{"ID":2866276,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.21690","arxiv_id":"2509.21690","title":"PACE: Physics Augmentation for Coordinated End-to-end Reinforcement Learning toward Versatile Humanoid Table Tennis","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: https://github.com/purdue-tracelab/TTRL-ICRA2026","short_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...","url_abs":"https://arxiv.org/abs/2509.21690","url_pdf":"https://arxiv.org/pdf/2509.21690v4","authors":"[\"Muqun Hu\",\"Wenxi Chen\",\"Wenjing Li\",\"Falak Mandali\",\"Zijian He\",\"Renhong Zhang\",\"Praveen Krisna\",\"Katherine Christian\",\"Leo Benaharon\",\"Dizhi Ma\",\"Karthik Ramani\",\"Yan Gu\"]","published":"2025-09-25T23:26:07Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[\"Reinforcement Learning\",\"LoRA\"]","has_code":false,"code_links":[{"ID":609358,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2866276,"paper_url":"https://arxiv.org/abs/2509.21690","paper_title":"PACE: Physics Augmentation for Coordinated End-to-end Reinforcement Learning toward Versatile Humanoid Table Tennis","repo_url":"https://github.com/purdue-tracelab/TTRL-ICRA2026","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}