Deep Learning for Detecting Robotic Grasps

From Ian Lenz, Honglak Lee, Ashutosh Saxena:

Abstract

We consider the problem of detecting robotic grasps in an RGB-D view of a scene containing objects. In this work, we apply a deep learning approach to solve this problem, which avoids time-consuming hand-design of features. This presents two main challenges. First, we need to evaluate a huge number of candidate grasps. In order to make detection fast and robust, we present a two-step cascaded system with two deep networks, where the top detections from the first are re-evaluated by the second. The first network has fewer features, is faster to run, and can effectively prune out unlikely candidate grasps. The second, with more features, is slower but has to run only on the top few detections. Second, we need to handle multimodal inputs effectively, for which we present a method that applies structured regularization on the weights based on multimodal group regularization. We show that our method improves performance on an RGBD robotic grasping dataset, and can be used to successfully execute grasps on two different robotic platforms... (homepage) (full pdf paper)

Comments (0)

This post does not have any comments. Be the first to leave a comment below.


Post A Comment

You must be logged in before you can post a comment. Login now.

Featured Product

Universal Robots Add a Sense of Touch in New e-Series Cobots with  Built-in Force/Torque Sensor and Re-Designed User Interface

Universal Robots Add a Sense of Touch in New e-Series Cobots with Built-in Force/Torque Sensor and Re-Designed User Interface

With the new e-Series cobot line, Universal Robots raises the bar for cobots, adding unique new features while significantly strengthening the four core principles defining collaborative robots: fast set-up, easy programming, flexible deployment, and safe operation. With a new built-in, tool-centric Force/Torque sensor the e-Series is ready to take on applications requiring force control right out of the box. A repeatability of 30 micron means the new cobots are suitable for very precise finishing, assembly and electronics tasks. A re-designed user interface decreases cognitive load and expedites program development, while a new externally accessible, 500Hz system bus enables more complex motion control algorithms or profiles.