Charles Hutchins MINDS CDT, 2020

Knowledge Exchange and Enterprise Fellow, University of Southampton

Charles Hutchins is a MINDS alumnus whose doctoral research focused on the security of flying ad hoc networks — the self-organising communication infrastructure used when swarms of unmanned aerial vehicles coordinate without fixed ground support.

His work addressed one of the harder problems in drone network security: grey hole attacks, where a compromised drone selectively drops packets rather than forwarding them. The challenge is that the effects closely resemble ordinary signal loss caused by a drone moving out of range, which means detection systems can be fooled and the network degraded before anyone acts. Charles built two AI models to address this, one to classify the type of threat from time series data, and one to decide which nodes in the network should receive information during an active incident. He also challenged how drone network security systems were being evaluated, arguing that optimising for a single performance metric often made systems less robust across the full range of real-world threat scenarios.

His placement, hosted at the Alan Turing Institute, took that security thinking into a different context: AI accelerator chips used in applications such as driverless vehicles, where fast and accurate image recognition is safety-critical. The work involved building these chips virtually, developing attacks that could trigger misclassification, and designing testing programmes to detect and remove those vulnerabilities before deployment.

Charles is now a Knowledge Exchange and Enterprise Fellow at the University of Southampton, working on a project developing products for TriboAI. The work uses tribology research and machine learning to detect anomalies and failure modes in machinery through electrostatic sensor data, in practice, that has meant close work with the Offshore Renewable Energies Catapult in Newcastle, testing materials, visiting operational sites and building the kind of industry relationships that applied research depends on.

The Centre for Doctoral Training created the environment to connect with researchers across disciplines