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

We focus on the task of object manipulation to an arbitrary goal pose, inwhich a robot is supposed to pick an assigned object to place at the goalposition with a specific orientation. However, limited by the execution spaceof the manipulator with gripper, one-step picking, moving and releasing mightbe failed, where a reorientation object pose is required as a transition. Inthis paper, we propose a learning-driven anytime prioritized search-basedsolver to find a feasible solution with low path cost in a short time. In ourwork, the problem is formulated as a hierarchical learning problem, with thehigh level finding a reorientation object pose, and the low level planningpaths between adjacent grasps. We learn an offline-training path cost estimatorto predict approximate path planning costs, which serve as pseudo rewards toallow for pre-training the high-level planner without interacting with thesimulator. To deal with the problem of distribution mismatch of the cost netand the actual execution cost space, a refined training stage is conducted withsimulation interaction. A series of experiments carried out in simulation andreal world indicate that our system can achieve better performances in theobject manipulation task with less time and less cost.

Efficient Object Manipulation to an Arbitrary Goal Pose: Learning-based Anytime Prioritized Planning