I am a Ph.D. candidate in Computer Science and Technology at the National University of Defense Technology (NUDT). Currently, I am a research assistantship at the Peking University - Psibot Joint Lab, working under the supervision of Prof. Yaodong Yang and Prof. Yuanpei Chen. My research focuses on Embodied AI and Robotics, encompassing systemic studies from high-level behavior planning (e.g., Task and Motion Planning (TAMP), behavior tree planning and generation) to low-level motion control (e.g., Vision-Language-Action Models (VLA) and real-world reinforcement learning).

News

  • 2025.01: Welcome to my homepage!

Publications

AAAI 2025
MRBTP

MRBTP: Efficient Multi-Robot Behavior Tree Planning and Collaboration

Yishuai Cai, Xinglin Chen, Zhongxuan Cai, Yunxin Mao, Minglong Li, Wenjing Yang, Ji Wang

AAAI 2025 Oral | Citations: 4

  • Proposes a multi-robot behavior tree planning algorithm with theoretical guarantees of soundness and completeness, achieving efficient collaboration through cross-tree expansion and intention sharing.
IJCAI 2024
Intent Understanding

Integrating Intent Understanding and Optimal Behavior Planning for Behavior Tree Generation from Human Instructions

Xinglin Chen*, Yishuai Cai*, Yunxin Mao, Minglong Li, Wenjing Yang, Weixia Xu, Ji Wang

IJCAI 2024 | Citations: 19

  • Proposes a two-stage framework for behavior tree generation that combines large language models with optimal behavior tree expansion to generate reliable behavior trees from human instructions.
NeurIPS 2025
DexFlyWheel

DexFlyWheel: A Scalable and Self-improving Data Generation Framework for Dexterous Manipulation

Kefei Zhu, Fengshuo Bai, YuanHao Xiang, Yishuai Cai, Xinglin Chen, Ruochong Li, Xingtao Wang, Hao Dong, Yaodong Yang, Xiaopeng Fan, Yuanpei Chen

NeurIPS 2025 Spotlight (Top 3.2%) | Citations: -

  • Proposes a scalable and self-improving data generation framework for efficiently generating high-quality dexterous manipulation data.
ICRA 2025
HBTP

HBTP: Heuristic Behavior Tree Planning with Large Language Model Reasoning

Yishuai Cai, Xinglin Chen, Yunxin Mao, Minglong Li, Shaowu Yang, Wenjing Yang, Ji Wang

ICRA 2025 | Citations: 2

  • Proposes a heuristic behavior tree planning framework that leverages large language model reasoning to generate heuristic paths, improving planning efficiency.
IROS 2023
Task2Morph

Task2Morph: Differentiable Task-inspired Framework for Contact-Aware Robot Design

Yishuai Cai, Shaowu Yang, Minglong Li, Xinglin Chen, Yunxin Mao, Xiaodong Yi, Wenjing Yang

IROS 2023 | Citations: 4

IROS 2023
Evolving Physical Instinct

Evolving Physical Instinct for Morphology and Control Co-Adaption

Xinglin Chen, Da Huang, Minglong Li, Yishuai Cai, Zhuoer Wen, Zhongxuan Cai, Wenjing Yang

IROS 2023 | Citations: 5

Honors and Awards

  • National Scholarship (Ph.D.), National Scholarship (Undergraduate), Outstanding Graduate of Zhejiang Province, Zhejiang Provincial Government Scholarship, Outstanding Student of National University of Defense Technology
  • 2023: First Prize, China Software Conference - Robot Large Model and Embodied Intelligence Challenge (National)
  • 2022: Special Prize, 6th Military Mathematical Modeling Competition (Team Leader)
  • 2020: First Prize, Mathematical Contest in Modeling (MCM/ICM)
  • 2019: First Prize, 9th Asia-Pacific Mathematical Contest in Modeling
  • 2019: First Prize, 10th China College Students Outsourcing Innovation and Entrepreneurship Competition (National)
  • 2019: Full Score in Programming Ability Test (PAT) Level A (1/1085)
  • 2018: Third Prize, China Collegiate Programming Contest (CCPC) (Team Leader)
  • 2018: Third Prize, Zhejiang Province ACM Collegiate Programming Contest

Professional Skills

  • Strong computer programming skills with training and award experience in CCPC and ACM collegiate programming competitions.
  • Published excellent papers at top-tier robotics and AI conferences including ICRA, IROS, AAAI, IJCAI, and NeurIPS.
  • Proficient in Linux and ROS operating systems, with good code standards and strong engineering implementation capabilities.
  • Familiar with simulation platforms such as IsaacGym/Sim and MuJoCo, with experience in Sim-to-Real transfer for reinforcement learning on real robots.
  • Strong problem-solving and self-learning abilities, able to quickly adapt to new technologies and tools. Excellent communication and collaboration skills, capable of working efficiently with development and operations teams. Strong presentation and organizational skills.

Education

  • In Progress, Ph.D. in Computer Science and Technology, National University of Defense Technology
  • Visiting, Peking University - Psibot Joint Laboratory, Advisor: Prof. Yaodong Yang