35312 - Automatic Flight Control (2nd cycle)

Learning outcomes

The course can be divided into two parts:

1. In the first part multivariable synthesis methods are presented. In particular the optimal control theory both for deterministic and stochastic systems. The case of linear time invariant systems with quadratic cost index (Linear Quadratic and LQ Gaussian case) is treated deeply. It is worth observing that also the basic theory of observers and optimal observers (Kalman Filter) is furnished.

Starting from a review of Dynamic Flight notions the classical autopilot design methods are firstly proposed. Afterwards, the design of Stability Augmentation Systems (SAS), Attitude Control Systems (ACS) and Flight Path Control Systems (FPCS) by means of optimal control methods (LQ and LQG) are proposed.

Course contents

Part 1

1. Optimal Control Law. Optimal Control Problem definition, Hamiltonian function, solution of the LQ optimal control problem; solution of the minimum energy and minimum time optimal control problem; design of the cost parameters; Pontryagin maximum principle.
2. Feedback Optimal Control. Solution of the optimum control problem: Differential Riccati equation and solution existence and uniqueness conditions, optimal feedback gain. Steady state optimal regulator: algebraic Riccati equation, stability of the steady state regular. Set points different from the origin.
3. Poles assignments. Coordinate change in input, state and output space. Canonical forms and poles assignment for MIMO systems
4. State observers and dynamic feedback. Identity observer, sub-optimal control problem by means of dynamic feedback
5. Notes on probability and stochastic process theory
6. Stochastic Optimal Observer. Optimal state estimate, Kalman Filter, differential Riccati equation and existence and uniqueness conditions.
7. Stochastic Optimal Regulator. Cost index, LQG regulator.
   

Part 2

1. Flight Dynamics notes. Six degree of freedom rigid body model. Military Flying qualities. Longitudinal and lateral-directional dynamics and model in case of coupled dynamics. Linearized models: longitudinal and lateral-directional modes. Dryden turbulence description and wind gust model.
2. Classical Autopilots. Stability Augmentation Systems (SAS): pitch rate SAS and other longitudinal dynamic SAS lateral-directional SAS: yaw dumper, roll rate dumper, spiral mode stabilization. Attitude Control Systems (ACS): pitch ACS, roll angle ACS, wing leveler, sideslip suppression ACS, turn coordination ACS. Flight Path Control System (FPCS): altitude hold system, speed control system, direction control system, heading control system, VOR-coupled automatic tracking system, ILS localizer coupled control system, ILS glide-path-coupled control system, automatic landing system
3. Design of Autopilot by means of Optimal Control LQ and LQG methods. Steady state control: determination of the inputs (command surfaces deflection and throttle) corresponding to a given steady state flight condition; determination of the optimal gain satisfying the flight qualities. Methods to arm and engage autopilots. Vertical mode autopilots (case of airbus 319/320/321) : climb rate and airspeed hold, altitude capture and hold, glide slope capture and hold, flare and touchdown. Lateral mode autopilots (case of airbus 319/320/321) : bank angle and sideforce ACS, heading capture and hold, track capture and hold, localizer capture and hold. Gain scheduling methodology. Notes on guidance methods: navigation mode. Notes on the design of an integrated navigation, guidance and control system. Notes on control of missiles. Notes on model based linear and non linear Fault Detection and Isolation methods.
   

Readings/Bibliography

Part 1

M. Tibaldi. Progetto di Sistemi di Controllo.Pitagora Editrice. Bologna
B.D.O. Anderson, J.B. Moore. Optimal Control: Linear Quadratic Methods. Prentice Hall Information and System Sciences Series

Part 2

Notes of the teacher (downloadableat AMS Campus)

D. McLean. Automatic Flight Control Systems. Prentice Hall Series in Sytems and Control Engineering
A.E. Bryson, Jr. Control of Spacecraft and Aircraft. Princeton University Press.

Teaching methods

Normal lessons in classroom plus laboratory autopilot design tested on matlab/simulink flight simulator.

Assessment methods

Witten and oral tests plus autopilot design in Matlab/Simulink

Teaching tools

matlab/Simulink software; Automatic Flight Control hangar laboratory.

Office hours

See the website of Paolo Castaldi