Case-Based Reasoning for Flight Conflicts Resolution in En-Route Short Term Conflict Alert Situation

Abstract

This paper introduces a knowledge-driven approach for solving short range conflicts in En-route STCA situa- tion in order to assist the air traffic controller to handle emergency and ensure the safety. In stead of applying classic con- flict resolution approach, case-based reasoning theory and technology is used to transfer the controllers' knowledge and experience to automate conflict resolution. Firstly on the basis of analysis of the controllers' conflict resolution knowledge with domain experts, a description of case is well established in a structural framework which has two layers, the first lay- er with 4 Slot features and the second layer with 19 Side features. Secondly, in case retrieval, the source cases are firstly filtered against constraint parameters then a nearest neighbour algorithm and global similarity measure based approach is applied to calculate the degree of similarity between the target case and the source cases. Finally, two scenario tests are designed to verify the performance of CBR-based conflict resolution model. The results indicate that this method is con- sistent with the solutions recommended by air traffic control experts and has a good effect in application.