Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

QUADRO: A Hybrid Quantum Optimization Framework for Drone Delivery

James B. Holliday, Darren Blount, Hoang Quan Nguyen, Samee U. Khan, Khoa Luu

TL;DR

This work introduces Quantum Unmanned Aerial Delivery Routing Optimization (QUADRO), a novel hybrid framework addressing the Energy-Constrained Capacitated Unmanned Aerial Vehicle Routing Problem and the Unmanned Aerial Vehicle Scheduling Problem, formulating these challenges as Quadratic Unconstrained Binary Optimization problems.

Abstract

Quantum computing holds transformative potential for optimizing large-scale drone fleet operations, yet its near-term limitations necessitate hybrid approaches blending classical and quantum techniques. This work introduces Quantum Unmanned Aerial Delivery Routing Optimization (QUADRO), a novel hybrid framework addressing the Energy-Constrained Capacitated Unmanned Aerial Vehicle Routing Problem and the Unmanned Aerial Vehicle Scheduling Problem. By formulating these challenges as Quadratic Unconstrained Binary Optimization problems, QUADRO leverages the Quantum Approximate Optimization Algorithm for routing and scheduling, enhanced by classical heuristics and post-processing. We minimize total transit time in routing, considering payload and battery constraints, and optimize makespan scheduling across various drone fleets. Evaluated on adapted Augerat benchmarks (16-51 nodes), QUADRO competes against classical and prior hybrid methods, achieving scalable solutions with fewer than one hundred qubits. The proposed results underscore the viability of hybrid quantum-classical strategies for real-world drone logistics, paving the way for quantum-enhanced optimization in the Noisy Intermediate Scale Quantum era.

QUADRO: A Hybrid Quantum Optimization Framework for Drone Delivery

TL;DR

This work introduces Quantum Unmanned Aerial Delivery Routing Optimization (QUADRO), a novel hybrid framework addressing the Energy-Constrained Capacitated Unmanned Aerial Vehicle Routing Problem and the Unmanned Aerial Vehicle Scheduling Problem, formulating these challenges as Quadratic Unconstrained Binary Optimization problems.

Abstract

Quantum computing holds transformative potential for optimizing large-scale drone fleet operations, yet its near-term limitations necessitate hybrid approaches blending classical and quantum techniques. This work introduces Quantum Unmanned Aerial Delivery Routing Optimization (QUADRO), a novel hybrid framework addressing the Energy-Constrained Capacitated Unmanned Aerial Vehicle Routing Problem and the Unmanned Aerial Vehicle Scheduling Problem. By formulating these challenges as Quadratic Unconstrained Binary Optimization problems, QUADRO leverages the Quantum Approximate Optimization Algorithm for routing and scheduling, enhanced by classical heuristics and post-processing. We minimize total transit time in routing, considering payload and battery constraints, and optimize makespan scheduling across various drone fleets. Evaluated on adapted Augerat benchmarks (16-51 nodes), QUADRO competes against classical and prior hybrid methods, achieving scalable solutions with fewer than one hundred qubits. The proposed results underscore the viability of hybrid quantum-classical strategies for real-world drone logistics, paving the way for quantum-enhanced optimization in the Noisy Intermediate Scale Quantum era.

Paper Structure

This paper contains 17 sections, 47 equations, 4 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background
  3. Drone Routing and Scheduling
  4. Quantum Approximate Optimization Algorithm
  5. Quadratic Unconstrained Binary Optimization (QUBO) to Quantum Approximate Optimization Algorithm (QAOA)
  6. Related Work
  7. Problem Formulation
  8. Drone Routing Problem
  9. Drone Scheduling Problem
  10. Our Proposed Method
  11. Routing Phase
  12. Scheduling Phase
  13. Implementation Details
  14. Experiments and Results
  15. Experimental Setup
  16. ...and 2 more sections

Figures (4)

  • Figure 1: QUADRO: Framework for Drone Routing and Scheduling
  • Figure 2: Drone delivery example with a two-drone fleet. The corresponding colored drone services colored routes. The central location is the depot, where the drones are loaded and recharged between routes.
  • Figure 3: The proposed flow of our framework
  • Figure 4: Comparison of average routing transit times (in minutes) across eight Augerat dataset instances for QUADRO, HQTS, and HGSA, showcasing the performance of our quantum-enhanced drone routing framework QUADRO against the prior hybrid method HQTS and a classical benchmark HGSA sajid2022routing. All methods exhibit rising transit times with increasing node counts, with QUADRO closely tracking HQTS while exceeding HGSA’s best-case results.