Ruihan Wu

Hey there!

I am Ruihan Wu, a senior undergraduate in Information Engineering at Southern University of Science and Technology (SUSTech).

At Southern University of Science and Technology, I work with Prof. Shuxiang Guo at the Guo Lab. I spent a semester at the University of Pennsylvania as an exchange student, and later I was lucky to work with Prof. Haimin Hu at Johns Hopkins University, where I joined the Laboratory for Computational Sensing & Robotics (LCSR) and the Alliance AI Lab.

My research interests broadly center on robotics and control. To me, exploring robotics is ultimately a process of understanding human nature and ourselves.

news

Sep 23, 2026	Our work, “Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning”, is now available on arXiv.
Aug 07, 2026	Presented our paper at ICMA in Jilin, China (Oral).
May 14, 2026	Our work, “Deep Hydro-Sim: Structured Residual Hydrodynamics for the RL-Based AUV Control,” was accepted to ICMA 2026.

selected publications

arXiv
Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning

Ruihan Wu^*, Rui Yang^*, Donggeon David Oh, and 2 more authors

arXiv preprint arXiv:2609.27312, 2026

Abs arXiv Bib HTML PDF Website

Robots deployed for competitive tasks must outmaneuver their opponents without sacrificing safety. Existing approaches, including safe reinforcement learning, train a single policy to achieve task success and avoid failures simultaneously, which can leave the learned policy exploitable by deliberate attacks. We propose Safety to Competence (S2C), a two-stage reinforcement learning framework that separates safety synthesis from competitive task learning. S2C learns a robust safety filter via adversarial reinforcement learning, embeds it during task policy training, and retains the same filter at deployment. In simulated touchdown games, S2C achieves the highest win rate and Elo rating and the lowest exploitability among safe reinforcement learning baselines; hardware stress tests against a human opponent further confirm its competence.
@article{wu2026competence, title = {Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning}, author = {Wu, Ruihan and Yang, Rui and Oh, Donggeon David and Nguyen, Duy and Hu, Haimin}, journal = {arXiv preprint arXiv:2609.27312}, year = {2026}, }
ICMA
Deep Hydro-Sim: Structured Residual Hydrodynamics for the RL-Based AUV Control

Ruihan Wu, Chunying Li, Qirong Lei, and 3 more authors

In 2026 IEEE International Conference on Mechatronics and Automation (ICMA), 2026

Abs Bib PDF Code

Deep Hydro-Sim bridges a high-fidelity underwater simulator and a fast rigid-body engine by learning residual hydrodynamic forces from Stonefish and injecting them into MuJoCo as a real-time plugin. The result improves open-loop replay fidelity and stabilizes downstream SAC training for AUV control without sacrificing throughput.
@inproceedings{wu2026deepsim, title = {Deep Hydro-Sim: Structured Residual Hydrodynamics for the RL-Based AUV Control}, author = {Wu, Ruihan and Li, Chunying and Lei, Qirong and Huang, Le and Li, Pengcheng and Guo, Shuxiang}, booktitle = {2026 IEEE International Conference on Mechatronics and Automation (ICMA)}, year = {2026}, }