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/

Zendy Plus

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

Zendy Tools

Zendy Open

Intelligent robot grasping is a very challenging task due to its inherentcomplexity and non availability of sufficient labelled data. Since makingsuitable labelled data available for effective training for any deep learningbased model including deep reinforcement learning is so crucial for successfulgrasp learning, in this paper we propose to solve the problem of generatinggrasping Poses/Rectangles using a Pix2Pix Generative Adversarial Network(Pix2Pix GAN), which takes an image of an object as input and produces thegrasping rectangle tagged with the object as output. Here, we have proposed anend-to-end grasping rectangle generating methodology and embedding it to anappropriate place of an object to be grasped. We have developed two modules toobtain an optimal grasping rectangle. With the help of the first module, thepose (position and orientation) of the generated grasping rectangle isextracted from the output of Pix2Pix GAN, and then the extracted grasp pose istranslated to the centroid of the object, since here we hypothesize that likethe human way of grasping of regular shaped objects, the center ofmass/centroids are the best places for stable grasping. For other irregularshaped objects, we allow the generated grasping rectangles as it is to be fedto the robot for grasp execution. The accuracy has significantly improved forgenerating the grasping rectangle with limited number of Cornell GraspingDataset augmented by our proposed approach to the extent of 87.79%. Experimentsshow that our proposed generative model based approach gives the promisingresults in terms of executing successful grasps for seen as well as unseenobjects.

Generating Quality Grasp Rectangle using Pix2Pix GAN for Intelligent Robot Grasping