MVGrasp: Real-Time Multi-View 3D Object Grasping in Highly Cluttered Environments

TL;DR

This paper introduces MVGrasp, a multi-view deep learning method enabling real-time, robust 3D object grasping in cluttered environments, effective on both simulated and real-world data without additional fine-tuning.

Contribution

It presents a novel multi-view approach that estimates pixel-wise grasp synthesis from point clouds, trained end-to-end on a small dataset for robust grasping in diverse scenarios.

Findings

High success rate in simulated and real-world tests

Effective grasping across isolated, packed, and cluttered objects

No fine-tuning needed for different scene configurations

Abstract

Nowadays robots play an increasingly important role in our daily life. In human-centered environments, robots often encounter piles of objects, packed items, or isolated objects. Therefore, a robot must be able to grasp and manipulate different objects in various situations to help humans with daily tasks. In this paper, we propose a multi-view deep learning approach to handle robust object grasping in human-centric domains. In particular, our approach takes a point cloud of an arbitrary object as an input, and then, generates orthographic views of the given object. The obtained views are finally used to estimate pixel-wise grasp synthesis for each object. We train the model end-to-end using a small object grasp dataset and test it on both simulations and real-world data without any further fine-tuning. To evaluate the performance of the proposed approach, we performed extensive sets of experiments in three scenarios, including isolated objects, packed items, and pile of objects. Experimental results show that our approach performed very well in all simulation and real-robot scenarios, and is able to achieve reliable closed-loop grasping of novel objects across various scene configurations.