Differentiable skill optimisation for powder manipulation in laboratory automation

Wei, Minglun, Yang, Xintong ORCID: https://orcid.org/0000-0002-7612-614X, Lai, Yukun ORCID: https://orcid.org/0000-0002-2094-5680 and Ji, Ze ORCID: https://orcid.org/0000-0002-8968-9902 2025. Differentiable skill optimisation for powder manipulation in laboratory automation. Presented at: IEEE/RSJ International Conference on Intelligent Robots and Systems, Hangzhou, China, 19-25 October 2025.

PDF - Submitted Pre-Print Version Download (1MB)

Abstract

Robotic automation is accelerating scientific discovery by reducing manual effort in laboratory workflows. However, precise manipulation of powders remains challenging, particularly in tasks such as transport that demand accuracy and stability. We propose a trajectory optimisation framework for powder transport in laboratory settings, which integrates differentiable physics simulation for accurate modelling of granular dynamics, low-dimensional skill-space parameterisation to reduce optimisation complexity, and a curriculum-based strategy that progressively refines task competence over long horizons. This formulation enables end-to-end optimisation of contact-rich robot trajectories while maintaining stability and convergence efficiency. Experimental results demonstrate that the proposed method achieves superior task success rates and stability compared to the reinforcement learning baseline.

Item Type:	Conference or Workshop Item (Paper)
Status:	Unpublished
Schools:	Schools > Engineering Schools > Computer Science & Informatics
Last Modified:	14 Nov 2025 12:45
URI:	https://orca.cardiff.ac.uk/id/eprint/182419

Actions (repository staff only)

Edit Item