I’m not a huge fan of the UAV delivery-to-your door concept. Aside from a personal dislike of small helicopters buzzing around me, I hate to see retail and delivery jobs eliminated. The people I know who do this work really need these jobs as a step up (or a landing to avoid going down).
On the other hand, I would like to see public transportation survive and thrive. A city with no public transit is a rotten place to grow up, and a difficult place to get started or be poor–or old,
The concept is simple.
Contemporary UAV delivery copters have a pretty limited range, especially when carrying meaningful cargo loads. This means that covering a city needs a lot of UAVs and a lot of recharging bases, spread across the whole area. Some operators have been experimenting with mobile vehicles, essentially delivery trucks that can dispatch UAVs. Such a solution would use a fleet of surface vehicles to extend the range of a smaller number of bases and UAVs.
The Stanford researchers explored the potential effectiveness of using existing mass transit systems instead of dedicated mobile bases. There are already large fleets of vehicles covering the city, so why not piggyback on them? And the roof of a city bus is pretty large and not used for very much. So, yeah, that could work.
The study focusses on the question of “how would you route deliveries if you could do it”? I.e., assuming that we can ride on the roof, how well would that work .
This is actually a moderately complicated optimization problem, because there are a lot of variables to consider. But, hey, optimization problems are what academic computer scientists are here to tackle!
The paper describes an efficient framework that can compute a schedule quickly (in seconds!), which means that you could try to keep up with the flow of a real city . I.e., you can recompute a new solution as things change.
The hitchhiking more than triples the effective range of the UAVs, which save power by riding the bus part of the way to their destination.
Obviously, there is some work to be done to get UAVs to autonomously land and take off from the roof of a bus. You probably need to know the motion of the bus (which might or might not be moving), and take care about obstacles (underpasses, overhead wires, who knows?) A city bus is a creature of the urban jungle, for sure.
I assume that we might have a charger station on the bus, too. And maybe more than one dronepad per vehicle, which would add air traffic control to the requirements.
I’m not sure how the rendezvous would happen, but at least some of the time the UAV might have to wait for the bus, just like the passengers. So, perhaps the bus stops would have dronepads with chargers, where the UAVs can safely nest.
Aside from extending the range of the aircraft, this concept has other potential advantages. Riding on a bus is probably a relatively safe and secure location, which offers a potential haven in case of emergency, bad weather, or malfunction. Worst case, the UAV can power down and ride to the end of the line for manual recovery.
But the best thing is that the UAVs would pay fares (maybe even refueling fees), sustaining the public transportation network with paying freight. This would also push the transit system to cover the whole area, in order to garner more freight traffic, and in the process serving more passengers. (And if the UAVs nest on bus shelters, there would be a demand to install and maintain shelters throughout the whole area, too.)
So, there are lots of wins, including plusses for the mass transit system and the public who rely on it.
And interesting idea.
- Evan Ackerman, Delivery Drones Could Hitchhike on Public Transit to Massively Expand Their Range, in IEEE Spectrum – Robotics, June 11, 2020. https://spectrum.ieee.org/automaton/robotics/drones/delivery-drones-could-hitchhike-on-public-transit-to-massively-expand-their-range
- Shushman Choudhury, Kiril Solovey, Mykel J. Kochenderfer, and Marco Pavone, Efficient Large-Scale Multi-Drone Delivery Using Transit Networks. arXive, 2020. https://arxiv.org/abs/1909.11840