• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input.
Zifan
Xu, Myoungkyu Seo, Dongmyeong Lee, Hao Fu, Jiaheng
Hu, Jiaxun Cui, Yuqian Jiang, Zhihan Wang,
Anastasiia Brund, Joydeep Biswas, and Peter
Stone.
In International Conference on Robotics and Automation (ICRA), June 2026.
Learning fast and robust ball-kicking skills is a critical capability forhumanoid soccer robots, yet it remains a challenging problem due to the need forrapid leg swings, postural stability on a single support foot, and robustnessunder noisy sensory input and external perturbations (e.g., opponents). Thispaper presents a reinforcement learning (RL)–based training pipeline that enableshumanoid robots to execute robust continual ball-kicking with adaptability todifferent ball-goal configurations. The pipeline extends a typicalteacher-student training framework---in which a ``teacher" policy is trained withground truth state information and the ``student" learns to mimic it with noisy,imperfect sensing---by including four training stages: (1) long-distance ballchasing (teacher); (2) directional kicking (teacher); (3) teacher policydistillation (student), and (4) student adaptation and refinement (student). Keydesign elements---including tailored reward functions, realistic noise modeling,and online constrained RL for adaptation and refinement---are critical forclosing the sim-to-real gap and sustaining performance under perceptualuncertainty. Extensive evaluations in both simulation and on a real robotdemonstrate strong kicking accuracy and goal-scoring success across diverseball–goal configurations. Ablation studies further highlight the necessity of theconstrained RL, noise modeling, and the adaptation stage. This work presents atraining pipeline for robust continual humanoid ball-kicking under imperfectperception, establishing a benchmark task for visuomotor skill learning inhumanoid whole-body control.
@InProceedings{zifan_xu_icra_2026,
author = {Zifan Xu and Myoungkyu Seo and Dongmyeong Lee and Hao Fu and Jiaheng Hu and Jiaxun Cui and Yuqian Jiang and Zhihan Wang and Anastasiia Brund and Joydeep Biswas and Peter Stone},
title = {Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input},
booktitle = {International Conference on Robotics and Automation (ICRA)},
year = {2026},
month = {June},
location = {Vienna, Austria},
abstract = {Learning fast and robust ball-kicking skills is a critical capability for
humanoid soccer robots, yet it remains a challenging problem due to the need for
rapid leg swings, postural stability on a single support foot, and robustness
under noisy sensory input and external perturbations (e.g., opponents). This
paper presents a reinforcement learning (RL)–based training pipeline that enables
humanoid robots to execute robust continual ball-kicking with adaptability to
different ball-goal configurations. The pipeline extends a typical
teacher-student training framework---in which a ``teacher" policy is trained with
ground truth state information and the ``student" learns to mimic it with noisy,
imperfect sensing---by including four training stages: (1) long-distance ball
chasing (teacher); (2) directional kicking (teacher); (3) teacher policy
distillation (student), and (4) student adaptation and refinement (student). Key
design elements---including tailored reward functions, realistic noise modeling,
and online constrained RL for adaptation and refinement---are critical for
closing the sim-to-real gap and sustaining performance under perceptual
uncertainty. Extensive evaluations in both simulation and on a real robot
demonstrate strong kicking accuracy and goal-scoring success across diverse
ball–goal configurations. Ablation studies further highlight the necessity of the
constrained RL, noise modeling, and the adaptation stage. This work presents a
training pipeline for robust continual humanoid ball-kicking under imperfect
perception, establishing a benchmark task for visuomotor skill learning in
humanoid whole-body control.
},
}
Generated by bib2html.pl (written by Patrick Riley ) on Mon May 18, 2026 15:31:22