Facebook’s solar-powered drone crashed on its very first flight. Google has announced it is closing down its solar-powered drone programme, though continuing to test its Project Loon balloons. However, doubts remain as to the real feasibility of such ‘loony’ projects.

Worldwide Internet access via drones has still some way to go

The idea looks very appealing: providing Internet access for all through drones. However, it appears that this goal is going to be really difficult to attain. What are known as High Altitude Pseudo-Satellites (HAPS) are proving to be a flop, or at least that is the conclusion drawn by observers following the failure of Facebook’s first solar-powered drone flight and Google’s decision to shut down its telecommunications drone project.

Daniel Hernandez, former assistant director in charge of future planning at CNES, the French National Centre for Space Studies, who is nowadays working as an independent consultant, was not impressed by these initial experiments. He points out: “Carrying out a test, a demonstration, is not particularly problematic. But as soon as these pseudo-satellites have to cover quite an extensive area – particularly for a long period –enormous technical problems arise, which have to be resolved, in addition to the political problems. Apart from potential complaints from residents, the operator will have to obtain authorisation to fly from each country concerned – unlike satellites, which have the right to do so.” However, salvation is potentially at hand from other ‘pseudo-satellites’ such as balloons and airships flying in the stratosphere.

The technical limitations of drones

The failure of the first flight of Facebook’s Aquila drone, developed by the firm’s research team, demonstrates the difficulties that the engineers are coming up against. Solar-powered unmanned aerial vehicles are very light but very large. Aquila, relying entirely on solar power, has just 5kW of power to fly on, although its wingspan approaches that of a Boeing 737. Coming in to land after a maiden flight lasting 90 minutes, a sudden gust of wind proved enough to unbalance the fragile machine, which speeded up and broke apart. Facebook has not announced any intention to give up on Aquila, but meanwhile rival Google has drawn conclusions from its own tests and thrown in the towel, even though its project was able to call on the expertise of Titan Aerospace, the firm Google bought in 2014. So the Mountain View giant has shut down its solar-powered drone programme, reassigning members of the Titan team to other projects.

With its Zephyr range, Airbus Defence and Space continues to focus on solar drones designed to fly in the stratosphere. To date, only the British Army has chosen Zephyr for surveillance applications.

At the moment, the most significant pseudo-satellite drone project is the Airbus Zephyr. The Zephyr 7 has demonstrated its capacity to fly long distances, having flown for an uninterrupted 340 hours, i.e. fourteen consecutive days. These are however modest figures when it comes to telecommunications services. Today only the UK Ministry of Defence has bought the UK-built Zephyr S, and not for use as a telecoms relay, but for long-endurance drone surveillance.

Google still betting on balloon constellations

Although Google has given up the idea of solar-powered drones, the company is – in spite of many obstacles – pursuing its ‘Project Loon’, developing high-altitude balloons to be placed in the stratosphere. Google X (now simply ‘X’) research and development facility staff are continuing testing, the goal being to implement a launch system capable of launching a new balloon every 30 minutes so that they can fly on a permanent basis above a given geographical zone so as to provide Internet connectivity. Engineers calculate that at an altitude of 20,000m – well above the commercial aircraft flight paths – a balloon could cover an area of 5000 km2, about the size of the US State of Delaware. The engineers are looking to pilot the balloons in the same way as hot air balloon pilots do, controlling the balloon’s position by changing altitude so as to seek out the winds blowing in the right direction. Following an initial pilot test in New Zealand in 2013, trials are now continuing in different areas of the globe. Google has already taken the record for continuous flying time to 187 days.

Having closed down its Titan Aerospace solar-powered drone programme, Google is now focusing on balloons designed to fly in the stratosphere, using technology that is already well known for its meteorological applications.

Google is currently launching balloons in South America, in Uruguay among other places. Photos posted on the social networks by people on the spot show that the engineers are still far from having an operational system up and running. The balloons’ trajectories are proving erratic and wreckage from pseudo-satellites is regularly found by walkers in the countryside, evidence that the risk of balloons falling out of the sky is still very real. "Launching a large number of these pseudo-satellites into the sky means you have to take into account the statistical probability of an accident somewhere,” argues Daniel Hernandez, underlining:”If the balloon load falls on someone, it could kill them, and even the actual envelope, though it’s very light, could potentially cause an accident if it fell on a vehicle.” So the twin challenge is firstly to manufacture and launch these balloons at scale, and secondly to pilot and bring them back down safely.

Airships set for a big comeback?

Given the difficulties facing drones and balloons, a third alternative, the airship, seems to be making a comeback. “The idea is nothing new. And we don’t have to go back to the tethered balloons of the First World War. The idea of deploying balloons as telecoms relays was proposed in 1997 by [American lawyer, author, and entrepreneur] Martine Rothblatt, with her Sky Station project,” Hernandez points out. But while this is not an entirely new idea, the technology has certainly evolved over the last twenty years and there is now a pressing need for universal global Internet access. At Thales Alenia Space, an aerospace company based in the southern French town of Cannes which is currently working on the ‘Stratobus’ airship, they are confident that the idea which failed to find favour back in 1997 will now have greater appeal for telecoms operators;. "We’re in touch with satellite operators, almost all of whom are showing great interest in the stratosphere, and also with terrestrial telecoms operators,” revealed Jean-Philippe Chessel, Stratobus Product Line Manager at Thales Alenia Space.



Explains Chessel: “Unlike Google Loon, whose untethered balloons are whisked off by the wind, the Stratobus is a motorised airship which uses its engines to remain stable even in winds of up to 90 kilometres per hour. It therefore remains stationary above a given position.” A Stratobus can remain in place for a whole year in any zone situated between the tropics, where the stratospheric winds reach a maximum of 90 km/h. At temperate latitudes, where the winds can be more violent, the Stratobus can currently only provide a seasonal service or cover ad hoc events. Based on Thales Alenia Space engineers’ calculations, a Stratobus could cover an area within a 200km radius. In addition, engineers are already thinking about hybrid constellations, combining traditional satellites and balloons. The balloons could for example communicate with the satellites using radio frequencies or laser beams.

In April 2016, a project along these lines was granted financial support under the PIA – the ‘Programme of Investments for the Future’ operated by French national investment bank Bpifrance. The two-yearresearch programme consists of trialling an initial prototype on a small scale, i.e. a balloon-like craft 40 metres in length and 12 across. It will be tethered and will be stationed only a few hundred meters up in the sky. While pseudo-satellites seem to be currently struggling, the advent of 5G could nevertheless revive interest in HAPS among both manufacturers and operators. The fact that HAPS can potentially offer a latency time of just a few milliseconds for data exchange should certainly appeal to 5G operators going forward.

By Alain Clapaud
Independent journalist specialising in the new technologies