Real-time Grasp Pose Estimation for Novel Objects in Densely Cluttered Environment

TL;DR

This paper introduces a real-time grasp pose estimation method for novel, complex-shaped objects in cluttered environments, improving grasp success rates over traditional centroid and major axis strategies.

Contribution

The paper presents a novel real-time technique that estimates object contours and skeletons to predict grasp poses, outperforming traditional methods for complex objects.

Findings

Achieved 88.16% grasp accuracy in distinct object placement scenarios.

Achieved 77.03% grasp accuracy in dense clutter scenarios.

Validated with a UR10 robot and WSG-50 gripper.

Abstract

Grasping of novel objects in pick and place applications is a fundamental and challenging problem in robotics, specifically for complex-shaped objects. It is observed that the well-known strategies like \textit{i}) grasping from the centroid of object and \textit{ii}) grasping along the major axis of the object often fails for complex-shaped objects. In this paper, a real-time grasp pose estimation strategy for novel objects in robotic pick and place applications is proposed. The proposed technique estimates the object contour in the point cloud and predicts the grasp pose along with the object skeleton in the image plane. The technique is tested for the objects like ball container, hand weight, tennis ball and even for complex shape objects like blower (non-convex shape). It is observed that the proposed strategy performs very well for complex shaped objects and predicts the valid grasp configurations in comparison with the above strategies. The experimental validation of the proposed grasping technique is tested in two scenarios, when the objects are placed distinctly and when the objects are placed in dense clutter. A grasp accuracy of 88.16\% and 77.03\% respectively are reported. All the experiments are performed with a real UR10 robot manipulator along with WSG-50 two-finger gripper for grasping of objects.