Ph.D. Student · Computer Science · Rutgers University

Xiaopeng Zhang

I'm interested in robotics and generative models.

Photo of Xiaopeng Zhang
01

About

I'm a Ph.D. student in Computer Science at Rutgers University, advised by Prof. Jingjin Yu. My research sits between robotics and generative modeling.

Before Rutgers, I was a visiting student at the Toyota Technological Institute at Chicago (TTIC), working with Prof. Matthew R. Walter in the RIPL lab, while completing my M.S. in Data Science at the University of Chicago. I received my B.S. in Mathematics and Applied Mathematics from the Institute for Advanced Study, Shenzhen University, where I was advised by Prof. Guojian Yin, graduated first in my class, and received China's National Scholarship (top 0.2% nationwide).

02

News

  1. StackGen3D was accepted to the IROS 2026 Workshop on Creativity in Assembly.

  2. Started my Ph.D. at Rutgers University with Prof. Jingjin Yu.

  3. Completed my M.S. in Data Science at the University of Chicago.

  4. Volunteered with RIPL at the South Side Science Festival in Chicago.

  5. Joined RIPL at TTIC as a visiting student, working with Prof. Matthew R. Walter.

  6. Graduated from Shenzhen University with a B.S. in Mathematics and Applied Mathematics.

03

Research

front side

IROS 2026 WorkshopCreativity in Assembly

StackGen3D: Stable 3D Block Stack Generation from Multiview Silhouettes

Xiaopeng Zhang*, Luzhe Sun*, Yiqing Xu, Matthew R. Walter · *equal contribution

Given silhouettes from several views and an inventory of blocks, StackGen3D generates 3D block stacks that stand on their own. Block poses come from conditional flow matching; training data is curated with a physics-informed GNN and verified in MuJoCo. A simulator-free, AABB-based inference step raises stability from 45.3% to 72.4%.

04

Tutorial

Open SourceInteractive tutorial · in Chinese

Planning Algorithms 中文互动教程: An Interactive Companion to LaValle's Planning Algorithms

This is a chapter-by-chapter tutorial for Steven M. LaValle's Planning Algorithms. It covers all 15 chapters: discrete planning, configuration spaces, sampling-based motion planning (RRT, PRM), combinatorial and feedback planning, decision theory, information spaces, and planning under differential constraints. Each chapter pairs the explanations with in-browser interactive figures, 200+ in all, plus a runnable Python reference implementation.

LaValle《规划算法》中文互动教程:15 章全覆盖,每章都有可交互图和 Python 参考实现。

05

Education

  1. 2026 – present

    Rutgers University, New Brunswick

    Ph.D. in Computer Science

    Advisor: Prof. Jingjin Yu

  2. 2025 – 2026

    Toyota Technological Institute at Chicago

    Visiting Student, Robot Intelligence through Perception Lab (RIPL)

    Advisor: Prof. Matthew R. Walter

  3. 2024 – 2026

    University of Chicago

    M.S. in Data Science, Data Science Institute

    GPA 3.85 / 4.00

  4. 2020 – 2024

    Shenzhen University

    B.S. in Mathematics and Applied Mathematics, Institute for Advanced Study

    Advisor: Prof. Guojian Yin

    Graduated with honor · Ranked 1st (1/17 in mathematics, 1/57 in IAS) · GPA 4.21 / 4.50

06

Honors & Awards

07

Teaching & Service