Unmanned vehicles and automation

From self-driving cars to motion control algorithms, our research uses data-driven techniques and algorithms to achieve real-world goals.

About the team

In the unmanned vehicles and automation team, we develop new control methods and technologies for unmanned vehicles used in areas such as self-driving cars, smart transportation, and agricultural robotics. By combining data-driven techniques, predictive control, estimation, and machine learning, we create distributed algorithms that allow groups of intelligent vehicles or robots to work together and accomplish common goals.

We also develop autonomous robot navigation systems and motion control algorithms for mobile robots operating in multi-terrain and dynamic farming environments.

Computer vision and deep learning algorithms are used in our research on precision livestock management including animal tracking and health monitoring and grass height measurement and biodiversity assessment in pasture-based farms from drones.

In addition, industrial automation using advanced and intelligent control techniques for different manufacturing processes is also a focus of our research.

Research topics

Explore all our unmanned vehicles and automation research:

Informative path planning algorithms

Informative path planning algorithms are critical in applications such as disaster management, where efficient information gathering is required in previously unknown environments. This problem is inherently complex, as it involves computing globally optimal paths that maximise information gain (e.g. constructing the most complete map with minimal travel distance) while relying on partial and uncertain local measurements.

Our research develops such algorithms for optimally mapping unexplored areas and for generating maps of agricultural fields infested with weeds. These algorithms integrate data-driven, stochastic, and machine learning techniques to address the challenges of uncertainty and complexity. Their performance is rigorously assessed using our in-house platform, RobiL, which leverages both real robots and their digital twins to evaluate distributed algorithms.

  • Academics
  • References

    [1] M. O. Orisatoki, M. Amouzadi and A. M. Dizqah, "A Heuristic Informative-Path-Planning Algorithm to Map Unknown Areas and A Benchmark Solution," 2024 IEEE Conference on Control Technology and Applications (CCTA), Newcastle upon Tyne, United Kingdom, 2024, pp. 254-261, doi: 10.1109/CCTA60707.2024.10666559.
    [2] M. O. Orisatoki, M. Amouzadi and A. M. Dizqah, "A Heuristic Informative-Path-Planning Algorithm for Autonomous Mapping of Unknown Areas," arXiv preprint 2023, doi: 10.48550/arXiv.2308.12209.

  • Funders
    • HEIF
  • Collaborators
    • Qatar University
    • Technical University of Denmark

Distributed path planning algorithms of connected and autonomous vehicles

Unlike today’s cars, connected and autonomous vehicles (CAVs) don’t need to stay in fixed lanes when crossing intersections. Instead, they can move through in a coordinated, lane-free way, which greatly improves traffic flow compared to stoplights.

Our research develops smart distributed path planning algorithms that help CAVs minimise travel time and fuel use. These algorithms use predictive control and machine learning to coordinate vehicles effectively. We also tackle the challenge of measuring just how much improvement lane-free intersections provide, since traditional traffic analysis tools don’t apply here. Our results show that lane-free coordination can boost intersection capacity by 127% compared to human-driven vehicles and 36% compared to signalised CAVs. We test and refine these methods with our custom-built tool, RobiL, which combines real robots and their digital twins to evaluate distributed multi-agent systems like fleets of CAVs.

  • Academics
  • References

    [1] M. Amouzadi, M. O. Orisatoki, and A. M. Dizqah, “Optimal lane-free crossing of cavs through intersections,” IEEE Transactions on Vehicular Technology, vol. 72, no. 2, pp. 1488–1500, 2023, doi: 10.1109/TVT.2022.3207054.

    [2] Mahdi Amouzadi, Mobolaji O. Orisatoki, and A. M. Dizqah, “Lane-free crossing of CAVs through intersections as a minimum-time optimal control problem,” IFAC-PapersOnLine, vol. 55, no. 14, pp. 28–33, 2022.

    [3] Amouzadi, Mahdi, Mobolaji O. Orisatoki, and Arash M. Dizqah, "Capacity Analysis of Intersections When CAVs Are Crossing in a Collaborative and Lane-Free Order" Future Transportation 2, no. 3: 698-710, 2022. doi: 10.3390/futuretransp2030039

  • Funders
    • HEIF
  • Collaborators

    •  Qatar University

Propulsion systems for modern vehicles

Connected vehicles can “see ahead” using their sensors and communication systems—and this look-ahead information can make engines and powertrains much smarter. In this research, we develop advanced control strategies that combine predictive road information with hierarchical economic model predictive control (HeMPC). For diesel engines, our controller uses both short- and long-term previews to manage fast and slow dynamics, while adjusting components like the turbocharger and exhaust gas recirculation in real time to reduce emissions and improve fuel economy.

We also study electrified vehicles, where the design of planetary gear systems has a big impact on energy efficiency. By creating an optimal energy management strategy, we determine the best sequence of clutch operations and torque distribution across drivetrains. Using our hardware-in-the-loop simulator, we show how different gear configurations can minimise overall energy consumption.

