An approach for augmenting existing distributed controller designs for large-scale interconnected systems with neural network based adaptive elements is proposed. It is assumed that the controllers are interconnected in the same way as the plant and based on the available measurements, a single hidden layer neural network is introduced for each subsystem to partially cancel the effects of the sub-system interconnections and modeling errors on tracking performance. Boundedness of error signals is shown through Lyapunov's direct method. Effectiveness of the approach is demonstrated on a wind tunnel experiment representing aerial vehicles flying in a highly aerodynamically coupled formation. Copyright © 2005 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.