English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Visual features can help predict if a manipulation behavior will succeed at agiven location. For example, the success of a behavior that flips lightswitches depends on the location of the switch. Within this paper, we presentmethods that enable a mobile manipulator to autonomously learn a function thattakes an RGB image and a registered 3D point cloud as input and returns a 3Dlocation at which a manipulation behavior is likely to succeed. Given a pair ofmanipulation behaviors that can change the state of the world between two sets(e.g., light switch up and light switch down), classifiers that detect wheneach behavior has been successful, and an initial hint as to where one of thebehaviors will be successful, the robot autonomously trains a pair of supportvector machine (SVM) classifiers by trying out the behaviors at locations inthe world and observing the results. When an image feature vector associatedwith a 3D location is provided as input to one of the SVMs, the SVM predicts ifthe associated manipulation behavior will be successful at the 3D location. Toevaluate our approach, we performed experiments with a PR2 robot from WillowGarage in a simulated home using behaviors that flip a light switch, push arocker-type light switch, and operate a drawer. By using active learning, therobot efficiently learned SVMs that enabled it to consistently succeed at thesetasks. After training, the robot also continued to learn in order to adapt inthe event of failure.

Autonomously Learning to Visually Detect Where Manipulation Will Succeed