Kanghui He (何康辉)

Kanghui He [Google Scholar] Control Group Department of Engineering Science University of Oxford Email: kanghui.he@eng.ox.ac.uk

About Me

I am a postdoctoral researcher in the Department of Engineering Science, University of Oxford, U.K. My advisor is Prof. Antonis Papachristodoulou. I obtained PhD under the supervision of Prof. Bart De Schutter, Prof. Ton van den Boom, and Dr. Shengling Shi at the Delft Center for Systems and Control, Delft University of Technology, the Netherlands. I received the B.S. degree in the School of Mechanical Engineering and Automation, Beihang University (BUAA), in 2018, and the M.S. degree (with an outstanding graduation thesis and Chinese national scholarship) in the Department of Dynamics and Control, Beihang University, in 2021. I was a research assistant in the group of Dr. Zhang Chen, in the Department of Automation, Tsinghua University, China, in 2021.

My Postdoc research: Integrated Learning and Optimisation for Safety-Critical Control is funded by the Rubicon Fellowship, NWO. My PhD research topic is about learning-based control under constraints, which is a part of an ERC Advanced Project “A novel control paradigm for large-scale hybrid networks” (CLariNet).

Research Interests

My main interests involve developing learning-based control methods with safety guarantees and computational efficiency, and their applications in mobile robots. I use tools from control theory (model predictive control, control barrier functions), optimization (mixed-integer programming, parametric programming, stochastic programming), and machine learning (reinforcement learning, Gaussian process).

Master's Projects

I am actively looking for supervision of master's thesis projects at the University of Oxford (can be remote). Please see the description of master's projects. Feel free to contact me if you are interested. The topics include

• generative policies for robot planning

• integrated RL and MPC for vehicle–traffic coordination

• machine learning for speeding up mixed-integer optimization with applications in multi-contact robots

• Safe learning and control in latent-space world Models,

Recent News