Forecasting, visibility and 3D route optimization for future air traffic management

• Main Author: Xie, Zhen
• Other Authors: Zhong Zhaowei
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2018
• Subjects: DRNTU::Engineering::Aeronautical engineering::Flight simulation
• Online Access: https://hdl.handle.net/10356/88514; http://hdl.handle.net/10220/45896
• Institution: Nanyang Technological University

Based on time-series Recurrent Neural Network (RNN) forecasting results, air traffic volumes and air passenger volumes will keep increasing for the coming decades. It will challenge the current airport capacity including runway capacity and terminal capacity. The airport group can improve the airport service efficiency and resource usage rate or build more facilities to meet the passenger demand. Moreover, air traffic volume forecasting demands better route structure as the number of aircraft goes up rapidly and current rigid route structure cannot meet the future traffic demand. Free flight concept emerged as a result of the development of advanced tracking, prediction and communication equipment. Artificial-Intelligence (AI)-based free flight increases the safety operation of the flight especially in adverse weather conditions and congested airspace as free route leads to more direct or wind-optimal trajectory. In the context of next generation of air traffic management (ATM) and implementation of automatic dependent surveillance – broadcast (ADS-B), the author proposed methods that can help to build safer and more efficient ATM systems. That is to say, the author applies a novel 3D multi-agent path planning method based on the free route airspace (FRA) concept in the ASEAN region and uses visual conspicuity experiments to guide the visual factors in order to obtain maximum visibility of the aircraft for tower controllers. On one hand, in order to better cater the increasing traffic volume, the author invented and validated a novel 3D multi-agent path-planning algorithm for the commercial aircraft to find a more efficient route, especially under adverse weather conditions. By implementing this path-planning algorithm in free-flight route structure, airspace capacity can be increased significantly. Since pilots can plan the aircraft freely between entry points and exit points, there is a decrease in workload for Air Traffic Control Officers (ATCOs). Moreover, the advantages of changing to this route structure include lower flight traveling time and decreased energy consumption. This route structure could contribute significantly to curb climate change impact due to the fact that greenhouse gases are mostly emitted from vehicles and aircraft. On the other hand, visual conspicuity for airplane approaching and departing in the vicinity of the airport or in the airspace is very important for the safety of aircraft. Studies on visual factors can improve visibility for the surrounding environment. Visual conspicuity field tests tend to cost a lot and require huge amount of human resources. Moreover, it is restricted by the schedules and regulations of the airport. Thus, lab simulation can replace field tests and generate the desired scenes easily. We use commercial image processing software to create the desired scenarios with the changing of the visual parameters based on the Airbus A320 and Boeing B737-800 SolidWorks models. The results show that visual conspicuity is related to visual factors including lightness, distance, chromatics, contrast, reflection and search time. The visibility of aircraft can be increased by tuning visual parameters.