Xintong (Ian) Yang

Research Associate (Postdoc) · Robotics & Manipulation · Cardiff University

I’m a postdoctoral research associate in the School of Engineering at Cardiff University. My research focuses on enabling robots to manipulate real-world objects — rigid, deformable, or granular — through both learning-based and model-based methods. Since my PhD, I have published learning/model-based robotic manipulation works in top-tier robotic journals, including IJRR, IEEE TRO, IEEE TNNLS, etc. I have first-hand experience with and advocate for harnessing and testing agentic programming tools (e.g., Claude Code, Codex, Copilot) to develop autonomous robotic systems, including general-purpose manipulation and laboratory robots. My ambition in the academic world is to develop safe, resilient and trustworthy real-world robotic systems that collaborate with humans.

I am passionate about open-source development and believe that knowledge should be shared freely. Before my PhD, I received my Bachelor’s and Master’s degrees in Mechanical and Industrial Engineering from Guangdong University of Technology, China.

Contact & Profiles

Email Google Scholar LinkedIn Zhihu

→ Research Projects & Papers

News

[2025 Nov.] Our paper titled “DDBot: Differentiable Physics-Based Digging Robot for Unknown Granular Materials” has been published on IEEE TRO! , .
[2025 Oct.] Our paper titled “Celebi’s Choice: Causality-Guided Skill Optimisation for Granular Manipulation via Differentiable Simulation” has been published at IROS 2025.
[2025 Feb.] Our paper titled “Differentiable Physics-based System Identification for Robotic Manipulation of Elastoplastic Materials” has been accepted by IJRR! , . project page.
[2025 Jan.] I have been recruited by a new BBSRC project as a research associate at Cardiff University, working on the development of an automatic/robotised biology lab.
[2025 Jan.] Our new preprint titled “AutomaChef: A Physics-informed Demonstration-guided Learning Framework for Granular Material Manipulation” is now online
[2024 Jan.] Our new preprint titled “Gam: General Affordance-Based Manipulation for Contact-Rich Object Disentangling Tasks” is now online , with video demo .

Representative Publications

2025	Wei M., Yang, X., Yan J., Lai Y.-K., Ji Z. Celebi's Choice: Causality-Guided Skill Optimisation for Granular Manipulation via Differentiable Simulation. IROS (Published).
2025	Yang, X., Minglun W., Lai Y., Ji Z. DDBot: Differentiable Physics-Based Digging Robot for Unknown Granular Materials. IEEE TRO (Published). (Preprint )
2025	Yang, X., Lai Y., Ji Z. Differentiable Physics-based System Identification for Robotic Manipulation of Elastoplastic Materials. IJRR (Published). (Preprint )
2024	Yang, X., Wu J., Lai Y., Ji Z. Gam: General Affordance-Based Manipulation for Contact-Rich Object Disentangling Tasks. Neurocomputing. (Preprint )
2023	Gao Y., Wu J., Yang, X., Ji Z. Efficient Hierarchical Reinforcement Learning for Mapless Navigation with Predictive Neighbouring Space Scoring. IEEE TASE.
2023	Yang, X. Robotic Manipulation via Hierarchical and Affordance Learning. PhD thesis.
2023	Yang, X., Ji, Z., Wu, J., Lai, Y. Recent Advances of Deep Robotic Affordance Learning: A Reinforcement Learning Perspective. IEEE TCDS. (Preprint )
2022	Yang, X., Ji Z., Wu J., Lai Y. Abstract Demonstrations and Adaptive Exploration for Efficient and Stable Multi-Step Sparse Reward Reinforcement Learning. ICAC.
2021	Yang, X., Ji, Z., Wu, J., Lai, Y. An Open-Source Multi-Goal Reinforcement Learning Environment for Robotic Manipulation with Pybullet. TAROS.
2021	Yang, X., Ji, Z., Wu, J., Lai, Y., Wei, C., Liu, G. & Setchi, R. Hierarchical Reinforcement Learning with Universal Policies for Multi-Step Robotic Manipulation. IEEE TNNLS.