AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimation

2026-02-16

AI Summary

This work addresses the challenge of achieving robust autonomous whole-body manipulation for humanoid robots in scenarios lacking prior human demonstrations, limited field of view, and occlusions. The authors propose an end-to-end adaptive policy based on reinforcement learning that integrates online recurrent state estimation, LiDAR-based localization, and residual operational-space control. Notably, this approach enables zero-shot sim-to-real transfer—without relying on imitation learning data—to accomplish fully autonomous navigation, grasping, and object delivery as a unified task. Experimental results demonstrate that the system significantly outperforms imitation learning–based baselines in real-world environments, maintaining high success rates and manipulation accuracy even under substantial occlusion.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This paper presents Adaptive Whole-body Loco-Manipulation, AdaptManip, a fully autonomous framework for humanoid robots to perform integrated navigation, object lifting, and delivery. Unlike prior imitation learning-based approaches that rely on human demonstrations and are often brittle to disturbances, AdaptManip aims to train a robust loco-manipulation policy via reinforcement learning without human demonstrations or teleoperation data. The proposed framework consists of three coupled components: (1) a recurrent object state estimator that tracks the manipulated object in real time under limited field-of-view and occlusions; (2) a whole-body base policy for robust locomotion with residual manipulation control for stable object lifting and delivery; and (3) a LiDAR-based robot global position estimator that provides drift-robust localization. All components are trained in simulation using reinforcement learning and deployed on real hardware in a zero-shot manner. Experimental results show that AdaptManip significantly outperforms baseline methods, including imitation learning-based approaches, in adaptability and overall success rate, while accurate object state estimation improves manipulation performance even under occlusion. We further demonstrate fully autonomous real-world navigation, object lifting, and delivery on a humanoid robot.

Problem

• humanoid robot
• whole-body manipulation
• object lifting
• autonomous navigation
• occlusion

Innovation

• reinforcement learning
• whole-body locomotion
• online state estimation
• humanoid robot
• zero-shot transfer