A dynamic data-driven stochastic state-awareness framework for the next generation of bio-inspired fly-by-feel aerospace vehicles

In this work, a novel Data-driven State Awareness (DSA) framework is introduced for the next generation of intelligent “fly-by-feel” aerospace vehicles. The proposed framework is based on two entities: (i) bio-inspired networks of micro-sensors that can provide real-time information on the dynamic aeroelastic response of the structure and (ii) a stochastic “global” identification approach for representing the system dynamics under varying flight states and uncertainty. The evaluation and assessment of the proposed DSA framework is based on a prototype bio-inspired self-sensing intelligent composite wing subjected to a series of wind tunnel experiments under multiple flight states. A total of 148 micro-sensors, including piezoelectric, strain, and temperature sensors, are embedded in the layup of the composite wing in order to provide the sensing capabilities. A novel data-driven stochastic “global” identification approach based on functionally pooled time series models and statistical parameter estimation techniques is employed in order to accurately interpret the sensing data and extract information on the wing aeroelastic behavior and dynamics. The method’s cornerstone lies in the new class of Vector-dependent Functionally Pooled (VFP) models which allow for the analytical inclusion of both airspeed and angle of attack (AoA) in the model parameters and, hence, system dynamics. Special emphasis is given to the wind tunnel experimental assessment under various flight states, each defined by a distinct pair of airspeed and AoA. The obtained results demonstrate the high achievable accuracy and effectiveness of the proposed state-awareness framework, thus opening new perspectives for enabling the next generation of “fly-by-feel” aerospace vehicles.