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

Visual navigation by mobile robots is classically tackled through SLAM plusoptimal planning, and more recently through end-to-end training of policiesimplemented as deep networks. While the former are often limited to waypointplanning, but have proven their efficiency even on real physical environments,the latter solutions are most frequently employed in simulation, but have beenshown to be able learn more complex visual reasoning, involving complexsemantical regularities. Navigation by real robots in physical environments isstill an open problem. End-to-end training approaches have been thoroughlytested in simulation only, with experiments involving real robots beingrestricted to rare performance evaluations in simplified laboratory conditions.In this work we present an in-depth study of the performance and reasoningcapacities of real physical agents, trained in simulation and deployed to twodifferent physical environments. Beyond benchmarking, we provide insights intothe generalization capabilities of different agents training in differentconditions. We visualize sensor usage and the importance of the different typesof signals. We show, that for the PointGoal task, an agent pre-trained on widevariety of tasks and fine-tuned on a simulated version of the targetenvironment can reach competitive performance without modelling any sim2realtransfer, i.e. by deploying the trained agent directly from simulation to areal physical robot.

An in-depth experimental study of sensor usage and visual reasoning of robots navigating in real environments