Multi-Objective Optimization of an Aircraft Trajectory between Cities using an Inverse Model Approach

Abstract

Increasing air traffic and the need for environmental-friendly flight operations require efficient Air Traffic Management (ATM) operations. The environmental objectives are the reduction of exhaustive emissions of the aircraft engines as well as the emitted noise. On the other hand there are economically and operationally driven objectives like reduction of flight time and fuel consumption. The general approach to simulate and optimize a trajectory from city to city is to use simplified two degrees of freedom aircraft models along with standard gradient-based optimizers. In order to more efficiently deal with the aforementioned objectives, the first contribution of this work is the design of a modular wrapper model suited for trajectory optimization. It incorporates a B-Spline based trajectory parameterization module, path controller and a more detailed inverse aircraft model with three degrees of freedom. With this combination of models, a low computation time could be achieved, whilst detailed information on the aircraft's behaviour is available. The second contribution is a multi-criterial optimization approach using a genetic-algorithm optimizer. It enables the user to efficiently compare and weigh different criteria like fuel consumption or NOx-emission against each other to design a trajectory that yields the best performance for the respective focus. In addition, the multiobjective optimization approach helps to identify possible gains and to determine the interconnections between different environmental related criteria.