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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Reinforcement Learning (RL), among other learning-based methods, representspowerful tools to solve complex robotic tasks (e.g., actuation, manipulation,navigation, etc.), with the need for real-world data to train these systems asone of its most important limitations. The use of simulators is one way toaddress this issue, yet knowledge acquired in simulations does not workdirectly in the real-world, which is known as the sim-to-real transfer problem.While previous works focus on the nature of the images used as observations(e.g., textures and lighting), which has proven useful for a sim-to-simtransfer, they neglect other concerns regarding said observations, such asprecise geometrical meanings, failing at robot-to-robot, and thus insim-to-real transfers. We propose a method that learns on an observation spaceconstructed by point clouds and environment randomization, generalizing amongrobots and simulators to achieve sim-to-real, while also addressing partialobservability. We demonstrate the benefits of our methodology on the point goalnavigation task, in which our method proves to be highly unaffected to unseenscenarios produced by robot-to-robot transfer, outperforms image-basedbaselines in robot-randomized experiments, and presents high performances insim-to-sim conditions. Finally, we perform several experiments to validate thesim-to-real transfer to a physical domestic robot platform, confirming theout-of-the-box performance of our system.

Point Cloud Based Reinforcement Learning for Sim-to-Real and Partial Observability in Visual Navigation