Ninth USA/Europe Air Traffic Management Research and Development Seminar (ATM2011) Relative Significance of Trajectory Prediction Errors on an Automated Separation Assurance Algorithm

Abstract — Trajectory prediction is fundamental to automated separation assurance. Every missed alert, false alert and loss of separation can be traced to one or more errors in trajec-tory prediction. These errors are a product of many different sources including wind prediction errors, inferred pilot intent er-rors, surveillance errors, navigation errors and aircraft weight es-timation errors. This study analyzes the impact of six different types of errors on the performance of an automated separation assurance system composed of a geometric conflict detection algo-rithm and the Advanced Airspace Concept Autoresolver resolu-tion algorithm. Results show that, of the error sources considered in this study, top-of-descent errors were the leading contributor to missed alerts and failed resolution maneuvers. Descent-speed er-rors were another significant contributor, as were cruise-speed er-rors in certain situations. The results further suggest that increas-ing horizontal detection and resolution standards are not effective strategies for mitigating these types of error sources