GAPG: Geometry Aware Push-Grasping Synergy for Goal-Oriented Manipulation in Clutter

TL;DR

This paper introduces a geometry-aware push-grasping framework that combines geometric analysis of point clouds with action evaluation to improve robotic manipulation in cluttered environments, enhancing safety, efficiency, and reliability.

Contribution

The novel framework integrates geometric reasoning into push and grasp evaluation, enabling more effective manipulation in cluttered scenes compared to prior methods.

Findings

Improved grasp stability and success rate in cluttered environments.

Enhanced safety and efficiency through geometry-aware action selection.

Good generalization to real-world scenes and unseen objects.

Abstract

Grasping target objects is a fundamental skill for robotic manipulation, but in cluttered environments with stacked or occluded objects, a single-step grasp is often insufficient. To address this, previous work has introduced pushing as an auxiliary action to create graspable space. However, these methods often struggle with both stability and efficiency because they neglect the scene's geometric information, which is essential for evaluating grasp robustness and ensuring that pushing actions are safe and effective. To this end, we propose a geometry-aware push-grasp synergy framework that leverages point cloud data to integrate grasp and push evaluation. Specifically, the grasp evaluation module analyzes the geometric relationship between the gripper's point cloud and the points enclosed within its closing region to determine grasp feasibility and stability. Guided by this, the push evaluation module predicts how pushing actions influence future graspable space, enabling the robot to select actions that reliably transform non-graspable states into graspable ones. By jointly reasoning about geometry in both grasping and pushing, our framework achieves safer, more efficient, and more reliable manipulation in cluttered settings. Our method is extensively tested in simulation and real-world environments in various scenarios. Experimental results demonstrate that our model generalizes well to real-world scenes and unseen objects.