Learning Push-Grasping in Dense Clutter

Kiatos, Marios; Sarantopoulos, Iason; Koutras, Leonidas; Malassiotis, Sotiris; Doulgeri, Zoe

doi:10.1109/lra.2022.3188437

Cited by 10 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [14], the RL was based on a hybrid discrete-continuous adaptation of soft actor-critic (SAC) for learning continuous and high-dimensional PG policies in robotic bin-picking. Different from one single network for pushing and grasping, Kiatos, et al [15] adopted two different networks, one for position and angle of the gripper, and the other one for opening size of the gripper. With two separate networks, Li, et al [16] proposed a position net for pushing and grasping positions, and an angle net for grasping angles.…”

Section: Deep-learning Modelsmentioning

confidence: 99%

An Exploratory Guided Push and Grasp

Tay,

Yan,

Chen

et al. 2024

Preprint

View full text Add to dashboard Cite

Teaching a robot to grasp cluttered and stacked objects is a common and challenging but important research topic that can potentially benefit many real-life applications. The challenge can be addressed by incorporating pushing action into grasping strategy. To make robot capable of pushing and grasping (PG) stacked objects, deep neural network (DNN), typically reinforcement learning, has been reported widely, while the limitations of long training time and low action efficiency are still obvious. In this work, an exploratory guided pushing and grasping approach was proposed using a self-supervised technique to expedite training cycle, and a memory buffer enhancement strategy to guide the training continuously. The network architecture, data collection and learning strategy were investigated and analysed to study their effects. Comprehensive experiments were conducted on both simulation and real environment to validate the proposed solution with high action efficiency, outperforming the state-of-the-art solutions with a large margin.

show abstract

Section: Deep-learning Modelsmentioning

confidence: 99%

An Exploratory Guided Push and Grasp

Tay,

Yan,

Chen

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…However, only depth image was considered in their work, and so the test results for novel unknown objects was not perfect. More recent research makes it possible to train robot to learn synergies between pushing and grasping in dense clutter [16][17][18][19]. These methods utilize visual observations for end-to-end decision-making without using object-specific knowledge.…”

Section: Related Workmentioning

confidence: 99%

“…Kalashnikov et al [20] introduce a scalable vision-based reinforcement learning framework named QT-Opt, which enables robots to learn how to pick up objects and execute nonprehensile pre-grasp actions. Kiatos et al [18] designed an experiment to learn a direct correlation between visual observations and actions, and it is trained in a comprehensive end-to-end manner. Without assuming a segmentation of the scene, the grasping policy accomplishes robust power grasps in cluttered environments.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Towards Multi-Objective Object Push-Grasp Policy Based on Maximum Entropy Deep Reinforcement Learning under Sparse Rewards

Zhang,

2024

Entropy

View full text Add to dashboard Cite

In unstructured environments, robots need to deal with a wide variety of objects with diverse shapes, and often, the instances of these objects are unknown. Traditional methods rely on training with large-scale labeled data, but in environments with continuous and high-dimensional state spaces, the data become sparse, leading to weak generalization ability of the trained models when transferred to real-world applications. To address this challenge, we present an innovative maximum entropy Deep Q-Network (ME-DQN), which leverages an attention mechanism. The framework solves complex and sparse reward tasks through probabilistic reasoning while eliminating the trouble of adjusting hyper-parameters. This approach aims to merge the robust feature extraction capabilities of Fully Convolutional Networks (FCNs) with the efficient feature selection of the attention mechanism across diverse task scenarios. By integrating an advantage function with the reasoning and decision-making of deep reinforcement learning, ME-DQN propels the frontier of robotic grasping and expands the boundaries of intelligent perception and grasping decision-making in unstructured environments. Our simulations demonstrate a remarkable grasping success rate of 91.6%, while maintaining excellent generalization performance in the real world.

show abstract

“…Recently, a multi-fingered push-grasping policy was proposed using deep Q-learning that creates enough space for the fingers to wrap around an object to perform a stable power grasp using a single primitive action [ 74 ]. A target-oriented robotic push-grasping system was proposed that is able to actively discover and pick up the impurities in dense environments with the synergies between pushing and grasping actions.…”

Section: Critical Reviewmentioning

confidence: 99%

Review of Learning-Based Robotic Manipulation in Cluttered Environments

Mohammed

Kwek

Chua

et al. 2022

Sensors

View full text Add to dashboard Cite

Robotic manipulation refers to how robots intelligently interact with the objects in their surroundings, such as grasping and carrying an object from one place to another. Dexterous manipulating skills enable robots to assist humans in accomplishing various tasks that might be too dangerous or difficult to do. This requires robots to intelligently plan and control the actions of their hands and arms. Object manipulation is a vital skill in several robotic tasks. However, it poses a challenge to robotics. The motivation behind this review paper is to review and analyze the most relevant studies on learning-based object manipulation in clutter. Unlike other reviews, this review paper provides valuable insights into the manipulation of objects using deep reinforcement learning (deep RL) in dense clutter. Various studies are examined by surveying existing literature and investigating various aspects, namely, the intended applications, the techniques applied, the challenges faced by researchers, and the recommendations adopted to overcome these obstacles. In this review, we divide deep RL-based robotic manipulation tasks in cluttered environments into three categories, namely, object removal, assembly and rearrangement, and object retrieval and singulation tasks. We then discuss the challenges and potential prospects of object manipulation in clutter. The findings of this review are intended to assist in establishing important guidelines and directions for academics and researchers in the future.

show abstract

Learning Push-Grasping in Dense Clutter

Cited by 10 publications

References 31 publications

An Exploratory Guided Push and Grasp

An Exploratory Guided Push and Grasp

Towards Multi-Objective Object Push-Grasp Policy Based on Maximum Entropy Deep Reinforcement Learning under Sparse Rewards

Review of Learning-Based Robotic Manipulation in Cluttered Environments

Contact Info

Product

Resources

About