A Deep Learning-Based Autonomous RobotManipulator for Sorting Application

TL;DR

This paper presents a deep learning-based autonomous robotic system for object sorting that integrates object detection, pose estimation, grasp planning, and collision-free motion planning, demonstrating effective real-world performance.

Contribution

It introduces an integrated deep learning approach for autonomous object sorting using RGB-D data, combining detection, grasp planning, and motion planning on a real robot.

Findings

Robust object detection and grasping with CNNs.

Fast and collision-free sorting trajectories.

Successful implementation on AUBO robotic manipulator.

Abstract

Robot manipulation and grasping mechanisms have received considerable attention in the recent past, leading to the development of wide range of industrial applications. This paper proposes the development of an autonomous robotic grasping system for object sorting application. RGB-D data is used by the robot for performing object detection, pose estimation, trajectory generation, and object sorting tasks. The proposed approach can also handle grasping certain objects chosen by users. Trained convolutional neural networks are used to perform object detection and determine the corresponding point cloud cluster of the object to be grasped. From the selected point cloud data, a grasp generator algorithm outputs potential grasps. A grasp filter then scores these potential grasps, and the highest-scored grasp is chosen to execute on a real robot. A motion planner generates collision-free trajectories to execute the chosen grasp. The experiments on AUBO robotic manipulator show the potentials of the proposed approach in the context of autonomous object sorting with robust and fast sorting performance.