Short Bio

🐣 Born in 2000, I grew up alongside the rise of intelligent machines. I received my B.Eng. degree from the School of Vehicle and Mobility, Tsinghua University, Beijing, China , in 2021. I am currently pursuing a Ph.D. in Mechanical Engineering at the same school. From 2025 to 2026, I am a visiting student scholar at the Department of Mechanical Engineering, UC Berkeley, California, United States . My research focuses on ⚙️ control theory, 🧠 reinforcement learning, and their applications in 🚗 autonomous vehicles and 🤖 robotics.

👋 Welcome to @molumitu’s Personal Page

GPA & Awards

🎓 I received my B.Eng. degree from Tsinghua University with a GPA of 3.78/4.0, and I am currently pursuing my Ph.D. with a graduate GPA of 3.93/4.0.
🏅 I have been recognized with the Comprehensive Excellence Award and named an Outstanding Graduate of Tsinghua University (Top 10\%).

🤝 Let’s Collaborate!

💡 I’m always open to research collaborations, particularly in areas such as Reinforcement Learning and Embodied AI. If you’re exploring innovative ideas, feel free to reach out — I’d love to hear from you and explore potential synergies!

📩 Contact Me:

• Email: zgj21@mails.tsinghua.edu.cn
• Email: zgj21@berkeley.edu
• Email: zishangzhan@gmail.com

🚀 Looking forward to collaborating with you!

Publications

Learn a Safe and Efficient Model Predictive Controller with Lagrange Multiplier Neural Network

Harmonic Constrained Reinforcement Learning

Bootstrap Off-policy with World Model

Continuous-Time Policy Gradient

Bicriteria Policy Optimization for High Accuracy Reinforcement Learning

Enhanced DACER Algorithm with High Diffusion Efficiency

Physics Informed Neural Pose Estimation for Real-Time Shape Reconstruction of Soft Continuum Robots

Distributional Soft Actor-Critic with Harmonic Gradient for Safe and Efficient Autonomous Driving in Multi-lane Scenarios

Understanding Connection between PMP and HJB Equations from the Perspective of Hamilton Dynamics

Transferable Latent-to-Latent Locomotion Policy for Efficient and Versatile Motion Control of Diverse Legged Robots

PODAR: A Collision Risk Model Offering Valid Signals for Vehicular Interactions

Predictive Lagrangian Optimization for Constrained Reinforcement Learning

Zeroth-Order Actor-Critic: An Evolutionary Framework for Sequential Decision Problems

An Explicit Discrete-Time Dynamic Vehicle Model with Assured Numerical Stability

Rocket Landing Control with Random Annealing Jump Start Reinforcement Learning

CIDC: A Flexible Formation Control Framework for Intelligent Connected Vehicles

Quantifying the Individual Differences of Drivers’ Risk Perception via Potential Damage Risk Model

Canonical Form of Datatic Description in Control Systems

A Transformation-Aggregation Framework for State Representation of Autonomous Driving Systems

Continuous-Time Policy Optimization

Data-Driven Safe Learning and Control with Safety Certificates

A Reinforcement Learning Benchmark for Autonomous Driving in General Urban Scenarios

Enhance Generality by Model-based Reinforcement Learning and Domain Randomization

Bridging the Gap between Newton-Raphson Method and Regularized Policy Iteration

Improve Generalization of Driving Policy at Signalized Intersections with Adversarial Learning

Languagempc: Large Language Models as Decision Makers for Autonomous Driving

Self-learned Intelligence for Integrated Decision and Control of Automated Vehicles at Signalized Intersections