Efficient Stacking and Grasping in Unstructured Environments
Open Access
- 1 April 2024
- journal article
- research article
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 110 (2) , 1-14
Abstract
Robotics has been booming in recent years. Especially with the development of artificial intelligence, more and more researchers have devoted themselves to the field of robotics, but there are still many shortcomings in the multi-task operation of robots. Reinforcement learning has achieved good performance in manipulator manipulation, especially in grasping, but grasping is only the first step for the robot to perform actions, and it often ignores the stacking, assembly, placement, and other tasks to be carried out later. Such long-horizon tasks still face the problems of expensive time, dead-end exploration, and process reversal. Hierarchical reinforcement learning has some advantages in solving the above problems, but not all tasks can be learned hierarchically. This paper mainly solves the complex manipulation task of continuous multi-action of the manipulator by improving the method of hierarchical reinforcement learning, aiming to solve the task of long sequences such as stacking and alignment by proposing a framework. Our framework completes simulation experiments on various tasks and improves the success rate from 78.3% to 94.8% when cleaning cluttered toys. In the stacking toy experiment, the training speed is nearly three times faster than the baseline method. And our method can be generalized to other long-horizon tasks. Experiments show that the more complex the task, the greater the advantage of our framework.Keywords
Funding Information
- the foundation of National Natural Science Foundation of China (61973065)
- Natural Science Foundation of China (52075531)
- Fundamental Research Funds for the Central Universities (N2104008)
- the Central Government Guides the Local Science and Technology Develop-ment Special Fund (2021JH6/10500129)
- Innovative Talents Support Program of Liaoning Provincial Universities (LR2020047)
This publication has 40 references indexed in Scilit:
- Learn to Grasp Unknown-Adjacent Objects for Sequential Robotic ManipulationJournal of Intelligent & Robotic Systems, 2022
- Intelligent problem-solving as integrated hierarchical reinforcement learningNature Machine Intelligence, 2022
- Towards Hierarchical Task Decomposition using Deep Reinforcement Learning for Pick and Place Subtasks2021
- Robotic Grasping using Deep Reinforcement Learning2020
- Grasping Objects Mixed With TowelsIEEE Access, 2020
- Planning Multi-fingered Grasps as Probabilistic Inference in a Learned Deep Network2019
- Learning Channel-Wise Interactions for Binary Convolutional Neural Networks2019
- Video Captioning via Hierarchical Reinforcement Learning2018
- Curiosity-Driven Exploration by Self-Supervised Prediction2017
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015