For now, regulations and technical issues make widespread drone deliveries impossible, which means an army of flying machines probably will not fetch your holiday gifts this year or even the next. Here’s what experts note as the major challenges to resolve before delivery by drone becomes a reality.
Sense-and-avoid technology Picture a delivery drone flying your package kilometers from its distribution center all the way to your doorstep—technology needs to make every step of this trip safe and reliable. Most importantly, a drone must be able to detect manned aircraft, other drones and any other obstacle (like birds and power lines) to avert potential crashes.
Drones’ “sense and avoid” capacity likely will combine technologies such as cameras, radar and a surveillance system called Automatic Dependent Surveillance–Broadcast (ADS-B), which transmits data about an aircraft’s position and velocity to air traffic controllers and other aircraft. All this should keep drones from, say, colliding with birds or the plane that you’re on. ADS-B technology is now lightweight enough for small drones to carry, but its tracking abilities aren’t very useful until everyone else in the air relies on the system as well. And that might not happen until 2020, which is when the Federal Aviation Administration (FAA) will require most aircraft to be equipped with it.
Not only do delivery drones need to be able to avoid objects at high altitudes—they also have to navigate that last 100 meters to your home, which often presents a veritable maze of power lines, trees, cable dishes, buildings, people—you name it. Parts of this maze can change daily, which makes things even harder for drones—you may have temporarily erected a ladder or put swing set in your yard that were not there during the last delivery. Unmanned aircraft will need to problem-solve those situations on their own, like a Mars rover. “We’ve learned that one of the hardest parts of deploying drones,” says Todd Humphreys, an associate professor of aerospace engineering at The University of Texas at Austin, “is giving them a level of autonomy that would enable them to make decisions when things get complicated in the last tens of meters.”
Humphreys says one solution is that drones could build an instant 3-D map of your yard by merging data from multiple sensors, like cameras and LiDAR (for, light detecting and ranging. This would allow drones to assess their surroundings and independently decide whether anything is in their way. The strategy requires quite a bit of processing power but the technology may be ready for use soon. “I think it’s pretty close,” Humphreys says. “Even with a smartphone-grade processor, you can do this 3-D reconstruction in real time.” His lab is also working on lightweight and inexpensive GPS with subdecimeter accuracy that could help guide drones more precisely.
