![[2602.12656] PMG: Parameterized Motion Generator for Human-like Locomotion Control](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.12656] PMG: Parameterized Motion Generator for Human-like Locomotion Control
Summary
The PMG paper presents a novel Parameterized Motion Generator for humanoid locomotion, addressing challenges in adapting motion tracking to diverse tasks with high-dimensional control commands.
Why It Matters
This research is significant as it proposes a solution to the limitations of existing humanoid locomotion systems, enabling more natural and adaptable movement in robots. It has implications for robotics, AI, and applications in virtual reality, enhancing the interaction between humans and machines.
Key Takeaways
- Introduces the Parameterized Motion Generator (PMG) for humanoid locomotion.
- Addresses challenges in adapting motion tracking to various task contexts.
- Utilizes a compact set of parameterized motion data for real-time control.
- Demonstrates effective sim-to-real transfer and VR-based teleoperation.
- Validates the approach on the humanoid prototype ZERITH Z1.
Computer Science > Robotics arXiv:2602.12656 (cs) [Submitted on 13 Feb 2026] Title:PMG: Parameterized Motion Generator for Human-like Locomotion Control Authors:Chenxi Han, Yuheng Min, Zihao Huang, Ao Hong, Hang Liu, Yi Cheng, Houde Liu View a PDF of the paper titled PMG: Parameterized Motion Generator for Human-like Locomotion Control, by Chenxi Han and 6 other authors View PDF HTML (experimental) Abstract:Recent advances in data-driven reinforcement learning and motion tracking have substantially improved humanoid locomotion, yet critical practical challenges remain. In particular, while low-level motion tracking and trajectory-following controllers are mature, whole-body reference-guided methods are difficult to adapt to higher-level command interfaces and diverse task contexts: they require large, high-quality datasets, are brittle across speed and pose regimes, and are sensitive to robot-specific calibration. To address these limitations, we propose the Parameterized Motion Generator (PMG), a real-time motion generator grounded in an analysis of human motion structure that synthesizes reference trajectories using only a compact set of parameterized motion data together with High-dimensional control commands. Combined with an imitation-learning pipeline and an optimization-based sim-to-real motor parameter identification module, we validate the complete approach on our humanoid prototype ZERITH Z1 and show that, within a single integrated system, PMG produces natural, ...