Robotic grasping in cluttered environments is a fundamental and challenging task in robotics research. The ability to autonomously grasp objects in cluttered scenes is crucial for robots to perform complex tasks in real-world scenarios. Conventional grasping is based on the known object model in a structured environment, but the adaptability of unknown objects and complicated situations is constrained. In this paper, we present a robotic grasp architecture of attention-based deep reinforcement learning. To prevent the loss of local information, the prominent characteristics of input images are automatically extracted using a full convolutional network. In contrast to previous model-based and data-driven methods, the reward is remodeled in an effort to address the sparse rewards. The experimental results show that our method can double the learning speed in grasping a series of randomly placed objects. In real-word experiments, the grasping success rate of the robot platform reaches 90.4%, which outperforms several baselines.

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.

Grasping objects in cluttered environments remains a significant challenge in robotics, particularly when dealing with novel objects that have not been previously encountered. This paper proposes a novel approach to address the problem of robustly learning object grasping in cluttered scenes, focusing on scenarios where the objects are unstructured and randomly placed. We present a unique Deep Q-learning (DQN) framework combined with a full convolutional network suitable for the end-to-end grasping of multiple adhesive objects in a cluttered environment. Our method combines the depth information of objects with reinforcement learning to obtain an adaptive grasping strategy to enable a robot to learn and generalize grasping skills for novel objects in the real world. The experimental results demonstrate that our method significantly improves the grasping performance on novel objects compared to conventional grasping techniques. Our system demonstrates remarkable adaptability and robustness in cluttered scenes, effectively grasping a diverse array of objects that were previously unseen. This research contributes to the advancement of robotics with potential applications, including, but not limited to, redundant manipulators, dual-arm robots, continuum robots, and soft robots.