Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimal Unmanned Aerial Vehicle Deployment for Macro-Micro Traffic Monitoring Fused with Connected Vehicles

Chaopeng Tan, Jiarong Yao, Meng Wang

TL;DR

This work addresses the challenge of estimating both macroscopic and microscopic traffic states by fusing limited UAV observations with sampled connected-vehicle data. It proposes a joint UAV placement framework that minimizes macro-path-flow uncertainty via PFRE, and microscopic queue-length and arrival-rate uncertainty via area-based metrics, solved with a quantum genetic algorithm (QGA). The core contributions include the first explicit CV-UAV collaboration for macro-micro monitoring, entropy-based and area-based uncertainty measures, and a practical UAV deployment optimization that scales to larger networks. Empirical results on a Qingdao-like network show that deploying a small number of UAVs can substantially reduce total uncertainty (up to 75.69%), with five UAVs sufficing to observe over 95% of paths and achieve meaningful macro-micro state improvements. The framework offers a principled, data-driven planning tool for urban traffic management that makes efficient use of limited UAV resources to improve reliability of traffic state estimation.

Abstract

Reliable estimation of macro and micro traffic states is essential for urban traffic management. Unmanned Aerial Vehicles, with their airborne full-sample continuous trajectory observation, bring new opportunities for macro- and micro-traffic state estimation. In this study, we will explore the optimal UAV deployment problem in road networks in conjunction with sampled connected vehicle data to achieve more reliable estimation of macroscopic path flow as well as microscopic arrival rates and queue lengths. Oriented towards macro-micro traffic states, we propose entropy-based and area-based uncertainty measures, respectively, and transform the optimal UAV deployment problem into minimizing the uncertainty of macro-micro traffic states. A quantum genetic algorithm that integrates the thoughts of metaheuristic algorithms and quantum computation is then proposed to solve the large-scale nonlinear problem efficiently. Evaluation results on a network with 18 intersections have demonstrated that by deploying UAV detection at specific locations, the uncertainty reduction of macro-micro traffic state estimation ranges from 15.28\% to 75.69\%. A total of 5 UAVs with optimal location schemes would be sufficient to detect over 95\% of the paths in the network considering both microscopic uncertainty regarding the intersection operation efficiency and the macroscopic uncertainty regarding the route choice of road users.

Optimal Unmanned Aerial Vehicle Deployment for Macro-Micro Traffic Monitoring Fused with Connected Vehicles

TL;DR

This work addresses the challenge of estimating both macroscopic and microscopic traffic states by fusing limited UAV observations with sampled connected-vehicle data. It proposes a joint UAV placement framework that minimizes macro-path-flow uncertainty via PFRE, and microscopic queue-length and arrival-rate uncertainty via area-based metrics, solved with a quantum genetic algorithm (QGA). The core contributions include the first explicit CV-UAV collaboration for macro-micro monitoring, entropy-based and area-based uncertainty measures, and a practical UAV deployment optimization that scales to larger networks. Empirical results on a Qingdao-like network show that deploying a small number of UAVs can substantially reduce total uncertainty (up to 75.69%), with five UAVs sufficing to observe over 95% of paths and achieve meaningful macro-micro state improvements. The framework offers a principled, data-driven planning tool for urban traffic management that makes efficient use of limited UAV resources to improve reliability of traffic state estimation.

Abstract

Reliable estimation of macro and micro traffic states is essential for urban traffic management. Unmanned Aerial Vehicles, with their airborne full-sample continuous trajectory observation, bring new opportunities for macro- and micro-traffic state estimation. In this study, we will explore the optimal UAV deployment problem in road networks in conjunction with sampled connected vehicle data to achieve more reliable estimation of macroscopic path flow as well as microscopic arrival rates and queue lengths. Oriented towards macro-micro traffic states, we propose entropy-based and area-based uncertainty measures, respectively, and transform the optimal UAV deployment problem into minimizing the uncertainty of macro-micro traffic states. A quantum genetic algorithm that integrates the thoughts of metaheuristic algorithms and quantum computation is then proposed to solve the large-scale nonlinear problem efficiently. Evaluation results on a network with 18 intersections have demonstrated that by deploying UAV detection at specific locations, the uncertainty reduction of macro-micro traffic state estimation ranges from 15.28\% to 75.69\%. A total of 5 UAVs with optimal location schemes would be sufficient to detect over 95\% of the paths in the network considering both microscopic uncertainty regarding the intersection operation efficiency and the macroscopic uncertainty regarding the route choice of road users.

Paper Structure

This paper contains 13 sections, 19 equations, 9 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. Problem Statement
  3. Methodology
  4. Location optimization model
  5. Entropy-based uncertainty of macroscopic traffic state
  6. Area-based uncertainty of microscopic traffic state
  7. Uncertainty of Queue length
  8. Uncertainty of vehicle arrivals
  9. Solution algorithm
  10. Evaluation
  11. Study site
  12. Uncertainty evaluation
  13. Conclusion and Future Work

Figures (9)

  • Figure 1: A sketch of the research scenario.
  • Figure 2: Uncertain area of queue length.
  • Figure 3: Uncertain area of vehicle arrivals.
  • Figure 4: Study site.
  • Figure 5: Uncertainty using different UAV numbers.
  • ...and 4 more figures