Methodology for Collision Risk Assessment of Aircraft with Diverse Collision Avoidance Capabilities
Date
2020
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This dissertation proposes a general dynamic event tree (DET) framework and evaluation methodology to assess collision risk for a variety of aircraft types and collision avoidance capabilities. The proposed DET framework consists of three levels – a high-level dynamic event tree that models multiple conflict detection and resolution (CD&R) systems that operate in a sequence to prevent a collision, a generic sub-tree modelling more specific sequences of events within each CD&R phase to resolve a conflict, and fault trees which model the component-based failure logic of each CD&R system. A solution approach is proposed combining analysis methodologies for dynamic event trees, phased-mission systems, and binary decision diagrams. The approach captures several different behaviors influencing collision risk such as time-varying conflict detection rates, pilot delays, component failures, and conflict geometry. The approach allows for ease of creating and modifying a model as well as quick evaluation. To illustrate the methodology, case studies are developed for collision risk between various types of aircraft with different collision avoidance capabilities in a hypothetical future airspace, e.g., Autonomous Flight Rules (AFR) and the Advanced Airspace Concept (AAC). In addition, sensitivity analyses on the model parameters including component failure probabilities, detection range of the sensors, and error rates of the CD&R systems are conducted. Case studies indicate that the reliability of aircraft transponders significantly drives collision risk since the CD&R systems and concepts considered highly rely on the transponders for surveillance. In addition, integrating unmanned aircraft with a limited CD&R system into the airspace would increase collision risk significantly.