Pro-HOI: Perceptive Root-guided Humanoid-Object Interaction

TL;DR

Pro-HOI introduces a robust, generalizable framework for humanoid robot loco-manipulation that leverages root trajectories and real-time perception to enhance reliability and adaptability in complex environments.

Contribution

The paper presents a novel training framework conditioning policies on root trajectories and a persistent object estimation module, advancing humanoid-robot interaction capabilities.

Findings

Outperforms baselines in generalization and robustness

Enables reliable long-horizon execution in real-world scenarios

Integrates real-time detection with Digital Twin for autonomous operation

Abstract

Executing reliable Humanoid-Object Interaction (HOI) tasks for humanoid robots is hindered by the lack of generalized control interfaces and robust closed-loop perception mechanisms. In this work, we introduce Perceptive Root-guided Humanoid-Object Interaction, Pro-HOI, a generalizable framework for robust humanoid loco-manipulation. First, we collect box-carrying motions that are suitable for real-world deployment and optimize penetration artifacts through a Signed Distance Field loss. Second, we propose a novel training framework that conditions the policy on a desired root-trajectory while utilizing reference motion exclusively as a reward. This design not only eliminates the need for intricate reward tuning but also establishes root trajectory as a universal interface for high-level planners, enabling simultaneous navigation and loco-manipulation. Furthermore, to ensure operational reliability, we incorporate a persistent object estimation module. By fusing real-time detection with Digital Twin, this module allows the robot to autonomously detect slippage and trigger re-grasping maneuvers. Empirical validation on a Unitree G1 robot demonstrates that Pro-HOI significantly outperforms baselines in generalization and robustness, achieving reliable long-horizon execution in complex real-world scenarios.