PILOT: A Perceptive Integrated Low-level Controller for Loco-manipulation over Unstructured Scenes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cui, Xinru, Feng, Linxi, Zhou, Yixuan, Han, Haoqi, Liu, Zhe, Wang, Hesheng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911396537565184
author Cui, Xinru
Feng, Linxi
Zhou, Yixuan
Han, Haoqi
Liu, Zhe
Wang, Hesheng
author_facet Cui, Xinru
Feng, Linxi
Zhou, Yixuan
Han, Haoqi
Liu, Zhe
Wang, Hesheng
contents Humanoid robots hold great potential for diverse interactions and daily service tasks within human-centered environments, necessitating controllers that seamlessly integrate precise locomotion with dexterous manipulation. However, most existing whole-body controllers lack exteroceptive awareness of the surrounding environment, rendering them insufficient for stable task execution in complex, unstructured scenarios.To address this challenge, we propose PILOT, a unified single-stage reinforcement learning (RL) framework tailored for perceptive loco-manipulation, which synergizes perceptive locomotion and expansive whole-body control within a single policy. To enhance terrain awareness and ensure precise foot placement, we design a cross-modal context encoder that fuses prediction-based proprioceptive features with attention-based perceptive representations. Furthermore, we introduce a Mixture-of-Experts (MoE) policy architecture to coordinate diverse motor skills, facilitating better specialization across distinct motion patterns. Extensive experiments in both simulation and on the physical Unitree G1 humanoid robot validate the efficacy of our framework. PILOT demonstrates superior stability, command tracking precision, and terrain traversability compared to existing baselines. These results highlight its potential to serve as a robust, foundational low-level controller for loco-manipulation in unstructured scenes.
format Preprint
id arxiv_https___arxiv_org_abs_2601_17440
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PILOT: A Perceptive Integrated Low-level Controller for Loco-manipulation over Unstructured Scenes
Cui, Xinru
Feng, Linxi
Zhou, Yixuan
Han, Haoqi
Liu, Zhe
Wang, Hesheng
Robotics
Humanoid robots hold great potential for diverse interactions and daily service tasks within human-centered environments, necessitating controllers that seamlessly integrate precise locomotion with dexterous manipulation. However, most existing whole-body controllers lack exteroceptive awareness of the surrounding environment, rendering them insufficient for stable task execution in complex, unstructured scenarios.To address this challenge, we propose PILOT, a unified single-stage reinforcement learning (RL) framework tailored for perceptive loco-manipulation, which synergizes perceptive locomotion and expansive whole-body control within a single policy. To enhance terrain awareness and ensure precise foot placement, we design a cross-modal context encoder that fuses prediction-based proprioceptive features with attention-based perceptive representations. Furthermore, we introduce a Mixture-of-Experts (MoE) policy architecture to coordinate diverse motor skills, facilitating better specialization across distinct motion patterns. Extensive experiments in both simulation and on the physical Unitree G1 humanoid robot validate the efficacy of our framework. PILOT demonstrates superior stability, command tracking precision, and terrain traversability compared to existing baselines. These results highlight its potential to serve as a robust, foundational low-level controller for loco-manipulation in unstructured scenes.
title PILOT: A Perceptive Integrated Low-level Controller for Loco-manipulation over Unstructured Scenes
topic Robotics
url https://arxiv.org/abs/2601.17440