View PDF HTML (experimental)

Abstract:Humanoid table tennis (TT) demands rapid perception, proactive whole-body motion, and agile footwork under strict timing--capabilities that remain difficult for end-to-end control policies. We propose a reinforcement learning (RL) framework that maps ball-position observations directly to whole-body joint commands for both arm striking and leg locomotion, strengthened by predictive signals and dense, physics-guided rewards. A lightweight learned predictor, fed with recent ball positions, estimates future ball states and augments the policy's observations for proactive decision-making. During training, a physics-based predictor supplies precise future states to construct dense, informative rewards that lead to effective exploration. The resulting policy attains strong performance across varied serve ranges (hit rate$\geq$96% and success rate$\geq$92%) in simulations. Ablation studies confirm that both the learned predictor and the predictive reward design are critical for end-to-end learning. Deployed zero-shot on a physical Booster T1 humanoid with 23 revolute joints, the policy produces coordinated lateral and forward-backward footwork with accurate, fast returns, suggesting a practical path toward versatile, competitive humanoid TT. We have open-sourced our RL training code at: this https URL

Submission history

From: Muqun Hu [view email]
[v1] Thu, 25 Sep 2025 23:26:07 UTC (3,749 KB)
[v2] Tue, 21 Oct 2025 17:21:42 UTC (3,750 KB)
[v3] Wed, 18 Mar 2026 17:59:50 UTC (1,990 KB)
[v4] Sat, 21 Mar 2026 03:06:51 UTC (1,990 KB)
[v5] Mon, 3 Aug 2026 22:14:20 UTC (1,990 KB)