Deep Reinforcement Learning for Robotic Pushing and Picking in Cluttered Environment

TL;DR

This paper presents a novel robotic grasping system that combines a composite hand with deep reinforcement learning to improve object picking success in cluttered environments.

Contribution

It introduces an active exploration mechanism guided by deep Q-Networks to optimize affordance maps for robotic grasping in cluttered scenes.

Findings

Significantly increased grasp success rate in cluttered scenes

Effective use of active exploration for affordance map optimization

Integration of suction and gripping for stable object handling

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.