I am Shuqi (Suky) Wang, a PhD student in the School of Automation and Intelligent Sensing, Shanghai Jiao Tong University (SJTU), advised by Prof. Xiang Yin in the CPS group. I work on safety control theory for AI-embedded cyber-physical systems, with a particular focus on using conformal prediction as a safety interface between learning-based perception/decision modules and physical control loops. More >

Prior to my PhD, I earned my bachelor's degree in Artificial Intelligence (Robotics) Honored Class from SJTU. I have also worked at the National University of Singapore (NUS) AI4SG Lab with Prof. Yi-Chieh (EJ) Lee.

Research Interests:

  • Conformal Safety Interfaces. Wrapping black-box AI modules (VLMs, neural predictors, learned dynamics) with distribution-free, finite-sample guarantees, and propagating these guarantees through planning and control layers for end-to-end safety.
  • Uncertainty-Aware Learning-Based Control. Chance-constrained neural MPC, sampling-based control (MPPI) with information-theoretic guarantees, and formal diagnosability of discrete-event systems under statistical uncertainty.

If you are interested in research collaboration, feel free to contact me.

Google Scholar| GitHub| LinkedIn| Email

#Research Highlights

CoRL 2026

CoSMO: Conformal Semantic Mapping for Open-World Safe Robot Navigation

Conformal semantic mapping turns vision-language embeddings into calibrated safety maps for open-world robot navigation, adapting to new hazards and distribution shifts with delayed expert feedback.

Conformal Prediction Robot Navigation Open World Safety Guaranteed VLM
NAHS 2026

Chance-Constrained Neural MPC under Uncontrollable Agents

Neural MPC provides formal probabilistic safety guarantees against stochastic, state-coupled uncontrollable agents. Region-wise robust conformal prediction bounds time-dependent uncertainty under policy-induced distribution shifts.

Neural MPC Robust CP Convex Optimization

#Selected Publications

CoRL
2026
CoSMO: Conformal Semantic Mapping for Open-World Safe Robot Navigation
Shuqi Wang, Yiwei Zheng, Pingrui Zhang, Yue Gao, Xiang Yin
Conference on Robot Learning (CoRL), 2026
NAHS
2026
Chance-constrained neural MPC under uncontrollable agents via sequential convex programming
Shuqi Wang, Mingyang Feng, Yu Chen, Yue Gao, Xiang Yin
Nonlinear Analysis: Hybrid Systems, vol. 62, 101751, 2026
npj Digit. Med.
2026
Mapping the role of artificial intelligence in health-related stigma: a scoping review
Tianqi Song, Jack Jamieson, Wataru Akahori, Han Meng, Shuqi Wang, Yi-Chieh Lee
npj Digital Medicine, 2026
arXiv
2024
SPARC: Prediction-based safe control for coupled controllable and uncontrollable agents with conformal predictions
Shuqi Wang, Siqi Wang, Shaoyuan Li, Xiang Yin
arXiv preprint arXiv:2410.15660, 2024
IFAC WC
2023
Failure-aware self-diagnostic task planning under temporal logic specifications
Jianing Zhao, Shuqi Wang, Xiang Yin
IFAC-PapersOnLine, vol. 56, no. 2, pp. 4582–4588, 2023

#Academic Services

Reviewer

  • IEEE Conference on Decision and Control (CDC) 2026
  • IEEE Transactions on Automatic Control 2026
  • Neural Information Processing Systems (NeurIPS) 2026
  • Conference on Robot Learning (CoRL) 2026

#Teaching

Teaching Assistant

  • CS2501H: Discrete Mathematics (Honor) (Zhiyuan Honor Course) 2026 Fall
  • AU7022H: Stochastic Methods in Systems & Control (Honor), Zhiyuan Honor Course 2025 Fall

#Awards & Honors

  • 2024  Outstanding Graduate of Shanghai
  • 2024  Ye Jun & Shen Nanpeng Zhiyuan Outstanding Scholarship (14 of ~4,000 SJTU students per year)
  • 2021 – 2024  Fan Hsu-chi Scholarship, three consecutive years (15 of ~4,000 SJTU students per year)
  • 2020 – 2024  Zhiyuan Honor Scholarship, four consecutive years

#About Me

In Isaac Asimov's Robots and Empire, the Spacers fell not because they used robots, but because they used robots to avoid becoming pioneers themselves. This story has stayed with me. I hope to contribute, even in small ways, to a future in which humans use robots not merely to preserve comfort, but to better understand robots, the world, and ourselves. Human–robot cooperation should help us move safely beyond our local optima, rather than trapping us inside them.

Email: wangshuqi@sjtu.edu.cn