Ma'ayan is a PhD candidate in the Department of Informatics in King's College London,  aligned with the UKRI Centre for Doctoral Training in Safe and Trusted Artificial Intelligence.

Her research focuses on leveraging common-sense knowledge for the improving the process of robot task planning. More specifically, it aims to improve the safety and trustworthiness of plans (sequences of actions) that are generated for a robot to execute, by using structured knowledge bases (knowledge graphs) and aligning the robot's behaviour with human expectations.

Before her PhD studies, Ma'ayan has completed her bachelor’s degree in computer science at King’s College London. Alongside her Bachelor studies, she was working as a Research Intern UKRI Hub for Trustworthy Autonomous Systems.

Project Title: Common Sense Planning for Robotics

Abstract: Robot task planning techniques are being developed constantly, aiming to maximise the feasibility and efficiency of the generated plans, to better achieve the desired goals. Large language models are being incorporated into planning strategies more frequently, and the use of LLMs to improve planners is now common. Due to LLMs training on textual corpora, in which common sense knowledge is rarely included, they lack certain knowledge which is intuitive to humans.

This PhD aims to leverage common sense knowledge graphs to increase the safety of planning approaches and the user’s trust in them. This is to be achieved by integrating knowledge about human behaviour into the planning process, whether for plan creation, evaluation, or repair.

Research Interests

• AI Task Planning
• Knowledge Representation
• Knowledge Graphs
• Robotics
• AI Safety
• Explainable AI

Research Centres or Groups

• Reasoning and Planning (RAP)
• LASR Robotics team
• King's Knowledge Graph Lab

PhD Supervision

• Primary Supervisor: Dr Gerard Canal
• Secondary Supervisor: Dr Albert Meroño-Peñuela

Website

• PURE Profile
• CDT Profile
• Google Scholar