Teaching robots athletic moves through realistic muscle simulation

Cheryl Wang, Chun Kwang Tan, Balint K. Hodossy, Eric Lyu, Jun Guo, Wentao Zhao, Huaping Liu, Chengkun Li, Merkourios Simos, Bianca Ziliotto, Alexander Mathis, Siyuan Liu, Jiahao Chen, Shanlin Zhong, Bo Jiang, Ci Song, Yaoye Zhu, Chenhui Zuo, Yanan Sui, Mohamed Irfan Refai, Massimo Sartori, Guillaume Durandau, Vikash Kumar, Vittorio Caggiano

MyoChallenge 2025 establishes a benchmark for motor control in sports by combining high-fidelity musculoskeletal models with machine learning. Two competition tracks test control of a biomechanical arm for table tennis and biomechanical legs for soccer penalty kicks. The competition attracted nearly 70 teams with over 560 submissions, yielding new control algorithms using physics-based planning, imitation learning, and hierarchical methods. Results are integrated into the open-source MyoSuite framework, enabling reproducible research across machine learning, biomechanics, and neuroscience.