Potential Functions for En-Route Air Traffic Management and Flight Planning

Abstract

This paper proposes an object oriented methodology and use of potential functions for efficient solutions to en-route air traffic management and flight planning in a complex airspace system. In this object oriented methodology, different objects, e.g., weather, Special Use Airspace (SUA), traffic, etc., and their attributes, e.g., weather orientation, SUA geometry, traffic density, etc., are defined by using potential functions. The interaction of the object of interest (host aircraft) and environment is modeled through an artificial potential field, whose interaction model is simple and well defined. This helps in framing a simple algorithm and produces efficient and complete solutions. The path planning and conflict detection and resolution algorithms developed for this methodology provide efficient solutions in the comprehensively defined environment model in which uncertainty is very limited. The algorithms have been implemented in a low-level flight planner. Results to validate the operation of the algorithm and feasibility of the methodology will be provided and a functional layout of the flight planner is provided. Introduction: The Distributed Air/Ground – Traffic Management (DAG-TM) concept [8] for Free Flight has been well defined for all phases of flight. The free flight in en-route airspace not only represents the largest portion of flight but also the most critical component to maintain continuous traffic flow in the National Airspace System (NAS). The DAG-TM Concept Elements (CE) 5, 6 and 7 dealing with en-route airspace has provided clear definition of the concept, requirements and solutions [3,5,6,7]. The problem in en-route air traffic has been believed to be issuance of inefficient resolution and difficulties in conformance to traffic flow management (TFM) constraints. The root cause for these problems has been identified as uncertainty in the trajectory information, lack of user preference and workload limitations [5,6,7]. The DAG-TM concept element 5, 6 and 7 proposes various efficient solutions by using decision support tools and collaboration. The decision support tools being developed for these problems should consider many factors that influence a problem and bring out efficient solutions that do not result in further complexity. One approach this paper suggests is an object oriented approach. Problems in a complex system like airspace can be best solved if an object oriented approach is applied. In this paper, potential functions are used which provide a convenient way to do this. Artificial potential fields have been used extensively in both robotics and air traffic [11,12,13,14,15,16]. Dynamic force fields have also been used, which are similar to potential functions and they carry more information than potential functions [10]. A methodology, named as potential functions methodology, will be introduced to use the potential functions to model the airspace in an object oriented approach. Potential functions of different shapes and strengths are used to represent various objects and their attributes in the airspace. The host-centric interaction between the host aircraft and the environment is modeled by using an appropriate potential charge, which exerts the force of attraction and repulsion. The 4D information of a potential function (3D shape and strength) is enough to formulate any problem which provides 4D trajectory information (3D location and time). Based on this methodology, two path planning algorithms are developed. One for flight route planning and the other for conflict detection and resolution and local trajectory generation. Many Conflict Detection and Resolution (CD&R) algorithms have been developed [3,4], but few have implemented coordinated resolution methods [1]. The CD&R algorithm developed under this methodology exhibits a rule based implicitly coordinated decision making for conflict resolution. The algorithms have been implemented in a low level flight planner, which features the basic functions of flight planning. This shows the