Ji Ma/ 马骥

I am a third-year Ph.D. student at ArcLab, The University of Hong Kong, advised by Prof. Peng LU. Previously, I received my B.Eng. degree in Information Engineering from Jilin University (2019-2023).

My research focuses on robust robot intelligence, especially at the intersection of reinforcement learning, legged locomotion, and motion planning.

Email · Github · Scholar · Zhihu · LinkedIn

Selected Work

Featured Research

IEEE T-FR

2026 · IEEE Transactions on Field Robotics

MASH: MAstering Safe and High-Speed Quadruped Planning and Tracking with Adaptive Gait Transitions

Ji Ma*, Yinzhao Dong*, Wanyue Li, Peng Lu

Video Project Page

Advanced Science

2026 · Advanced Science (Adv. Sci.)

MGDP: Mastering a Generalized Depth Perception Model for Quadruped Locomotion

Yinzhao Dong*, Ji Ma*, Yidan Lu, Jiahui Zhang, Wanyue Li, Yeke Chen, Xuechao Chen, Zhangguo Yu, Wanyue Li, Peng Lu

JCR Q1 · Impact Factor 14.1

Paper Video Project Page Code

IEEE T-RO

2025 · IEEE Transactions on Robotics (T-RO)

MARG: MAstering Risky Gap Terrains for Legged Robots with Elevation Mapping

Yinzhao Dong*, Ji Ma*, Liu Zhao, Wanyue Li, Peng Lu

JCR Q1 · Impact Factor 10.5

Paper Video Project Page

IROS

2025 · IEEE/RSJ International Conference on Intelligent Robots and Systems

Like Playing a Video Game: Spatial-Temporal Optimization of Foot Trajectories for Controlled Football Kicking in Bipedal Robots

Wanyue Li*, Ji Ma*, Minghao Lu, Peng Lu

Best RoboCup Paper Award

Paper Video Project Page

IEEE RA-L

2025 · IEEE Robotics and Automation Letters (RA-L)

Learning Autonomous and Safe Quadruped Traversal of Complex Terrains Using Multi-Layer Elevation Maps

Yeke Chen, Ji Ma, Zeren Luo, Yimin Han, Yinzhao Dong, Bowen Xu, Peng Lu

Featured in IEEE Spectrum

Paper Video IEEE Spectrum

Latest Archive

All Publications

2026 · arXiv

Co-jump: Cooperative Jumping with Quadrupedal Robots via Multi-Agent Reinforcement Learning

A multi-agent reinforcement learning framework for communication-free cooperative jumping that pushes quadrupeds beyond their solo limits.

Open arXiv PDF

2026 · arXiv

Learning Human-Like Badminton Skills for Humanoid Robots

A progressive imitation-to-interaction pipeline for agile, human-like badminton striking with zero-shot transfer to humanoid hardware.

Open arXiv PDF

Recognition

Academic Honors

2025

Best RoboCup Paper Award
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
2024

ICRA Quadruped Robot Challenges
2nd Place
2022

ICRA RoboMaster University AI Challenge (RMUA)
3rd Place
2022

ICRA RoboMaster University Sim2Real Challenge (RMUS)
2nd Place
2022

Robomaster AI Challenge Open Source Outstanding Prize

Community

Academic Service

Journal

Reviewer
IEEE Robotics and Automation Letters (RA-L)
Conf.

Reviewer
IROS / ICRA