Flight Management System Pathfinding Algorithm for Automatic Vertical Trajectory Generation

Ramon Andreu Altava (1) , Jean Claude Mere (1) , Daniel Delahaye (2) , Thierry Miquel (2)

1 Airbus Operation S.A.S.
2 ENAC - Ecole Nationale de l'Aviation Civile

Air Traffic has evolved over the years with current commercial fleet expected to double in the near future, increasing airspace congestion. Jet fuel prices volatility, market competition and drones integrating the same airspace require a modernization of the Air Transportation System. In this context, Airbus proposes modern avionics and cockpit designs to fulfil these requirements. Flight Management System (FMS), in-service since the early 1980s, is one of these candidate avionic systems to be revisited. During high workload flight phases as descent and approach are, best-in-class FMSs compute a fixed vertical reference trajectory based on a lateral flight plan and applicable procedures. Then, flight guidance is responsible for tactical replanning so that deviations are absorbed. FMS reference profile is constructed through a series of hypotheses, which could be enhanced by taking into account other factors such as aircraft optimal energy repartition, ATC restrictions, surrounding traffic, wind errors and mass biases. This paper proposes and compares two methodologies based on a modified version of A* algorithm that solves the Optimal Control Problem in the vertical plane. Fixed and variable speed aircraft trajectories are compared in order to quantify the benefits with respect to current FMS design. The problem is formulated through a relaxed point-mass model with a real performance database for a modern commercial aircraft. It accounts for flight constraints as well as aircraft dynamics for trajectory generation. Fuel consumption is optimized without excessively penalizing flight time. These trajectories are compared with those generated by Airbus FMS simulator. Results show that, for modern arrival procedures such as those defined for Continuous Descent Operations, aircraft energy management (potential and kinetic) is enhanced, producing continuously idle trajectories that consume up to 30 % less fuel than current operations, as obtained for this particular procedure. In that case, flight time would be stretched by few minutes yielding to 15% extra time, whose trade-off still seems interesting from the airline perspective.

Mots clés: Pathfinding; Graph Search; Optimal Control; Aircraft Trajectory Optimization

Domaines: Automatique

Ramon Andreu Altava, Jean Claude Mere, Daniel Delahaye, Thierry Miquel. Flight Management System Pathfinding Algorithm for Automatic Vertical Trajectory Generation. DASC 2018, 37th AIAA/IEEE Digital Avionics Systems Conference, Sep 2018, Londres, United Kingdom. ⟨10.1109/DASC.2018.8569254⟩. ⟨hal-01886355⟩