Intelligent Cobotic Systems with LiDAR Perception and LLM-Based Reasoning

About the Project

OPEN FOR UK AND REPUBLIC OF IRELAND STUDENTS ONLY

SPEAR Research Programme: Fully Funded PhD Opportunities

The Semiconductor and Photonics Education and Research (SPEAR) Centre, funded by PEACEPLUS and managed by the Special EU Programmes Body (SEUPB), is a cross-border project that will provide Ulster University and ATU with access to an all-island network of research groups and industry partners. The project receives strategic support from Tyndall and advisory support from Seagate Technology. The SPEAR Centre is a photonics research, training, and innovation response to the challenges outlined in the EU Chips Act 2023, while also addressing existing deficits in high skill/high-value employment and research infrastructure in the border region.

A key element of the project is a doctoral training initiative comprising of 15 PhD students, delivered in collaboration with ATU, Tyndall National Institute, and Seagate Technology, a global leader in data storage and photonics innovation. Three PhD students will be based at Ulster University (Derry~Londonderry campus) and will join a collaborative Doctoral College alongside PhD students at ATU (Letterkenny campus) and Tyndall National Institute. This initiative involves co-supervised research, joint training activities, summer schools, industry engagement, and access to advanced infrastructure. The following PhD studentship is now open for recruitment to the SPEAR Doctoral College.

This PhD project focuses on the development of intelligent collaborative robotic (cobotic) systems that can safely and effectively operate alongside humans in shared environments. The project will integrate LiDAR-based perception, artificial intelligence, robotics, and Large Language Models (LLMs) to enable real-time human detection, behaviour prediction, and intelligent robot decision support.

The research will develop human-aware cobotic systems capable of understanding human movement, interpreting context, and supporting adaptive robot behaviour in dynamic environments. This work will contribute to next-generation Industry 5.0 systems, enabling seamless and intelligent human–robot collaboration.

Research Objectives

• Develop LiDAR-based human detection and tracking, using 3D deep learning (point cloud processing, transformers, etc.)
• Design AI models for human behaviour and trajectory prediction (e.g. sequence learning, trajectory prediction)
• Develop LLM-based reasoning framework for interpreting human activity and context.
• Enable natural language interaction between humans and cobots.