CPS Seminar: Tuersday 16 July 2026, 14:00 - 15:00

Building 53, room 4025

Speaker: Dr Tomasz Szydlo, Newcastle University, UK

Title: Towards Tiny Autonomic Computing

Abstract: Modern IoT devices increasingly use TinyML, embedded machine learning, and edge analytics, but real-world deployment can disrupt expected behaviour through sensor noise, changing environments, limited energy, unreliable connectivity, hardware degradation, evolving workloads, and security threats. Therefore, edge intelligence should be adaptive, context-aware, and resilient.

This talk presents initial results on Tiny Autonomic Computing, which extends the well-established autonomic computing approach to resource-constrained microcontroller systems. It explores lightweight self-management loops, TinyML models, anomaly detection, and contextual adaptation to support self-configuration, self-healing, self-optimisation, and self-protection at the extreme edge.

Bio: Dr Tomasz Szydlo is a Senior Lecturer at Newcastle University specialising in intelligent and distributed systems, with expertise in edge-cloud computing, IoT, and resilient AI infrastructures. His research focuses on adaptive, resource-aware systems for deploying AI in dynamic, real-world environments, including cyber-physical and smart infrastructure applications.

His work is informed by industrial experience supporting the translation of advanced research into practical impact.

He also serves as a Co-Lead in the EPSRC-funded National Edge AI Hub, where he works on TinyML algorithms. He is the main lead of the FogML framework, enabling TinyML on resource-constrained embedded devices.