Computational Modelling Group

Design of Unmanned Air Vehicles

22nd September 2014
Research Team
Robert Entwistle
James Scanlan

One of the universities latest unmanned aircraft for scientific applications.

This project will the exploit the success the university has recently had in attracting over £3M in funding for equipment to support autonomous systems and the next generation of 3D printing to produce “structronics” for platforms.

This project will apply computational techniques to the problem of designing unmanned air vehicles to meet the safety and collision avoidance requirements of the certification authorities. It will investigate the safety and reliability of unmanned systems in the context of shared airspace. The research will look at various “sense and avoid” technologies and strategies.

An outline of the work is as follows:

  1. Creation of a 3D airspace simulation environment.

  2. Modelling of various types of traffic including VFR/ IFR and recreational (gliders, hot air balloons, private flying) manned aircraft.

  3. Introduction of various UAV systems into this environment with varying degrees of sense and avoid capabilities.

  4. The environment will be expanded to model reliability and redundancy of systems including communications links and safety critical avionics systems to analytically assess whether the system meets the 10-9 certification safety levels including collision probabilities.

  5. The environment will model other items such as weather, terrain and areas of high population density/ surface transportation features.

  6. This environment will be developed such that it can be integrated into a design analysis/optimisation environment to be used for platform design and decision support.

  7. Design and build example small unmanned systems to help validate the computation models.


Algorithms and computational methods: Agents

Programming languages and libraries: C, Matlab, Python

Computational platforms: Linux, Mac OS X, Windows

Transdisciplinary tags: Complex Systems, Design, NGCM