Project title: Cooperative UAVs and CAVs Integration for Efficient Autonomous Service Delivery in Smart Cities
Eligibility: Indonesian nationality
Duration: Full-Time, 4 years fixed term
Application deadline: 30 April 2026
Interview date: 4-8 May 2026
Start date: September 2026
Contact:

  • Prof. Soufiene Djahel (ae3095@coventry.ac.uk), Centre for Future Transport Cities, Coventry University
  • Prof. Dr. Ir. Hari Muhammad, Ph.D. (hari@itb.ac.id), Flight Mechanics and Operations Research Group, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung
  • Dr. -Ing. Ir. Javensius Sembiring, S.T., M.T. (javensius.sembiring@itb.ac.id), Flight Mechanics and Operations Research Group, Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung

Introduction

Unmanned Aerial Vehicles (UAVs), commonly known as drones, were originally developed for military use where they eliminate the risk of pilot loss during missions in hostile environments. Their flexible deployment, relatively low cost, and modest fixed-infrastructure requirements have made them increasingly attractive for civilian applications, including search-and-rescue, critical-infrastructure monitoring, automated traffic monitoring, and last-mile delivery.

These emerging smart-city applications require UAVs with higher autonomy so that missions can be planned and dynamically adapted under strict constraints, including limited flight time, constrained on-board computation, sensor analytics, weather uncertainty, hardware anomalies, and rapidly changing operational contexts.

This project envisions cooperative swarms of UAVs that communicate with each other and with ground Connected and Autonomous Vehicles (CAVs). The CAVs act as mobile edge-computing platforms, enabling UAVs to redistribute tasks, offload processor-intensive computation, and reorganise mission plans efficiently during autonomous service delivery.

Continuous improvements in UAV construction, embedded computing, communications, sensing, and autonomous navigation are bringing the deployment of autonomous UAV fleets within reach. The research responds to the need for reliable, scalable, and safety-aware UAV-CAV cooperation in smart cities.

Project Details

This PhD project investigates the design, integration, and validation of cooperative UAVs and CAVs for efficient autonomous service delivery in Smart Cities. A heterogeneous fleet of UAVs cooperates with ground CAVs that act as mobile edge-computing platforms to jointly optimise path planning, task redistribution, and computation offloading under dynamic environments and strict time constraints.

The research combines Graph Neural Networks for cooperative path planning, Markov Decision Processes for task redistribution, and coalitional-game theory for offloading. A dedicated risk-quantification layer is introduced to incorporate safety-aware constraints into mission planning and offloading decisions.

The work addresses the central research question: Given a fixed number of UAVs and a set of time-bound tasks, how may an autonomous fleet of UAVs, supported by CAVs, cooperatively plan their flight routes and task-offloading strategies in response to rapid changes in their operational context?

The research will be validated through simulation and a small-scale indoor UAV-CAV testbed using DJI Tello EDU, Sphero RVR+, and Raspberry Pi nodes. The four-year programme is organised around foundation and cooperative path planning, task redistribution and offloading, risk-aware integration and testbed implementation, and final validation, dissemination, and thesis completion.

Possible Research Objectives

  • Develop a near-optimal cooperative path planning solution for a heterogeneous fleet of UAVs that minimises total mission downtime under dynamic and partially-known environments.
  • Design efficient task redistribution and offloading strategies, enabling processor-intensive tasks that exceed UAV capabilities to be offloaded to clusters of CAVs acting as mobile edge-computing platforms.
  • Develop a small-scale indoor UAV-CAV integration testbed using DJI Tello EDU, Sphero RVR+, and Raspberry Pi nodes to validate cooperative planning and offloading strategies and demonstrate a working prototype.
  • Quantify operational and safety risks of cooperative UAV-CAV missions, including collision, communication loss, battery exhaustion, and task failure.
  • Incorporate risk-aware constraints into path planning and offloading algorithms to improve the safety and reliability of autonomous service delivery.
  • Evaluate the proposed framework across smart-city use cases such as traffic monitoring, search-and-rescue, and last-mile delivery.

Funding

Tuition fees and bursary from LPDP, PDDI or potentially ITB/CU.

Benefits

The candidate will receive research training in autonomous UAV systems, CAV-assisted mobile edge computing, intelligent transportation, graph neural networks, Markov decision processes, game-theoretic optimisation, risk-aware mission planning, simulation, and hardware-in-the-loop validation.

The project is expected to produce at least two Q1 journal papers, two conference or demo papers, and a PhD dissertation. Target publication venues include IEEE Transactions on Vehicular Technology, IEEE Transactions on Intelligent Transportation Systems, IEEE Globecom, IEEE ICC, and IEEE VNC.

The candidate will work within an interdisciplinary research environment connecting the Centre for Future Transport Cities at Coventry University and the Flight Mechanics and Operations Research Group at the Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung.

Entry Requirements

  • Applicants should have a strong academic background in a relevant engineering, science, or computing discipline, and should demonstrate the potential to conduct independent doctoral research in autonomous systems, intelligent transportation, UAV operations, optimisation, or data-driven decision making.
  • The project is particularly suited to candidates with interests in UAV autonomy, cooperative robotics, CAV integration, smart cities, edge computing, machine learning, operations research, and safety-aware optimisation.
  • Applicants are expected to be able to complete a four-year full-time PhD programme and contribute to peer-reviewed publications and experimental research outputs.

Academic Requirements

A master degree or strong first degree in a relevant area is desirable. Relevant fields include Aerospace Engineering, Mechanical Engineering, Computer Science, Robotics, Artificial Intelligence, Intelligent Transportation Systems, Operations Research, Mathematics, or related disciplines.

Candidates should have strong analytical and quantitative skills. Experience in one or more of the following areas would be advantageous: graph neural networks, reinforcement learning, Markov decision processes, game theory, optimisation, simulation, UAV control, vehicular networks, mobile edge computing, or embedded platforms.

Experience with tools or platforms such as AirSim, Gazebo, ns-3, Python, MATLAB, ROS, Raspberry Pi, DJI Tello EDU, or Sphero RVR+ would support the experimental and simulation components of the project.

Candidates should also have strong written communication skills, motivation to work in an interdisciplinary research setting, and interest in producing high-quality academic publications.