Efficient Large-Scale Multi-Drone Delivery using Transit Networks

Abstract

We consider the problem of routing a large fleet of drones to deliver packages simultaneously across broad urban areas. Besides flying directly, drones can use public transit vehicles such as buses and trams as temporary modes of transportation to conserve energy. Adding this capability to our formulation augments effective drone travel range and the space of possible deliveries but also increases problem input size due to the large transit networks. We present a comprehensive algorithmic framework that strives to minimize the maximum time to complete any delivery and addresses the multifaceted computational challenges of our problem through a two-layer approach. First, the upper layer assigns drones to package delivery sequences with an approximately optimal polynomial time allocation algorithm. Then, the lower layer executes the allocation by periodically routing the fleet over the transit network, using efficient, bounded suboptimal multi-agent pathfinding techniques tailored to our setting. We demonstrate the efficiency of our approach on simulations with up to 200 drones, 5000 packages, and transit networks with up to 8000 stops in San Francisco and the Washington DC Metropolitan Area. Our framework computes solutions for most settings within a few seconds on commodity hardware and enables drones to extend their effective range by a factor of nearly four using transit.

On the Throughput of the Common Target Area for Robotic Swarm Strategies

“…In addition, the first two arrival times were not plotted because the first one is zero, yielding an indeterminate output by the throughput definition, and the second one is still too small in relation to the others, making the resultant throughput too high compared with the rest, thus producing an incomprehensible graph. Observe that the values from (16) are almost equal to the values from simulation, except for some points. They are different because of the floating-point error in the divisions and trigonometric functions before the use of floor function used on (16)-already mentioned in the introduction of Section 4-as well as the time measurement errors for the arrival of the robots on the target area as explained at the beginning of this section.…”

“…Even recent related works on multi-agent systems [13][14][15] also deal with this problem in a similar way. In [16,17], they also deal with multi-agents and pathfinding, but not in a situation where the target of every agent is the same area. Furthermore, distributed solutions are considered here where agents only have local information, while [16,17] propose centralised solutions.…”

“…In [16,17], they also deal with multi-agents and pathfinding, but not in a situation where the target of every agent is the same area. Furthermore, distributed solutions are considered here where agents only have local information, while [16,17] propose centralised solutions. Xia et al [18] investigate the topology of the neighbourhood relations between multiple unmanned surface vehicles in a swarm.…”

On the throughput of the common target area for robotic swarm strategies

“…Related works on multi-agents systems (Carlino et al, 2013;Sharon and Stone, 2017;Cui et al, 2021) deals with a similar problem, but they consider autonomous cars navigating over lanes and roads, and coordination is needed at the junctions. In (Choudhury et al, 2021) and (Shahar et al, 2021), they also deal with multi-agents and pathfinding, but not in a situation in which the target of every agent is the same area. Shahar et al (2021) present theoretical analysis for their proposed matter, alongside experimentation by simulations corroborating the results as we do.…”

“…Shahar et al (2021) present theoretical analysis for their proposed matter, alongside experimentation by simulations corroborating the results as we do. Furthermore, we are considering agents with only local information and distributed solutions, while Choudhury et al (2021) and Shahar et al (2021) propose centralised solutions. Xia et al (2021) investigate the topology of the neighbourhood relations between multiple unmanned surface vehicles in a swarm.…”