AI News, Closing the loop for robotic grasping

Closing the loop for robotic grasping

QUT's Dr Jürgen Leitner said while grasping and picking up an object was a basic task for humans, it had proved incredibly difficult for machines.

By mapping what is in front of it using a depth image in a single pass, the robot doesn't need to sample many different possible grasps before making a decision, avoiding long computing times,' Mr Morrison said.

'In our real-world tests, we achieved an 83% grasp success rate on a set of previously unseen objects with adversarial geometry and 88% on a set of household objects that were moved during the grasp attempt.

'For example, in the Amazon Picking Challenge, which our team won in 2017, our robot CartMan would look into a bin of objects, make a decision on where the best place was to grasp an object and then blindly go in to try to pick it up,' he said 'Using this new method, we can process images of the objects that a robot views within about 20 milliseconds, which allows the robot to update its decision on where to grasp an object and then do so with much greater purpose.

#2 - Robot Grasping for New Objects with Dr. Juxi Leitner

Dr. Juxi Leitner talks about robot grasping for new objects. Podcast topics: Robotic grasping and manipulation in both structured and unstructured environments, ...

QUT Closes the Loop for Robotic Grasping

Roboticists at QUT have developed a faster and more accurate way for robots to grasp objects, including in cluttered and changing environments, which has the ...

Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach

Paper to appear at RSS 2018. Code available at This paper presents a real-time, .

Robot Grasping Pose Selection with R-CNN Initialization

RobotFusion: Grasping with a Robotic Manipulator via Multi-view Reconstruction.

RobotFusion: Grasping with a Robotic Manipulator via Multi-view Reconstruction.

Robot grasping cylindrical objects by the ellipse detection

We propose a novel approach to manipulating cylindrical objects in robotic applications and automatic production lines.

Grasping - Laparoscopic Grasper Robot

Vision Based Grasp Learning

My entry for the Hackaday Prize 2017: A semi-autonomous robotic grasping system using a Convolutional Neural Network to determine grasping strategies for ...

Pick & Pack an Individual Order on ShipHero Mobile

ShipHero gives warehouse team members the ability to manage inventory, pick, pack and ship orders, batch pick orders and receive inventory across devices.

IROS 2018: Data-efficient Robotic Clothing Assistance using Bayesian Nonparametric LVMs

Motor-skill learning for complex robotic tasks is a challenging problem due to the high task variability. Robotic clothing assistance is one such challenging ...