Deep Reinforcement Learning for Robotic Pushing and Picking in Cluttered Environment

Yuhong Deng, Xiaofeng Guo, Yixuan Wei, Kai Lu, Bin Fang +3 more
2/21/2023
cs.ROcs.CVcs.LG

No official code found. Best implementation: weiyx16/Active-Perception (19 stars, MIT, last commit 2020-06-02).

Abstract

In this paper, a novel robotic grasping system is established to automatically pick up objects in cluttered scenes. A composite robotic hand composed of a suction cup and a gripper is designed for grasping the object stably. The suction cup is used for lifting the object from the clutter first and the gripper for grasping the object accordingly. We utilize the affordance map to provide pixel-wise lifting point candidates for the suction cup. To obtain a good affordance map, the active exploration mechanism is introduced to the system. An effective metric is designed to calculate the reward for the current affordance map, and a deep Q-Network (DQN) is employed to guide the robotic hand to actively explore the environment until the generated affordance map is suitable for grasping. Experimental results have demonstrated that the proposed robotic grasping system is able to greatly increase the success rate of the robotic grasping in cluttered scenes.

View on arXivView PDF

Code Implementations(1)

weiyx16/Active-PerceptionCommunity74%
MIT

Deep Reinforcement Learning for Robotic Pushing and Picking in Cluttered Environment

196Lua, PythonJan 22, 20196 years agoMIT
active-perceptiondeep-reinforcement-learningdqnrobotic-manipulationu-net+1 more

Similar Papers

Corner-Grasp: Multi-Action Grasp Detection and Active Gripper Adaptation for Grasping in Cluttered Environments

84% match2025

Learning Dual-Arm Coordination for Grasping Large Flat Objects

83% match2025

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

82% match2024

Learning Extrinsic Dexterity with Parameterized Manipulation Primitives

82% match2023

Pseudo-Labeling and Contextual Curriculum Learning for Online Grasp Learning in Robotic Bin Picking

82% match2024

GraspClutter6D: A Large-scale Real-world Dataset for Robust Perception and Grasping in Cluttered Scenes

81% match1 repo332025

Cite this paper

@article{deng2023deep,
  title  = {Deep Reinforcement Learning for Robotic Pushing and Picking in Cluttered Environment},
  author = {Yuhong Deng and Xiaofeng Guo and Yixuan Wei and Kai Lu and Bin Fang and Di Guo and Huaping Liu and Fuchun Sun},
  year   = {2023},
  eprint = {2302.10717},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.10717v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.10717.svg)](https://codeofpaper.com/paper/2302.10717)

Discussion