Mengze Tian

I am an MSc student in Mechanical Engineering at EPFL, advised by Sylvain Calinon and Auke Ijspeert. My research focuses on robot learning across motion planning, manipulation, and locomotion. Previously, I received my B.E. in Automation from Zhejiang University, where I worked with Fei Gao.

Research

Hierarchically Depicting Vehicle Trajectory with Stability in Complex Environments

Zhichao Han†, Mengze Tian† (co-first), Fei Gao*, et al. · Science Robotics, 2025

We present a hierarchical approach that integrates the spatial extraction capabilities of neural networks with the robust convergence of numerical optimization, enabling efficient and stable trajectory generation in diverse environments.

Paper Code

Spline Policy: A Structured Representation for Robot Policies

Mengze Tian†, Yiming Li†, Sichao Liu, Auke Ijspeert, Sylvain Calinon · (under review)

We propose Spline Policy, a structured policy representation that models robot behaviors as spline trajectories rather than discrete action chunks, supporting flexible temporal evaluation, parameter-space constraints, and uncertainty propagation.

Paper

Learning Whole-Body Control for a Salamander Robot

Mengze Tian†, Qiyuan Fu†, Chuanfang Ning, Javier Jia Jie Pey, Auke Ijspeert · (under review)

We develop a reinforcement learning framework for coordinated axial–limb control on a highly articulated salamander robot, achieving stable sim-to-real transfer across flat and uneven terrains and hybrid terrestrial–aquatic locomotion.

Paper Video