The current air transportation system (ATS) is being pushed to operate ever closer to its critical capacity, leading to increases in both the frequency and duration of delays. This strain is expected to worsen in the coming years alongside an increase in the demand for air travel. As it stands, human air traffic controllers direct the bulk of traffic to travel along a limited number of predetermined routes. To improve efficiency and increase capacity, there has been a push for the next ATS to allow aircraft greater autonomy when planning routes and during flight.
To support this vision, Kevin Spieser, Dimos Dimarogonas and Professor Emilio Frazzoli, in collaboration with a team from Georgia Tech., are working towards the development of a health monitoring tool for averting severe degradations in the health of a next-gen ATS. The team at MIT is concerned with quantifying the fundamental limitations governing traffic health. To this end, we model the ATS as a multi-agent system and use tools from operations research, queuing theory, and algorithmic robotics to relate degradations in the environment to key performance metrics, including delay and system throughput. In this way, we can better understand the effect of high traffic densities, inclement weather, and navigational uncertainty on system performance. This work will serve to better quantify the safety of the national airspace, develop new policies for managing air traffic, and establish benchmarks in the achievable performance of large scale, autonomous, mobile, multi-agent systems.