The Georgia Tech Aerial Robotics (GTAR) team finished first in the 2008 International Aerial Robotics Competition at the McKenna Soldier Battle Lab facility in Fort Benning, Georgia. Georgia Tech's aerial vehicle completed the greatest portion of the prescribed mission, completing the first, second and part of the final phase of the mission. This mission must be completed by an aerial vehicle system with no assistance from human operators during mission attempts, representing a difficult engineering challenge.

"This competition includes one of the most difficult missions of any engineering competition," said Eric Johnson. "To attempt it, we came with a system that combined proven components developed over the past six years with some new components that were put together over the past year in a challenging system integration effort."

The GTMax-helicopter is based on the Yamaha RMAX helicopter. It was equipped with two general purpose computers, differential GPS, inertial navigation and two cameras. The slung-load system consisted of redundant release mechanisms, a bump-switch to detect hitting a wall to enable retries, a data-link relay and a magnetometer to measure its twist. The rover system included a high-resolution camera, a general-purpose computer to control driving and manage images, and infrared sensors to enable it to effectively move around rooms indoors.

Software was developed by the team for helicopter navigation and flight control, three different image processing and tracking systems (locating correct building based on sign, locating openings and tracking the opening during approach to the window), automated mission management and simulation tools.

"We were delighted that all parts of the system were demonstrated multiple times during our mission attempts," said Johnson. "My worst fear was that the initial part of the mission would fail and all the work that went into later phases would be for naught - like a rover designed to drive on Mars that fails to get off the launch pad."

The Georgia Tech team made four attempts at the complete mission. On all four of these attempts, the GTAR research UAV system (referred to as the GTMax, a small helicopter) automatically flew a three-kilometer flight to a small village and flew a search pattern, looking at the exterior walls of the buildings within. It automatically located a pre-specified sign on one of the buildings, identifying the correct "building of interest" on all but one of the attempts.

The vehicle then automatically flew a search pattern looking for openings into the building. Having selected a suitable opening, it then dropped a 12-foot-long boom on a 90-foot two-wire slung load, allowed to gently descend by use of a spool/damper system. A camera on the boom was then used to steer the boom to the opening on the building. The boom missed the opening in all three attempts, coming within feet of an open door on one attempt.

The plan was to have a small ground robot drop from the boom inside the opening. This rover would then drive within the building and take a picture of a specific item to complete the mission. Due to missing the openings, the rover was dropped outside the building on these attempts, and so it drove around outside the building transmitting images relayed by the 'mother ship' helicopter back to the launch point.

"To finish the mission completely in the required time would have been great, but we are completely happy with the first place finish," said Johnson. "The mission itself can now retire undefeated, for it will be something different next year."

This is the final year for this contest mission, one that teams have attempted since 2001. Since no team completed the entire mission, $80,000 in prize money was distributed among the teams according to how far their system progressed in the mission in 2008. Having come the closest, the Georgia Tech team, getting the closest, will receive a $27,200 prize for its performance.

The competition was sponsored by the Association for Unmanned Vehicle Systems International. The team consists of graduate and undergraduate students from Aerospace and Electrical & Computer Engineering. The team wishes to thank other sponsors of GTAR 2008: Lockheed Martin, Adaptive Flight Inc., and NovAtel.