Adaptive Neural Control for Active Flutter Suppression

Abstract

Abstract Under a joint research and development effort conducted by McDonnell Douglas Aerospace (MDA) and the National Aeronautics and Space Administration, Langley Research Center (NASA LaRC), a neural network-based adaptive control system has been developed and demonstrated for active wing flutter suppression. The adaptive control system, which uses a neural network embedded within a Model Predictive Control (MPC) framework, is referred to as Neural Predictive Control (NPC). During Phase II of the Adaptive Neural Control of Aeroelastic Response (ANCAR) program, the NPC system was applied for active flutter suppression using NASA’s Benchmark Active Controls Technology (BACT) wing model in the Transonic Dynamics Tunnel (TDT). The ability of the NPC system to self-configure a control law for stabilizing an unstable wing during flutter was demonstrated. This paper defines the basic architecture of the NPC approach and discusses the flutter suppression results obtained during the wind tunnel demonstration.