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

Grasping an object when it is in an ungraspable pose is a challenging task,such as books or other large flat objects placed horizontally on a table.Inspired by human manipulation, we address this problem by pushing the objectto the edge of the table and then grasping it from the hanging part. In thispaper, we develop a model-free Deep Reinforcement Learning framework tosynergize pushing and grasping actions. We first pre-train a VariationalAutoencoder to extract high-dimensional features of input scenario images. OneProximal Policy Optimization algorithm with the common reward and sharinglayers of Actor-Critic is employed to learn both pushing and grasping actionswith high data efficiency. Experiments show that our one network policy canconverge 2.5 times faster than the policy using two parallel networks.Moreover, the experiments on unseen objects show that our policy can generalizeto the challenging case of objects with curved surfaces and off-centerirregularly shaped objects. Lastly, our policy can be transferred to a realrobot without fine-tuning by using CycleGAN for domain adaption and outperformsthe push-to-wall baseline.

Reinforcement Learning Based Pushing and Grasping Objects from Ungraspable Poses