Swarm-Based Drone-as-a-Service for Delivery

Abstract

There has been a growing interest in the applications of drones as a cost-effective, efficient, and environmentally friendly alternative in various domains. Particularly in the context of delivery services, the demand for contactless and efficient delivery solutions has surged. Drone delivery offers faster and greener deliveries. However, existing methods focus primarily on point-to-point delivery, limiting their potential for optimisation. This thesis proposes a novel approach to servitise drone delivery by operating through a skyway network composed of building rooftops, enabling drones to traverse between source and destination while recharging at intermediate nodes. Although single drone delivery offers numerous advantages, it faces significant challenges in scenarios where multiple packages require simultaneous delivery. Flight regulations, which often limit the carrying capacity of individual drones, necessitate the exploration of alternative solutions. Therefore, this thesis presents a novel Swarm-Based Drone-as-a-Service (SDaaS) model and framework for multiple package delivery. The proposed framework prioritises the composition of services that optimise Quality of Service (QoS) factors, such as delivery time and energy consumption. This thesis identifies swarm-specific constraints and leverages the unique characteristics of drone swarms. It explores swarm formations, in-flight wireless charging between drones, and allocation problems to maximise drone utilisation for consumer deliveries. Furthermore, this research investigates the recommendation of services to consumers based on their preferences, aiming to increase their satisfaction. Moreover, the framework addresses the resilience of SDaaS by addressing issues related to drone soft failures and their impact on other swarm members. Ultimately, this work paves the way for the widespread adoption and optimisation of swarm-based drone services in the context of last-mile delivery.