analiza aiqbench
This paper addresses a critical gap in humanoid robotics by introducing ConstrainedMimic, a framework that enforces safety constraints like collision avoidance and joint limits within reinforcement learning policies without sacrificing agility. By integrating operational space control and control barrier functions, the method achieves real-time constraint satisfaction above 1 kHz, bridging the divide between high-performance RL agents and safe, deployable hardware systems.
Constrained Whole-Body Tracking for Humanoid Robots
arXiv:2606.00374v2 Announce Type: replace Abstract: Recent advances in reinforcement learning (RL) have demonstrated impressive whole-body agility for humanoid robots, yet ensuring safety and satisfying constraints -- particularly those specified after training -- remains a challenge. Towards this goal, we present ConstrainedMimic, a control framework that leverages whole-body kinematics and dynamics for real-time constraint enforcement within RL tracking policies. By integrating principles from operational space control and control barrier functions, we enable the satisfaction of arbitrary runtime constraints on both the kinematic reference motion and the underlying dynamics, with efficient computation at frequencies above 1 kHz. In whole-body motion-tracking experiments in simulation and hardware, we demonstrate collision avoidance (both with the robot body and external obstacles), joint limits, and center of mass stability constraints. By remaining consistent with the current contact mode and tracking objectives, we minimally restrict the capabilities of the policy when constraints are active. Project page: https://stanfordasl.github.io/constrainedmimic.
Abstract:Recent advances in reinforcement learning (RL) have demonstrated impressive whole-body agility for humanoid robots, yet ensuring safety and satisfying constraints -- particularly those specified after training -- remains a challenge. Towards this goal, we present ConstrainedMimic, a control framework that leverages whole-body kinematics and dynamics for real-time constraint enforcement within RL tracking policies. By integrating principles from operational space control and control barrier functions, we enable the satisfaction of arbitrary runtime constraints on both the kinematic reference motion and the underlying dynamics, with efficient computation at frequencies above 1 kHz. In whole-body motion-tracking experiments in simulation and hardware, we demonstrate collision avoidance (both with the robot body and external obstacles), joint limits, and center of mass stability constraints. By remaining consistent with the current contact mode and tracking objectives, we minimally restrict the capabilities of the policy when constraints are active. Project page: this https URL.
Robotics (cs.RO)
arXiv:2606.00374 [cs.RO]
(or arXiv:2606.00374v2 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.00374 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 237 cuvinte
Subiecte conexe
Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking
arXiv · 5 oct. 2026
Extreme-RGMT: Continual Learning of Highly Dynamic Skills for Robust Generalist Humanoid Control
arXiv · 23 iul. 2026
ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation
arXiv · 21 sept. 2026
LadderMan: Learning Humanoid Perceptive Ladder Climbing
arXiv · 6 iun. 2026