New applications enabled or enhanced by 5G for drones in public safety, security, precision agriculture and mobile communications are being evaluated in various trials and testbeds around the world. The main points are that by including a specification for latency for the first time in a cellular standard and also enabling smooth handover between cells for fast moving clients, 5G will greatly extend the scope of drones along various dimensions.
Drones strictly include all unattended vehicles including boats, self-driving cars and even missiles, but here we discuss just UAVs (unattended aerial vehicles) such as the familiar consumer quadcopters, because these have particular constraints which 5G can help overcome either directly or indirectly.
The primary challenges then for UAVs are power and control, which are slightly related since the ability to navigate totally autonomously requires onboard computation which consumes electricity. The development of lithium ion batteries provided some relief just as it did for autonomous vehicles on the ground, but given the greater weight constraints of flying, coupled with effective inability to take on electric charge without landing, that remains a serious limitation.
For this reason, most drones have been, until recently at least, primarily remote controlled with limited scope for autonomous navigation driven by onboard sensors. The role of sensory devices has been to date largely for reconnaissance and remote control, with cameras the first sensory devices to be incorporated. Indeed, semiconductor advances over two decades or more have brough cameras of increasing quality and reducing weight. That has also enables accelerometers to come on board, while reducing the weight of gyroscopes, which are essential for stable flight.
Most recently, other imaging systems have become feasible, such as Lidar for precise measurement of distance from surrounding objects without need for external light so that it can operate at night. Previously confined to larger systems such as cars or aircraft, lidar, which uses pulses of infrared or ultraviolet light like radar to measure distances from the sensor, is now available in very small form for drones. The rapid progress in miniaturization is witnessed by Apple’s inclusion of lidar in its latest iPhone 12 Pro.
This is a huge development for drones because it gives them the capability to navigate much more precisely in the absence of remote control. This is where 5G can score by enabling some of the associated processing to be offloaded to the cloud at sufficiently low latency. This will greatly expand the scope of drones, according to Sarah Yost, senior business development manager at NI (formerly National Instruments), a US maker of automated test equipment and software involved in a 5G testbed evaluating potential applications.
This is part of the Platforms for Advanced Wireless Research (PAWR) program spearheaded by the US National Science Foundation (NSF) to bolster the nation’s 5G effort. Other partners include leading technology and semiconductor players such as Samsung, Qualcomm, Intel and Facebook, as well as the nation’s three leading MNOs AT&T, Verizon and T-Mobile, which collectively have invested $100m in PAWR.
As Yost put it, NI is one of the PAWR team assessing how best to integrate drones into 5G networks with benefits for both. Applications being looked at include public safety in getting closer to dangerous events or situations than humans could without risk, for more detailed assessment or even remedial actions. Drones could descend towards forest fires to help determine where the flames will go next, or play roles in de-escalation, as in disarming a bomb.
The latter requires robotic capability and companies such as SteelRock Technologies and Chemring Technology Solutions are already working on systems designed to neutralize improvised explosive devices (IEDs) in environments that are dangerous or hard to access for human disposal experts.
Another category being evaluated is the broad one of air protection, which can be a euphemism for military applications and also takes drones into arms races with criminals or other countries, especially over jamming.
On the one hand, jamming can be employed by airports for example to disable rogue drones entering their air space, as has already happened with disruptive effect in on or two instances. On the other hand, there is work ongoing with systems such JAM-ME from Hamad bin Khalifa University in Qatar to develop autonomous jamming-assisted navigation that allows a drone to accomplish a mission even in areas where anti-drone jamming protection is employed. Clearly availability of such technology might negate anti-jamming protection employed by airports.
The PAWR program, like some others, is also looking at the use of drones to extend the 5G network itself by operating as floating cells at high altitude for coverage in remote areas. This field of high altitude cells has a checkered history involving balloons and solar powered drones, as well as this latest larger hydrogen machine.
Google’s Project Loon to deliver internet services from twin airborne balloons, one filled with air and the other helium, has been in prolonged gestation since its launch in 2015. Facebook’s rival project Aquila, using solar-powered drones, was abandoned in June 2018 as being unviable and not cost effective.
The other option is satellite, but this struggles to meet latency targets, as well as having concerns over ability to handle large numbers of simultaneous unicast connections cost effectively. Nevertheless, there is interest in use of LEO (low earth orbit) satellites to fill in 5G coverage in remote land areas and over the ocean. SpaceX, with its Starlink, plans to target the 3% “hardest to reach customers” in rural areas beyond terrestrial 5G, according to its owner Elon Musk, aiming for latencies below 20 milliseconds which would be good enough for all but the most demanding use cases.
China is offering competition with a seemingly even more ambitious plan to provide global 5G connectivity from a constellation of 144 LEO satellites by 2022 made by Beijing based start-up Galaxy Space.
Given this satellite competition and the backdrop of failed projects so far, there is some scepticism over whether drone-based approaches will be economically viable, despite claims that it will work out cheaper per cell than terrestrial 5G when dense millimeter w
Wave deployments are taken into account. However, the concept is starting to obtain buy-in from leading operators, with Deutsche Telekom a notable enthusiast and investor. Telekom in 2019 acquired 31% of UK firm called Stratospheric Platforms Ltd (SPL), later increased to 38%, which in partnership with Cambridge Consultants has claimed a breakthrough in antenna technology with a three-meter-wide design capable of beamforming from altitude to reuse spectrum and deliver airborne capacity from a single platform equivalent to up to 1,000 or more ground-based cells.
The two firms reckon that could cover the whole UK with 60 drones flying at about 14 kilometers altitude just under the stratosphere, depending on how much back-up is provisioned. Deutsche Telekom and SPL have now demonstrated 4G voice and data connectivity over the platform integrated into the operator’s commercial mobile network in Bavaria with a remote-controlled aircraft system working at 2.1 GHz.
SPL developed the aircraft, antenna and hydrogen fuel cell to keep it aloft. The test flights delivered 4G connections at 70 Mbps downstream and 20 Mbps up, which would need to be tripled for 5G. However, unlike satellite and balloon alternatives, the 3GPP compatibility allows access from current handsets without installation of new terrestrial equipment.
Despite Telekom’s enthusiasm, such drone-based systems will also find it hard to be profitable and will certainly have to make significant headway into sectors such as shipping and aviation beyond the niche of remote land areas. They will also struggle to get close to the high bit-rates and ultra-low latencies of mmWave on the ground, given those frequencies will not be available from the air. It is unlikely that these drone-based systems will displace 5G infrastructure on the ground, with the best hope being to complement that around the edges.