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

Testing black-box perceptual-control systems in simulation faces twodifficulties. Firstly, perceptual inputs in simulation lack the fidelity ofreal-world sensor inputs. Secondly, for a reasonably accurate perceptionsystem, encountering a rare failure trajectory may require running infeasiblymany simulations. This paper combines perception error models -- surrogates fora sensor-based detection system -- with state-dependent adaptive importancesampling. This allows us to efficiently assess the rare failure probabilitiesfor real-world perceptual control systems within simulation. Our experimentswith an autonomous braking system equipped with an RGB obstacle-detector showthat our method can calculate accurate failure probabilities with aninexpensive number of simulations. Further, we show how choice of safety metriccan influence the process of learning proposal distributions capable ofreliably sampling high-probability failures.

Testing Rare Downstream Safety Violations via Upstream Adaptive Sampling of Perception Error Models