FALCON: Learning Force-Adaptive Humanoid Loco-Manipulation

TL;DR

FALCON introduces a dual-agent reinforcement learning framework for humanoid robots that enhances force-adaptive loco-manipulation, enabling precise control and robustness in forceful tasks across different robots.

Contribution

The paper presents a novel dual-agent RL approach that decomposes whole-body control into stable locomotion and precise end-effector tracking with adaptive force compensation.

Findings

Achieves 2x more accurate upper-body joint tracking

Maintains robust locomotion under external force disturbances

Enables cross-robot forceful loco-manipulation in real-world tasks

Abstract

Humanoid loco-manipulation holds transformative potential for daily service and industrial tasks, yet achieving precise, robust whole-body control with 3D end-effector force interaction remains a major challenge. Prior approaches are often limited to lightweight tasks or quadrupedal/wheeled platforms. To overcome these limitations, we propose FALCON, a dual-agent reinforcement-learning-based framework for robust force-adaptive humanoid loco-manipulation. FALCON decomposes whole-body control into two specialized agents: (1) a lower-body agent ensuring stable locomotion under external force disturbances, and (2) an upper-body agent precisely tracking end-effector positions with implicit adaptive force compensation. These two agents are jointly trained in simulation with a force curriculum that progressively escalates the magnitude of external force exerted on the end effector while respecting torque limits. Experiments demonstrate that, compared to the baselines, FALCON achieves 2x more accurate upper-body joint tracking, while maintaining robust locomotion under force disturbances and achieving faster training convergence. Moreover, FALCON enables policy training without embodiment-specific reward or curriculum tuning. Using the same training setup, we obtain policies that are deployed across multiple humanoids, enabling forceful loco-manipulation tasks such as transporting payloads (0-20N force), cart-pulling (0-100N), and door-opening (0-40N) in the real world.