WSpeed: Drone Energy Optimization for Multiple-Package Delivery Considering Weight Changes

Drones are increasingly used in daily or urgent package delivery with their outstanding flexibility and rapid response capability, especially in scenarios like post-disaster relief, medical supply dispatch, etc. However, due to the limitations of drone battery size and load capacity, the delivery coverage scope and the number of packages can be delivered per time are highly constrained. In addition, the delivery order of packages also affects the quality of service due to the different urgency levels of packages. Since the flight energy dominates the total UAV energy consumption, it would be very beneficial to find energy-saving methods for the package delivery scenarios, through more energy-efficient drone speed and trajectory scheduling. In this paper, we propose WSpeed, an energy optimization algorithm that jointly considers the drone speed and flight path optimization with urgency differences of delivery points, for the multiple-package delivery scenarios. WSpeed is based on a novel drone energy consumption model, which provides a key finding, that as the total drone weight changes, the drone energy consumption per unit distance, as well as the most energy-efficient speed also changes. Therefore, during a multi-package delivery flight, the UAV speed should be adjusted adaptively to achieve the best energy efficiency. Moreover, WSpeed clusters delivery points based on the load capacity and the distance between points, for more efficient path scheduling. Besides, WSpeed also considers the different package urgency priorities during the delivery, which is closer to the real-life scenarios. Experiments show that WSpeed can save as much as 27.2% energy and 36.5% cumulative flight time over existed approaches while achieving the minimum service utility.