Singular versus boundary arcs for aircraft trajectory optimization in climbing phase

Abstract

Verbatim from arXiv. Not paraphrased, not summarized.

In this article, we are interested in optimal aircraft trajectories in climbing phase. We consider the cost index criterion which is a convex combination of the time-to-climb and the fuel consumption. We assume that the thrust is constant and we control the flight path angle of the aircraft. This optimization problem is modeled as a Mayer optimal control problem with a single-input affine dynamics in the control and with two pure state constraints, limiting the Calibrated AirSpeed (CAS) and the Mach speed. The candidates as minimizers are selected among a set of extremals given by the maximum principle. We first analyze the minimum time-to-climb problem with respect to the bounds of the state constraints, combining small time analysis, indirect multiple shooting and homotopy methods with monitoring. This investigation emphasizes two strategies: the common CAS/Mach procedure in aeronautics and the classical Bang-Singular-Bang policy in control theory. We then compare these two procedures for the cost index criterion.

Authors

• Olivier Cots IRIT-APO, IRIT
• Joseph Gergaud IRIT-APO, IRIT
• Damien Goubinat TAES
• Boris Wembe IRIT-APO, IRIT

Keywords

• math.OC