[School of Engineering PhD Scholarships] Evidence Before Action: Grounding Large Language Models in Multimodal Physical Evidence for Self-Correcting Robotic Manufacturing

About the project

About the Project Large language and foundation models are starting to play their role in manufacturing decision-making. They can retrieve process knowledge, explain abnormal behaviour and suggest corrective actions. Recent work has shown that sensing, model reasoning and machine action can be connected in a closed loop. However, these systems still rely heavily on the evidence available to them. If the evidence is incomplete or conflicting, they may still suggest actions that are not feasible or safe. A more reliable system should recognise when the available evidence is not enough and actively collect the information needed before taking action. Project Scope: This project will develop an evidence-before-action framework for self-correcting robotic manufacturing. When the cause of a defect is unclear, the robot will collect additional observations to help distinguish between the remaining possible causes. Corrective actions will then be selected from a validated action library and checked against robot and process constraints before execution. Multi-axis large-scale additive manufacturing will be used as the demonstrator. This is a suitable test case because robot motion, material delivery and errors from previous layers can produce similar defects. Large workpieces also make fixed-view inspection difficult. Research Objectives: Develop a multimodal diagnosis method that can distinguish between different defect causes and report uncertainty. Enable the robot to actively select useful inspection viewpoints and sensor poses while considering reachability, collision and visibility. Link diagnosis to safe and feasible corrective actions, including process parameter adjustment, local path correction and material compensation. Demonstrate and evaluate the complete framework on a multi-axis robotic additive manufacturing system. Expected Outcomes: A multimodal dataset with known defect causes generated through controlled fault experiments. A diagnosis model that uses physical evidence and provides uncertainty estimates. An active robotic inspection method for collecting additional evidence when required. A validated action framework integrated into a self-correcting robotic printing demonstrator, together with the evaluation method that can also be adapted to other robotic manufacturing processes. The work is expected to support publications in leading journals and conferences in manufacturing, robotics and automation. Training Opportunities: Multimodal Sensing and Machine Learning: machine vision, 3D scanning, process signal acquisition, multimodal data fusion and uncertainty-aware diagnosis. Foundation Models and Decision Systems: using language models together with physical process evidence, and retrieval of process knowledge and design of decision mechanisms. Robotics and Motion Planning: robot kinematics and inspection path planning, and hands-on experience with robotic system and simulation tools. Advanced Additive Manufacturing: multi-axis large-scale printing and process control, controlled fault experiments, in-process monitoring and quality evaluation. Industrial and Research Networks: engagement with UK-RAS and the Centre for Robotics and Artificial Intelligence and opportunities to take part in workshops, summer schools and research collaborations. Professional and Research Development: department PGR training courses and research seminars, and academic training in technical writing, research integrity and open data. This project is expected to start in September 2027. Before you apply: We strongly recommend that you contact the supervisors for this project before you apply. How to apply: To be considered for this project you must complete a formal application through our online application portal. If you already have an applicant account this link will directly open an application for PhD School of Engineering Scholarships . If you don’t already have an applicant account, please follow the instructions here. . When applying, please specify the full title and supervisor/s of the project, details of your previous study, and names and contact details of two referees. You must also upload a Supporting Statement describing the motivation to apply to the project, your CV and transcripts of awarded and in-progress university qualifications . Please note late or incomplete applications will not be considered. Equality, diversity and inclusion are fundamental to the success of The University of Manchester and central to all our activities. A diverse research community strengthens creativity, productivity and quality, while increasing the societal and economic impact of our work. We welcome applicants from all career paths, backgrounds and sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation or transgender status. We welcome applications from candidates returning to study after a career break or experience in other roles. Flexible study arrangements may be available, including part-time study at 50%, 60% or 80%, subject to the requirements of the project and funder. Eligibility : The standard academic entry requirement for this PhD is an upper second-class (2:1) honours degree (or international equivalent) in Mechanical and Mechatronic Engineering, Manufacturing Engineering, Computer Science or related disciplines OR any upper-second class (2:1) honours degree and a Master’s degree at merit (or international equivalent) in Mechanical and Mechatronic Engineering, Manufacturing Engineering, Computer Science or related disciplines. Experience in autonomous system, manufacturing/robotics and language/vision foundation model development will be an advantage. This project will remain open until filled. If your application is submitted by 1 st November 2026, you can expect a decision by 18 th December 2026. If your application is submitted by 15 th January 2027, you can expect a decision by 30 th March 2027. Self or externally funded students can also be considered for this project. FSESoE

Research areas