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

During the development of autonomous systems such as driverless cars, it isimportant to characterize the scenarios that are most likely to result infailure. Adaptive Stress Testing (AST) provides a way to search for themost-likely failure scenario as a Markov decision process (MDP). Our previouswork used a deep reinforcement learning (DRL) solver to identify likely failurescenarios. However, the solver's use of a feed-forward neural network with adiscretized space of possible initial conditions poses two major problems.First, the system is not treated as a black box, in that it requires analyzingthe internal state of the system, which leads to considerable implementationcomplexities. Second, in order to simulate realistic settings, a new instanceof the solver needs to be run for each initial condition. Running a new solverfor each initial condition not only significantly increases the computationalcomplexity, but also disregards the underlying relationship between similarinitial conditions. We provide a solution to both problems by employing arecurrent neural network that takes a set of initial conditions from acontinuous space as input. This approach enables robust and efficient detectionof failures because the solution generalizes across the entire space of initialconditions. By simulating an instance where an autonomous car drives while apedestrian is crossing a road, we demonstrate the solver is now capable offinding solutions for problems that would have previously been intractable.

Efficient Autonomy Validation in Simulation with Adaptive Stress Testing