Lucas Barcelos

Papers

Total Citations

H-Index

About

Lucas Barcelos is a researcher advancing the frontiers of motion planning and control for complex robotic systems. His work lies at the intersection of trajectory optimisation, probabilistic inference, and simulation-based control, tackling the fundamental challenge of navigating robots through intricate, obstacle-dense environments. Barcelos’s key contribution is the development of novel frameworks that overcome the notorious problem of local minima in trajectory optimisation. In his highly cited 2024 paper, "Path signatures for diversity in probabilistic trajectory optimisation," he introduces a method that uses path signatures to generate a diverse set of high-quality trajectory candidates, dramatically improving solution robustness in geometrically complex spaces. His earlier work, "DISCO: Double Likelihood-free Inference Stochastic Control" (2020), addresses the critical gap between advanced simulators and real-world deployment by enabling control strategies to be developed and tested without requiring analytically tractable models. Though early in his career, Barcelos’s innovative fusion of probabilistic methods and control theory is already shaping how researchers approach motion planning, promising safer and more reliable autonomous systems.