Abstract:We propose an active learning-based grasp method for accurate industrial manipulation that combines the high accuracy of geometrically driven grasp methods and the generalization ability of data-driven grasp methods. Our grasp sequence consists of pregrasp stage and grasp stage which integrates the active perception and manipulation. In pregrasp stage, the manipulator actively moves and perceives the object. At each step, given the perception image, a motion is chosen so that the manipulator can adjust to a proper pose to grasp the object. We train a convolutional neural network to estimate the motion and combine the network with a closed-loop control so that the end effector can move to the pregrasp state. In grasp stage, the manipulator executes a fixed motion to complete the grasp task. The fixed motion can be acquired from the demonstration with nonexpert conveniently. Our proposed method does not require the prior knowledge of camera intrinsic parameters, hand-eye transformation, or manually designed feature of objects. Instead, the training data sets containing prior knowledge are collected through interactive perception. The method can be easily transferred to new tasks with a few human interventions and is able to complete high accuracy grasp task with a certain robustness to partial observation condition. In our circuit board grasping tests, we could achieve a grasp accuracy of 0.8 mm and 0.6°. Note to Practitioners—The research in this paper is motivated by the following practical problem. Manipulators on industrial lines can complete high accuracy tasks with hand-crafting features of objects. The perception is only used for object detection and localization. It is not flexible since the prior knowledge differs from tasks, which takes a long time to deploy in a new task. Besides, only well-trained experts are qualified to complete the deployment process. Our grasp method uses a convolutional network to estimate the motion for manipulator directly from images. The camera is mounted on the manipulator and can perceive the object actively. The training data set of the network is specific for different objects that can be automatically collected with a few human interventions. Our method simplifies the deployment process and can be applied in 3C industry (computers, communications, and consumer electronics) where the products upgrade frequently.

A Novel Robotic Grasp Detection Framework Using Low-Cost RGB-D Camera for Industrial Bin Picking

LiteGrasp: A Light Robotic Grasp Detection Via Semi-Supervised Knowledge Distillation

Active Learning-Based Grasp for Accurate Industrial Manipulation.

A Robotic Semantic Grasping Method for Pick-and-place Tasks

OptiGrasp: Optimized Grasp Pose Detection Using RGB Images for Warehouse Picking Robots

Efficient End-to-End Detection of 6-DoF Grasps for Robotic Bin Picking

A Novel RGB-D Cross-Background Robot Grasp Detection Dataset and Background-Adaptive Grasping Network

Vision-Based Robotic Object Grasping—A Deep Reinforcement Learning Approach

Robotic picking in dense clutter via domain invariant learning from synthetic dense cluttered rendering

RGBGrasp: Image-based Object Grasping by Capturing Multiple Views during Robot Arm Movement with Neural Radiance Fields

Object Detection and Pose Estimation from RGB and Depth Data for Real-time, Adaptive Robotic Grasping

Robotic Grasping With Multi-View Image Acquisition and Model-Based Pose Estimation

A Practical Robotic Grasping Method by Using 6-D Pose Estimation With Protective Correction

Fast and Robust Bin-picking System for Densely Piled Industrial Objects

6D Pose Estimation of Industrial Parts Based on Point Cloud Geometric Information Prediction for Robotic Grasping

Deep learning for detecting robotic grasps

Bin-picking of novel objects through category-agnostic-segmentation: RGB matters

Robotic pick-and-place of novel objects in clutter with multi-affordance grasping and cross-domain image matching

POLLINATION STUDY IN A NATURAL POPULATION OF MIMULUS GUTTATUS

Efficient Fully Convolutional Network and Optimization Approach for Robotic Grasping Detection Based on RGB-D Images

A Sim-to-Real Object Recognition and Localization Framework for Industrial Robotic Bin Picking