Physics-Embedded Perception and Predictive Modelling for Resilient Human-Robot Collaboration

About the Project

Human-robot collaboration (HRC) is becoming central to manufacturing, healthcare and service robotics, where robots must share dynamic spaces with human partners. Despite recent advances in perception, SLAM and motion prediction, most HRC systems still treat perception, prediction and decision-making as loosely coupled modules. This makes them fragile under cluttered scenes, occlusions and unpredictable human behaviour, where perceptual noise propagates into unstable predictions and overly conservative or unsafe robot actions.

This PhD will explore the unified framework that tightly couples physics-embedded scene reasoning with uncertainty-aware modelling of human motion, enabling proactive and resilient collaboration. Two challenges will be addressed: (1) Physics-embedded scene reasoning, fusing multimodalities with physical priors and domain-specific scene primitives (e.g., manufacturing, care environments), so that the robot understands not only what and where objects are, but how they move, deform and afford interaction; and (2) Uncertainty-aware human motion modelling and control, developing probabilistic, physics-informed predictors conditioned on the embedded scene, propagating uncertainty into joint human-robot occupancy estimates and feeding these into a conflict-aware decision layer that balances short-term safety against long-term task efficiency. The framework will be validated in simulation and on real robotic platforms at UoM.

Eligibility

Applicants should have or expect to achieve at least a UK 2.1 honours degree in Mechanical and Mechatronic Engineering, Electrical and Electronic Engineering, Computer Science or related disciplines. Experience in robotics, autonomous system and machine vision development will be an advantage.

Funding

This is a 3.5-year PhD. Excellent candidates will be nominated for competence-based faculty funding through the School of Engineering. The funding covers tuition fees and provides a tax-free stipend based on the UKRI rate (£21,805 for 2026/27). We expect the stipend to increase each year.

Self-funded students are also welcome to apply.