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/

Zendy Plus

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

Zendy Tools

Zendy Open

Empowering robots to navigate in a socially compliant manner is essential forthe acceptance of robots moving in human-inhabited environments. Previously,roboticists have developed geometric navigation systems with decades ofempirical validation to achieve safety and efficiency. However, the manycomplex factors of social compliance make geometric navigation systems hard toadapt to social situations, where no amount of tuning enables them to be bothsafe (people are too unpredictable) and efficient (the frozen robot problem).With recent advances in deep learning approaches, the common reaction has beento entirely discard these classical navigation systems and start from scratch,building a completely new learning-based social navigation planner. In thiswork, we find that this reaction is unnecessarily extreme: using a large-scalereal-world social navigation dataset, SCAND, we find that geometric systems canproduce trajectory plans that align with the human demonstrations in a largenumber of social situations. We, therefore, ask if we can rethink the socialrobot navigation problem by leveraging the advantages of both geometric andlearning-based methods. We validate this hybrid paradigm through aproof-of-concept experiment, in which we develop a hybrid planner that switchesbetween geometric and learning-based planning. Our experiments on both SCANDand two physical robots show that the hybrid planner can achieve better socialcompliance compared to using either the geometric or learning-based approachalone.

Rethinking Social Robot Navigation: Leveraging the Best of Two Worlds