DockAnywhere: Data-Efficient Visuomotor Policy Learning for Mobile Manipulation via Novel Demonstration Generation

2026-04-16

AI Summary

This work addresses the limited viewpoint generalization of visual policies in mobile manipulation caused by varying docking positions. To overcome this, the authors propose DockAnywhere, a framework that decouples base motion from viewpoint-invariant contact-based manipulation skills. From a single demonstration, the method automatically generates diverse feasible docking configurations and their corresponding trajectories. It leverages structure-preserving data augmentation and point-cloud-based 3D visual synthesis to expand training viewpoints while maintaining geometric consistency of the task. This approach achieves, for the first time, automatic generalization from a single demonstration to arbitrary feasible docking poses. Evaluated on both the ManiSkill simulation benchmark and real robotic platforms, the method significantly improves policy success rates, effectively generalizes to unseen docking viewpoints, and substantially enhances robustness for real-world deployment.

Technology Category

• Intelligent Robots: Manipulation
• Computer Vision: Motion & Tracking
• Machine Learning: Learning with Manifolds

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Mobile manipulation is a fundamental capability that enables robots to interact in expansive environments such as homes and factories. Most existing approaches follow a two-stage paradigm, where the robot first navigates to a docking point and then performs fixed-base manipulation using powerful visuomotor policies. However, real-world mobile manipulation often suffers from the view generalization problem due to shifts of docking points. To address this issue, we propose a novel low-cost demonstration generation framework named DockAnywhere, which improves viewpoint generalization under docking variability by lifting a single demonstration to diverse feasible docking configurations. Specifically, DockAnywhere lifts a trajectory to any feasible docking points by decoupling docking-dependent base motions from contact-rich manipulation skills that remain invariant across viewpoints. Feasible docking proposals are sampled under feasibility constraints, and corresponding trajectories are generated via structure-preserving augmentation. Visual observations are synthesized in 3D space by representing the robot and objects as point clouds and applying point-level spatial editing to ensure the consistency of observation and action across viewpoints. Extensive experiments on ManiSkill and real-world platforms demonstrate that DockAnywhere substantially improves policy success rates and easily generalizes to novel viewpoints from unseen docking points during training, significantly enhancing the generalization capability of mobile manipulation policy in real-world deployment.

Problem

• mobile manipulation
• view generalization
• docking variability
• visuomotor policy
• real-world deployment

Innovation

• data-efficient learning
• viewpoint generalization
• demonstration generation
• mobile manipulation
• visuomotor policy