaptive airborne sen-

sor network, which fuses the onboard sensors information acquired from multiple agents

to monitor environmental processes over space and time. We use a fleet of small and

affordable unmanned aerial vehicles (UAVs) as the carrier platform for the mobile net-

work nodes. The network will be able be continuously reconfigure in a tridimensional

space according to the circumstances (e.g., continuously evolving scientific phenomena)

to optimize the location of individual nodes. As proof of concept and validation, we will

apply the proposed sensing approach to monitoring diffusing volcanic plumes. This pro-

posed work will (i) develop fast converging mathematical algorithms that can predict the

volcanic plume in real time using single and multiple autonomous agents. (ii) validate

the the a proposed algorithms using hardware in the loop computer based simulations

and physical autonomous agents. Lastly we present a sensor fusion algorithm for UAV

navigation in GPS degraded environments.