HR: 16:35h
AN: IN14A-03 INVITED     [Abstracts]
TI: Cooperative Mobile Sensing Systems for In Situ Measurements in Hazardous Environments
AU: * Argrow, B
EM: brian.argrow@colorado.edu
AF: Research and Engineering Center for Unmanned Vehicles, Aerospace Engineering Sciences, University of Colorado, Boulder, CO 80309 United States
AB: Sondes are typically deployed from manned aircraft or taken to altitude by a balloon before they are dropped. There are obvious safety and physical limitations that dictate where and how sondes are deployed. These limitations have severely constrained sonde deployment into highly dynamic and dangerous environments. Additionally, conventional parachute dropsondes provide no means for active control. The "smartsonde" idea is to integrate miniature sonde packages into micro air vehicles (MAVs). These MAVs will be ferried into the hard to reach and hazardous environments to provide in situ measurements in regions that have been heretofore out of reach. Once deployed, the MAV will provide some means of control of the sonde, to enable it to remain aloft and to provide some measure of directional control. Preliminary smartsonde communications experiments have been completed. These experiments focused on characterizing the capabilities of the 802.11.4 wireless protocol. Range measurements with 60-mW, 2.4-GHz radios showed 100% throughput rate over 2.7 km during air to ground tests. The experiments also demonstrated the integration of an in-house distributed computing system that provides the interface between the sensors, UAV flight computers, and the telemetry system. The University of Colorado's Research and Engineering Center for Unmanned Vehicles (RECUV) is developing an engineering system that integrates small mobile sensor attributes into flexible mobile sensor infrastructures to be deployed for in situ sensing in hazardous environments. There are three focus applications: 1) Wildfire, to address sensing, communications, situational awareness, and safety needs to support fire-fighting operations and to increase capabilities for dynamic data acquisition for modeling and prediction; 2) Polar, where heterogeneous mixes of platforms and sensors will provide in-situ data acquisition from beneath the ocean surface into the troposphere; 3) Storm, to address the challenges of volumetric in-situ data acquisition in the extremely dynamic environments of severe storms. The common thread among these applications is the need for a cooperative mobile sensing system, where sensor packages are integrated into custom platforms that enable targeting of areas of interest through the cooperative control, with varying levels of autonomy, of small unmanned vehicles. RECUV has demonstrated mobile ad hoc networks using WiFi (802.11b) radios simultaneously deployed in fixed and mobile ground nodes and unmanned aerial vehicles (UAVs). Recently, an autonomous UAV was deployed with a miniature sensor package that returned real-time temperature, pressure, and humidity data, through the ad hoc communications network. The UAV demonstrated the ability to autonomously make flight-path decisions based on the sensor data that was monitored by the flight computer. Current work is now focused on integrating the sensor package into a smartsonde to be deployed from a UAV mothership. Benign scenarios for upcoming tests to validate the collaborative mobile sensing system paradigm include scenarios with features similar to those that will be encountered in the hazardous and dynamic environments a of the wildfire, polar, and storm applications. These include a fly-through of a dust devil on the planes of eastern Colorado and deployment of a dual-mode smartsonde that transmits at high data rates while airborne then, upon landing, it switches to quiet, power-saving mode , where in situ data is logged and only transmitted when the sonde package is queried during overflights of a UAV mothership.
UR: http://recuv.colorado.edu
DE: 9800 GENERAL OR MISCELLANEOUS
SC: Earth and Space Science Informatics [IN]
MN: Fall Meeting 2005