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

Cyber-physical systems (CPS) designed in simulators behave differently in thereal-world. Once they are deployed in the real-world, we would hence like topredict system failures during runtime. We propose robust predictive runtimeverification (RPRV) algorithms under signal temporal logic (STL) tasks forgeneral stochastic CPS. The RPRV problem faces several challenges: (1) theremay not be sufficient data of the behavior of the deployed CPS, (2) predictivemodels are based on a distribution over system trajectories encountered duringthe design phase, i.e., there may be a distribution shift during deployment. Toaddress these challenges, we assume to know an upper bound on the statisticaldistance (in terms of an f-divergence) between the distributions at deploymentand design time, and we utilize techniques based on robust conformalprediction. Motivated by our results in [1], we construct an accurate and aninterpretable RPRV algorithm. We use a trajectory prediction model to estimatethe system behavior at runtime and robust conformal prediction to obtainprobabilistic guarantees by accounting for distribution shifts. We preciselyquantify the relationship between calibration data, desired confidence, andpermissible distribution shift. To the best of our knowledge, these are thefirst statistically valid algorithms under distribution shift in this setting.We empirically validate our algorithms on a Franka manipulator within theNVIDIA Isaac sim environment.

Robust Conformal Prediction for STL Runtime Verification under Distribution Shift