Overview
This project focuses on the intersection of Artificial Intelligence (AI) and Formal Methods (FM), with an emphasis on automated planning and synthesis.
About this opportunity
There has been a fast-growing trend in developing autonomous artificial intelligence (AI) systems operating in dynamic, partially known, and unpredictable environments. Autonomous AI systems must self-deliberate their behaviours to accomplish tasks, especially in challenging situations when human instructions are lacking or delayed, e.g., autonomous driving cars. In today’s world, numerous sectors are utilizing autonomous systems, including Robotics, Smart Manufacturing, Cybersecurity, Internet of Things, and Space Exploration, but these systems do not always yield good results due to a lack of formal guarantees.
This deficiency poses a significant hurdle in establishing the trustworthiness of AI systems, particularly impeding their widespread adoption in safety-critical and security-critical domains.
This PhD project aims to advance the design of trustworthy-by-design autonomous AI systems by integrating methodologies from formal verification, automated planning, and recent advances in large language models (LLMs). The project will investigate automated synthesis and planning techniques that generate system controllers whose behaviour provably satisfies high-level specifications. It will explore how LLMs can be used to align system behaviour with human intent, enabling functionalities such as task refinement, natural-language-based goal specification, and multi-task scheduling. Furthermore, the project will develop probabilistic verification techniques to diagnose the root causes of task failures and iteratively refine task plans to improve reliability and robustness. Through this integration, the project seeks to build autonomous systems capable of explaining and adapting their behaviour transparently and verifiably.
Who is this opportunity for?
Candidates will have, or be due to obtain, a Bachelor’s/Master’s Degree or equivalent from a reputable University in an appropriate field of Science and Engineering.
Qualifications:
- Research skills, critical thinking, and analytical abilities
- Strong written and verbal communication skills
- Ability to work independently and as part of a research team
- Knowledge of logical methods, automated reasoning and planning, reactive synthesis
Strong interest in exploring research problems