ATHLETE: Learning Reactive Humanoid Tennis via Trajectory-Guided Motion Matching
ATHLETE: Learning Reactive Humanoid Tennis via Trajectory-Guided Motion Matching
ATHLETE is a framework for learning reactive humanoid tennis from generic locomotion data and ordinary tennis videos. It combines LAFAN1 locomotion with video-derived striking motions using Trajectory-Guided Motion Matching (TGMM) to construct diverse whole-body references. These are refined through physics-based teacher-student learning and PPO fine-tuning, achieving 90.00% and 86.67% hit rates for forehand and backhand shots, respectively, and enabling multi-shot rallies with human players.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- code: Available Artifact link Open source S5
- paper: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
istillation, followed by PPO fine-tuning for the tennis task. The resulting policy responds directly to robot and ball observations without requiring reference-motion inputs at deployment. On a real humanoid robot, ATHLETE achieves forehand and backhand hit rates of $90.00\%$ and $86.67\%$, respectively, and enables multi-shot rallies with human players. These results demonstrate that ATHLETE can learn dynamic humano
Open source S4
id tennis skills from reusable locomotion data and video demonstrations. Result videos and code can be found at https://anonymous.4open.science/w/ATHLETE2027/.
Open source S5
arXiv:2610.13572v1 Announce Type: new Abstract: Humanoid tennis requires reactive whole-body coordination under rapidly changing ball dynamics, yet collecting task-specific motion-capture demonstrations that cover diverse approach states, contact locations, and timing conditions is costly. Inspired by how humans acquire new skills by combining reusable motor experience with task-specific demonstrations, we present AT
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org