The key innovation here is the HeMPC approach. Unlike traditional predictive control, which simply follows a set trajectory, HeMPC actively finds the best trajectories and tracks them seamlessly, adapting on the fly to maximise efficiency.

  • Academics
  • References

    [1] Z. Liu, A. M. Dizqah, J. M. Herreros, J. Schaub, and O. C. L. Haas, “A hierarchical economic model predictive controller that exploits look-ahead information of roads to boost engine performance,” IEEE Transactions on Control Systems Technology,
    vol. 31, no. 6, pp. 2632–2643, 2023, doi: 10.1109/TCST.2023.3282051.

    [2] D. Rajput, J. M. Herreros, M. Innocente, J. Bryans, J. Schaub, A. M. Dizqah, “Impact of the Number of Planetary Gears on the Energy Efficiency of Electrified Powertrains,” Applied Energy, vol. 323, p. 119531, 2022, doi: 10.1016/j.apenergy.2022.119531.

    [3] Z. Liu, A. M. Dizqah, J. M. Herreros, J Schaub, O. Haas, “Simultaneous Control of NOx, Soot and Fuel Economy of a Diesel Engine with Dual-Loop EGR and VNT using economic MPC,” Control Engineering Practice, vol. 108, p. 104701, 2021, doi: 10.1016/j.conengprac.2020.104701.

  • Funders
    • FEV, Germany
  • Collaborators
    • University of Birmingham
    • FEV, Germany
    • Coventry University

Autonomous robot navigation in dynamic farming environments

Precision and smart farming techniques using robotic systems have played a significant role in different agricultural applications as they reduce human labour and enhance the operation safety. Pasture-based farms with variations of terrains and the humans and animals surrounding them, make robot operation more challenging. This research focuses on autonomous navigation and control challenges of robotic systems operating in dynamic outdoor farming environment with the presence of animals and humans.

Animal tracking and monitoring in outdoor farming environments

This project develops a real-time animal tracking and monitoring system using computer vision and deep learning techniques that can be used in both indoor and outdoor farms.

Grass height measurement and biodiversity assessment in pasture-based farms

Pasture quality is one of the critical inputs to pasture livestock farming. Various methods have been used to measure grass height, which is an important indicator of growth conditions, including mechanical / electronic rising plate meter, ultrasonic sensor, mounted / trailed devices.

However these measurements are labour-intensive and time-consuming as most of them are carried out manually on a weekly basis throughout the grazing season. This project is to develop new methods to measure vegetation height and investigate biodiversity levels in pasture-based farms using computer vision systems and deep learning techniques. This proposed measurement system will aid farmers’ decision-making and to make the livestock farming industry more efficient, resilient and sustainable.

  • Academics
  • References

    [1] Wang J., Ying RH., Liu CX., Wei GY., Birch P., Nguyen BK., A robust architecture search for sward height estimation from UAV-derived Digital Surface Models, Computers and Electronics in Agriculture, Volume 238, 2025, 110789.

    [2] J. Wang, N. Oishi, R. Ying, G. Wei, P. Birch and B. K. Nguyen, "Keep the Key Part: Exploring Drone-Captured Digital Elevation Model Data Augmentation for Deep Learning-based Crop Height Estimation," 2024 IEEE International Conference on Visual Communications and Image Processing (VCIP), Tokyo, Japan, 2024, pp. 1-5, doi: 10.1109/VCIP63160.2024.10849876.

    [3] Wang, J.; Oishi, N.; Birch, P.; Nguyen, B.K. G-DMD: A Gated Recurrent Unit-Based Digital Elevation Model for Crop Height Measurement from Multispectral Drone Images. Machines 2023, 11, 1049.

  • Funders
    • HEIF
    • KEI
  • Collaborators
    • Saddlescombe Farm

Sensor-base ultrasonic viscosity control for the extrusion of recycled plastics – ultravisc project

This is a European FP7 project, led by QUB in partnership with five SMEs and two RTDs across Europe, to develop a real time control system to control the viscosity of recycled polymers in extrusion processes using ultrasonic modulation.

  • Academics
  • References

    1] Nguyen B K, McNally G, Clarke A, Automatic extruder for processing recycled polymers with ultrasound and temperature control system. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering. 2014, 230(5):355-370.

    [2] Nguyen B K, McNally G, Clarke A, Real time measurement and control of viscosity for extrusion processes using recycled materials, Polymer Degradation and Stability, 2014, Vol. 102, pp. 212-221.

  • Funders
    • EU FP7
  • Collaborators
    • DKI Plast (Denmark)
    • Plastitehase (Estonia)
    • Polinter (Spain)
    • TSM (Ireland)
    • UKMaTri
    • QUB (UK)
    • Sirris (Belgium)