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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Robotic navigation has been approached as a problem of 3D reconstruction andplanning, as well as an end-to-end learning problem. However, long-rangenavigation requires both planning and reasoning about local traversability, aswell as being able to utilize general knowledge about global geography, in theform of a roadmap, GPS, or other side information providing important cues. Inthis work, we propose an approach that integrates learning and planning, andcan utilize side information such as schematic roadmaps, satellite maps and GPScoordinates as a planning heuristic, without relying on them being accurate.Our method, ViKiNG, incorporates a local traversability model, which looks atthe robot's current camera observation and a potential subgoal to infer howeasily that subgoal can be reached, as well as a heuristic model, which looksat overhead maps for hints and attempts to evaluate the appropriateness ofthese subgoals in order to reach the goal. These models are used by a heuristicplanner to identify the best waypoint in order to reach the final destination.Our method performs no explicit geometric reconstruction, utilizing only atopological representation of the environment. Despite having never seentrajectories longer than 80 meters in its training dataset, ViKiNG can leverageits image-based learned controller and goal-directed heuristic to navigate togoals up to 3 kilometers away in previously unseen environments, and exhibitcomplex behaviors such as probing potential paths and backtracking when theyare found to be non-viable. ViKiNG is also robust to unreliable maps and GPS,since the low-level controller ultimately makes decisions based on egocentricimage observations, using maps only as planning heuristics. For videos of ourexperiments, please check out our project pagehttps://sites.google.com/view/viking-release.

ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints