Paul K. Grant

About

Paul K. Grant’s research bridges the frontiers of Bayesian statistics and autonomous robotics, with a focus on scalable inference and intelligent navigation. His most influential work, “Efficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems” (2019, 10 citations), introduces a flexible framework for modeling complex, hierarchical variability in nonlinear systems—a significant advance for fields like pharmacokinetics and population dynamics. This work generalizes nonlinear mixed-effects models, enabling efficient, amortized inference that scales to large datasets. Earlier, Grant contributed to mobile robotics with “Local path planning: a brute force approach” (2002, 5 citations), where he developed a generic, domain-independent local planning module for the TURNIP robot, addressing obstacle avoidance and modular autonomy. This work underscores his commitment to practical, robust solutions in real-world navigation. With a career spanning theoretical statistics and applied robotics, Grant’s contributions demonstrate a rare ability to unify rigorous mathematical frameworks with tangible engineering challenges, making his research valuable for both methodologists and practitioners seeking to push the boundaries of autonomous systems and Bayesian computation.

Research Focus

Key Achievements

Top Papers

1. 1
   Efficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems

   10 citations · 2019
2. 2
   Local path planning: a brute force approach

   5 citations · 2002

Key Collaborators

Contact & Links