How to design the future of air traffic
I recently discovered my love for reporting on research projects sponsored by public funds (your taxes at work).
I will be reviewing research projects from around the world, both recently-completed and ongoing, and preparing a condensed summary (less than 5-minute read) in the hopes to bridge the gap between the-thinkers and the-doers (read: entrepreneurs) quicker than traditional channels. I hope that my thoughts and questions can spur discussions and improvements, collaborations and investments.
My core areas of interest are transportation, bioeconomy and nanotechnology. And everything else.
The Research Chronicles: #001 — METROPOLIS
The METROPOLIS project that ran between October 2013 and April 2015 for a total cost of €760,197 (co-ordinated by the Technical University of Delftin in the Netherlands). Their forward-thinking project aimed to keep aircrafts from crashing into each other once the airspace became crowded.
In the US alone, the FAA expects the number of hobbyist drones to triple in size from 1.1 million UAVs to 3.55 million by 2021. (UAV = Unmanned Aerial Vehicle). And in fact, there are no technological hurdles left to owning a PAV (Personal Air Vehicle)
In fact, the biggest hurdle for society is the need for efficient and safe Air Traffic Management.
Imagine a not-so-distant Jetsons future of more than 1 billion UAVs and PAVs in the cities above us (surveillance drones, delivery drones, personal transport aircrafts, news/media coverage drones). What would be the optimal way to manage that traffic while prioritizing safety? Choose from one of the images below.

If you selected (a) “Flock of birds”, go ahead and do the happy dance.

The Flock of Birds concept, referred to in the research project as a “Layered” concept provided an optimal environment to improve safety and efficiency in a spatially spread demand.

The improvements in safety come at a cost of reduced efficiency. Shorter flights at lower altitudes (higher fuel burn) while longer flights at higher altitudes.
The simulation city was a fictional “Metropolis” inspired by the 1927 expressionist film and structured similar to present-day Paris (although I think most cities today have a similar structure)

By the way, if you guessed Swarm of Bees, you were a close second.
Referred to in the research as a “Full Mix” concept, it finds the most direct route from origin to destination, allowing for direct horizontal routing & optimum cruise altitudes. I’m guessing safety goes down?

Interestingly, the most structured approach (Tubes) took double the time increasing noise pollution and third-party risk. Concentrating traffic flows along pre-determined paths leads to congestions — and this is actually the air traffic management concept in place around the world today. (Now perhaps you understand why you almost missed your connecting flight that time you flew to Lisbon)
My questions (feel free to respond in the comments or to me):
(Update Aug 28 2017 — I received responses to these questions directly from Jacco M. Hoekstra, the METROPOLIS project and Full professor CNS/ATM Communication, Navigation, Surveillance/Air Traffic Management at the Delft University of Technology so I have updated those answers below)
(i) The software used to simulate these tests was the Traffic Manager (TX) developed by the National Aerospace Laboratory of the Netherlands (NLR) capable of simulating up to 5000 aircrafts simultaneously. Is there an open-source version of a similar software that interested startups and researchers can use to improve this model?
JMH: Yes, there is an open source version of this tools, which has been developed by us at TU Delft and is used in projects where we do not have the TMX. It is called BlueSky and can be found here: https://github.com/ProfHoekstra/bluesky/archive/master.zip
Info on installation and use can be found in the Wiki: https://github.com/ProfHoekstra/bluesky/wiki
This tool has been used in follow-up research after the closure of the project.
(ii) If all the flight logs in the history of flights were made available to the public (as a data set for research) would that help or would it just encourage us to designing for the past instead of designing for the future?
JMH: If you are referring to current flights, we do collect those (almost) globally via a co-operation with flightradar24. But regarding UAVs or PAVs, of course more data results in more insight, so yes.
(iii) Would the results be different if the structure of the city was different? Are urban planners currently taking into consideration air traffic management (over city airspace) in their plans?
JMH: Yes and no. The basic principles derived are universally true. In follow-up research we have derived the relation mathematically, which we have tested using the experimental data from the 9.5 million flights in Metropolis. So some universal conclusions are:
1. In unstructured demand for point to point traffic, the track-based layered airspace is the one which yields the highest capacity.
2. Other methods which lower the relative velocity will also work. We are investigating several of them right now. And only in some special cases (departure/arrival, congested corridors) they might even beat the layers.
3. 4D TBO-like concepts (tubes) is limited by the difficulty of the prediction of conflicts, which is especially sensitive as it comes to conflicts compared to trajectory prediction.
4. Decentralised approaches beat centralised approaches in high traffic densities.
5. Using distributions of track angles and altitudes an estimate can be made of the conflict probability (this is our follow-up research.
Almost forgot. Whilst care has been taken to ensure that this information is accurate, it may be out of date or incomplete. Neither the METROPOLIS Consortium participants nor the European Community are liable for any use that may be made of the information contained herein (or above)
