Zhuoqun Chen

I'm an incoming PhD student at Duke University advised by Prof. Boyuan Chen. Prior to that, I got my M.S. in Electrical and Computer Engineering at UC San Diego, advised by Prof. Xiaolong Wang. I've also been fortunate to work with Prof. Hao Su and Prof. Chen Wang before the doctoral study.

I'm interested in making robots interact with the physical world and complex environments with human-like intelligence by exploring from various data.

Email  /  CV  /  Scholar  /  Twitter  /  Github  /  LinkedIn

Publications & Preprints

* indicates equal contribution

By utilizing action queue of increasing noises and sequential rollout, we enable responsive (higher success rate in envs with rich contact dynamics) and efficient (1 NFEs / action) action generation.

We propose a real-time zero-shot vision-language navigation approach with spatial reasoning that integrates pixel-wise vision-language features and curiosity-based exploration for robots.

Our system learns a Single Instructable Motion Prior from a diverse reference motion dataset.

Miscellanea

coursework, toy projects and some random stuff

By formulating a minigrid path planning as a deterministic optimal control problem, it can be easily solved using dynamic programming algorithm.

Solving simple planning problems in 3D space using A*, RRT, and their variants.

Solving a periodic reference trajectory tracking problem using Model Predictive Control (MPC) and Generalized Policy Iteration (GPI).

With Jun Wang, a project to validate a simple idea, that is, demonstrations collected from specialist policies can help improve base models that are already stucked in a local optimum.

Professional Service

Conference reviewer for CVPR 2025; ICRA 2024, 2025.

Template borrowed from Jon Barron's website.