In our commercial UAV report, Valour Consultancy laid out a timeline for future Drone Industry Development, both drivers and inhibitors. One of the first items it listed as a driver for development was Individual Drone Identification:
“Commercial companies will welcome and whole-heartedly endorse drone registration and individual drone identification (they will do this for purely selfish reasons). Probability 95% + 5% – 10% Timeline 2016 Q4 to 2017 Q4”
Last week, DJI produced a well-reasoned and thoughtful whitepaper entitled ‘A Call for a Balanced Remote Identification Approach’. http://www.dji.com/newsroom/news/dji-proposes-electronic-identification-framework-for-small-drones
In this paper, they argue that the privacy of the operator should be respected. To achieve this and respect the interests of privacy of third parties and of other airspace users and to comply with existing legislation and community principles, they suggest a Non-Network Localised ID. It appears that they suggest something similar to an IP address or other identification code being broadcast on the 2.4 GHz or 5.8 GHz bands. In itself, this is an ideal solution for drone manufacturing companies.
In a controlled airspace, a conflict of airspace is said to exist when aircraft are separated by less than 9km horizontally and 1km vertically. In uncontrolled areas, depending on the visibility, 300 metres horizontally and some vertical separation is all that is required. There are some issues that an identification code broadcast will not address however. In urban and suburban conurbations, this would seem to be ideal but it is in these areas that drones are likely to come into the operating space of other air users such as air ambulances, police helicopters, press helicopters and senior executives returning from late lunches at the golf course by company helicopter.
Wi-Fi broadcasts are pretty much line-of-sight and of limited distance. At 300m in clear field communication, the quality of the communication has been measured at 56%. This is reasonable as the communication (broadcast of ID) will be frequent and the content simple and so easily checked.
If we wanted to alert any other air vehicle to the presence of our drone at, say, 600m or around buildings (any buildings at 600m will have lots of glass so the likelihood is that signals will pass through), and in inclement weather, then we need a slightly different system. Experiments of radio transmission through concrete suggest that a thickness of 40cm (or 2 x 20cm) is enough to degrade transmission so that it becomes unreadable. Weather is also an issue as heavy moisture content in the air degrades a signal. More research is needed before a suitable broadcast can be finalised.
And then there is the issue of power. Broadcasting ID over 300m (say) at 10s intervals (say) in addition to normal command and control and even video streaming, can up power consumption for the control portion of power expenditure considerably. This shortens flight time, not significantly and not enough to disturb anyone using a drone to take photos or just enjoying the pleasure of flying one. However, if you are using the drone as a tool and you now require one or two extra sorties in an 8 hour work day to achieve the same surveying coverage, this increases cost for spare/stand-by batteries and man-power costs. Any holistic appraisal of drone productivity needs to be cognisant of these factors.
When we say the flight time is less or rather the power consumption increases but not significantly, we are talking around 0.5% increase (using a very rough rule-of-thumb calculation). That may not seem significant but remember that it is only the insignificant curvature of the Earth by 0.00025% that stops the Earth from being flat.