Reconfigurable handling and flying qualities for degrade flight systems using model predictive control

Project D.A. Joosten (Diederick), T.J.J. van den Boom (Ton), M. Verhaegen (Michel)
Keywords:Model predictive control, Data driven and fault tolerant control, Transportation and infrastructure
Sponsored by:STW

We shall develop a framework for the real-time synthesis, simulation and implementation of aircraft
vehicle health and reconfiguration management systems. On the base of a flexible, powerful
and theoretically well founded command and control infrastructure a relevant and cost effective in
depth study on the handling and flying qualities of crippled aircraft can be performed. We shall
make use and, where necessary, extend the latest theoretical results from nonlinear model based
predictive control, linear parameter varying control, reconfiguration and optimal trajectory generation in order to develop an on-line self-repairing command and control systems for crippled aircraft. A system as such will enhance the on-board vehicle functionality up to autonomy extending in
this way the operational safety of crippled aircraft. The innovation in this project in threefold, 1)
in the incorporation of the failure phenomena into the model of the aircraft, 2) the estimation of
the achievable level of performance and 3) the control system reconfiguration to allow flight with
degraded handling qualities. We aim at demonstrating the MPC concept on the flight 1882 El-Al
accident, for which a high fidelity mathematical Boeing-747 model, a complete damage model and
full Black-Box flight data recordings are at our disposition.

© Copyright Delft Center for Systems and Control, Delft University of Technology, 2017.