Using drones (or unmanned aerial vehicles and flying robots) for delivering packages has been a hot topic lately. One might expect that our pizzas, Amazon orders, burritos, medical supplies, and Cherry Ripes will soon be ferried to our doorstep slung under a drone.

There’s a lot to like about the idea. Deliveries could not only be brought to our doorstep, they could be brought to wherever we are right now.

However, the reality is that there is much progress to be made in regulations, safety and technology before the idea of delivery by drones can go beyond trials and become routine.

Regulations in Australia

One of the first reasons delivery-by-drone is not imminent in Australia is the aviation safety regulations administered by the Civil Aviation Safety Authority (CASA). Under CASR (Civil Aviation Safety Regulations) Part 101, which covers unmanned aircraft, a drone may not fly:

  • Over a populous area.
  • Beyond line-of-sight of the operator.
  • More than 400 feet (about 120 meters) above the ground.
  • Within 30 meters of the general public.
  • In controlled airspace, or within 3 nautical miles (about 5.5 kms) of an aerodrome or heliport.
  • In or into cloud, at night, or in conditions other than VMC (Visual Meteorological Conditions).
  • For commercial purposes, unless the pilot holds a UAV Controller Certificate issued by CASA and the pilot is employed by a company that holds a UAV Operator Certificate.

For drone delivery to become real, these regulations would have to change, or exemptions would have to be granted; neither of which will happen quickly or lightly.

Safety

For the regulations to change, the components upon which a drone relies to keep in the air and navigate would need to become much more reliable to address the safety concerns. Motors, batteries, GPS receivers, telemetry, and compasses would need to become tolerant of failure, and immune to interference from radio frequency sources and strong magnetic fields the drone would encounter in urban or industrial environments.

A delivery drone would also need a collision avoidance system, so that in an unexpected encounter with an obstacle (such as a building, another aircraft, a tree or a person) the drone could take evasive action and resume its course.

As we all want our packages now, drone traffic management would be required to define orderly ‘air highways’ across cities so that delivery drones did not collide with each other or with full-size aircraft. Minimum separation rules would need to be applied to drones as they are to full-size aircraft.

Package destinations will present the full spectrum of challenges for an autonomous vehicle such as a delivery drone. Not all destinations will be a house on the standard quarter-acre suburban block with a concrete driveway. Some destinations will be in alleys, multi-tenanted apartments, businesses in industrial estates, and commercial buildings in the CBD. A delivery drone will need enough autonomy to navigate to the vicinity of the destination and select a descent path and landing spot that is free of obstructions and clear of people. It will also need to be clever enough to work out what to do when the situation changes, such as a toddler wandering underneath as the delivery drone descends.

The technology to address these challenges is available in various forms of maturity, but at a high price, not only in dollars, but importantly in weight. Not only must a delivery drone carry the package we ordered, but it must also carry all the electronics that help make it safe to do so. This means a bigger battery to generate the power to lift it all, which in turn means more weight, or limited range.

Weather

The final reason I do not believe delivery drones are imminent is an elemental one: the weather. All drones are affected by wind and rain.

There has been some great progress in making a waterproof drone. However, most drones are not waterproof as they have open motors and electronics. The efficiency of the propellers that produce the downward thrust required for flight may also be affected by rain.

A light drone will be blown about by wind. A drone with larger propellers designed for long endurance will usually be affected by wind even more so.

A weather station at the drone’s point of departure could monitor conditions and delay takeoff until the conditions are suitable. If the delivery path is long, it may be more difficult to ensure that good conditions will prevail for the outward and return journey. A network of drones reporting local weather conditions to each other has been suggested, but for safety, a delivery drone will probably need to have some degree of weather protection, which again means more weight and compromised flight duration.

Delivery by drone will happen, just not soon

I feel confident and hopeful all these obstacles will be overcome, because there is great benefit in solving them. Medicines can be delivered to patients in emergency situations. Blood samples could be delivered to a pathology laboratory for rapid analysis. Chocolate bars could be sped across the city from the supermarket to quell a craving. My sense is these drone delivery scenarios are years away, however there are many worthwhile needs that drones can serve in the meantime. I’ll be writing more about those shortly.