Intelligent agent for aircraft collision avoidance

Abstract

The trend of the air traffic system is toward a free flight environment. Free flight offers greater flexibility in planning for flights than the current air traffic control and management system. In free flight, operators are allowed to fly under instrument flight rules and choose their own flight path and speed in real time. One of the requirements to make the free flight environment feasible is an aircraft collision avoidance agent, also known as a traffic agent. One widely accepted concept of aircraft to aircraft communication for free flight environment is Automatic Dependent Surveillance Broadcast. In this research, the focus is on constructing a traffic agent, utilizing aircraft to aircraft information for flight management system. The agent includes a traffic conflict detection module and collision avoidance module. The method used by the modules is a combination of knowledge based expert system and optimal control. The expert system is the primary decision-maker and determines the appropriate actions required for conflict detection and avoidance. Optimal control is used to generate the optimum avoidance trajectory that adheres to the criteria assigned by the expert system. Results of various test cases presented in the research demonstrate that the combination of the two methods provides an efficient and effective way to obtain optimal traffic avoidance trajectories.