Collision Avoidance in Commercial Aircraft Free Flight via Neural Networks and Non-Linear Programming

Abstract

In recent years there has been a great effort to convert the existing Air Traffic Control system into a novel system known as Free Flight. Free Flight is based on the concept that increasing international airspace capacity will grant more freedom to individual pilots during the enroute flight phase, thereby giving them the opportunity to alter flight paths in real time. Under the current system, pilots must request, then receive permission from air traffic controllers to alter flight paths. Understandably the new system allows pilots to gain the upper hand in air traffic. At the same time, however, this freedom increase pilot responsibility. Pilots face a new challenge in avoiding the traffic shares congested air space. In order to ensure safety, an accurate system, able to predict and prevent conflict among aircraft is essential. There are certain flight maneuvers that exist in order to prevent flight disturbances or collision and these are graded in the following categories: vertical, lateral and airspeed. This work focuses on airspeed maneuvers and tries to introduce a new idea for the control of Free Flight, in three dimensions, using neural networks trained with examples prepared through non-linear programming.

Does ATM Need Centralized Coordination? Autonomous Conflict Resolution Analysis in a Constrained Speed Environment

“…In the early 2000s, a mathematical model using Mixed Integer Linear Programming, which could be solved by CPLEX and ensured the global optimality of the solution, was proposed [19], [20]. The model was extended in 3D in 2008, but required constant speed during climbing phases [21]. Uncertainties on the trajectory heading were added to the horizontal model in 2009, but all maneuvers still needed to be executed at the same time (at every optimization step) [22].…”

Dynamic collision avoidance using local cooperative airplanes decisions

“…The size of the configuration space has led most research to discretize the configuration space, either for the maneuvers or the airspace, and explore it with graph search or evolutionary algorithms, or by using heuristics to guide the exploration. Meta-heuristics using discretization of maneuvers, such as genetic algorithms [10] or ant colony algorithms [11], along with artificial intelligence algorithm as neuronal networks [12], give interesting results but they scale poorly. Indeed, [10,11] are one of the few that handle more than 20 airplanes and still find a global optimum.…”

Constant speed optimal reciprocal collision avoidance

“…In the early 2000s, a mathematical model using Mixed Integer Linear Programming, which could be solved by CPLEX and ensured the global optimality of the solution, was proposed Pallottino et al (2001Pallottino et al ( , 2002. The model was extended in 3D in 2008, but required constant speed during climbing phases Christodoulou and Kontogeorgou (2008). Uncertainties on the trajectory heading were added to the horizontal model in 2009, but all maneuvers still needed to be executed at the same time (at every optimization step) Gariel and Feron (2009).…”