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

The task of vision-and-language navigation in continuous environments(VLN-CE) aims at training an autonomous agent to perform low-level actions tonavigate through 3D continuous surroundings using visual observations andlanguage instructions. The significant potential of VLN-CE for mobile robotshas been demonstrated across a large number of studies. However, most existingworks in VLN-CE focus primarily on transferring the standard discretevision-and-language navigation (VLN) methods to continuous environments,overlooking the problem of collisions. Such oversight often results in theagent deviating from the planned path or, in severe instances, the agent beingtrapped in obstacle areas and failing the navigational task. To address theabove-mentioned issues, this paper investigates various collision scenarioswithin VLN-CE and proposes a classification method to predicate the underlyingcauses of collisions. Furthermore, a new VLN-CE algorithm, named Safe-VLN, isproposed to bolster collision avoidance capabilities including two keycomponents, i.e., a waypoint predictor and a navigator. In particular, thewaypoint predictor leverages a simulated 2D LiDAR occupancy mask to prevent thepredicted waypoints from being situated in obstacle-ridden areas. Thenavigator, on the other hand, employs the strategy of `re-selection aftercollision' to prevent the robot agent from becoming ensnared in a cycle ofperpetual collisions. The proposed Safe-VLN is evaluated on the R2R-CE, theresults of which demonstrate an enhanced navigational performance and astatistically significant reduction in collision incidences.

Safe-VLN: Collision Avoidance for Vision-and-Language Navigation of Autonomous Robots Operating in Continuous Environments