Software Engineering

Focus areas

AI-based, autonomous and intelligent-adaptive systems

We design, develop, maintain, and test smart systems that are AI-based, autonomous or intelligent-adaptive.

AI technology uses reasoning to detect patterns and anomalies and advise a human user or autonomous system about what action should be taken. Autonomous systems have the power to make decisions, which are often based on AI-based algorithms, and to change their behaviour accordingly. Decisions made by autonomous systems are typically based on a predetermined set of rules, which is interfaced with real-life via technology such as sensors and other data feeds.

Two key areas that we focus on in development of such systems are the planning, scheduling and execution of complex software-controlled tasks, and recommender systems, which link to AI-based or autonomous systems to advise on actions that should or should not be taken.

Exhaustive testing of software systems is crucial, especially in safety-critical scenarios. We develop AI-based algorithms, which rely on – and preserve the properties of – historical data, to generate simulated data that can be used to test systems. Using simulated data means we can test the system under a range of conditions and can add randomness and other variables.

Cyber-physical system and digital twins

We work on cyber-physical systems and digital twins, which usually incorporate AI and autonomous components, and are composed of multiple distributed connected components.

We develop cyber-physical systems: complex AI-based software systems made up of distributed components. A system may comprise multiple devices, surveillance cameras, sensors, computers, as well as centralised computer resources that communicate over a physical network.

We also work on digital twins: software models representing a physical system and its components, such as a manufacturing process. Digital twins are used to help understand the complex processes and interactions occurring in the physical twin and to make decisions – for example, for determining production levels or scheduling maintenance.

Secure and reliable software environments

Another key area for us is ensuring that AI is trustworthy – safe, technically robust and operates in an ethical way. During the software engineering process, we focus on cybersecurity through identification and repair of vulnerabilities; ensuring software is resilient, with the ability to detect threats and self-heal; and continuous analysis of observation data.

Quantum software engineering

Currently, a totally new paradigm of computing is emerging, based on the concepts of quantum mechanics. Such quantum computers will be fundamentally different in design and architecture from classical computers, meaning that several decades’ worth of software engineering concepts and methods need to be revised and extended. Quantum computers are still several years away from being a go-to resource for industry but, together with international partners, we are pioneering software engineering for the quantum age.

Effective digitalisation of the public sector

In this centre we work on digitalisation of current and future services provided by the public sector. Our research approach is empirical, including controlled experiments, observational studies, case studies and surveys with a focus on benefits management in the development of IT systems; flexible work processes and organisation of software development; management and leadership of projects and product development; competence evaluation and resource management; and cooperation and forms of agreement between actors. The research aims to provide knowledge about what leads to successful digitalisation in the public sector and to disseminate knowledge that leads to higher efficiency and more value creation.