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

The widescale deployment of Autonomous Vehicles (AV) appears to be imminentdespite many safety challenges that are yet to be resolved. It is well-knownthat there are no universally agreed Verification and Validation (VV)methodologies guarantee absolute safety, which is crucial for the acceptance ofthis technology. The uncertainties in the behaviour of the traffic participantsand the dynamic world cause stochastic reactions in advanced autonomoussystems. The addition of ML algorithms and probabilistic techniques addssignificant complexity to the process for real-world testing when compared totraditional methods. Most research in this area focuses on generatingchallenging concrete scenarios or test cases to evaluate the system performanceby looking at the frequency distribution of extracted parameters as collectedfrom the real-world data. These approaches generally employ Monte-Carlosimulation and importance sampling to generate critical cases. This paperpresents an efficient falsification method to evaluate the System Under Test.The approach is based on a parameter optimisation problem to search forchallenging scenarios. The optimisation process aims at finding the challengingcase that has maximum return. The method applies policy-gradient reinforcementlearning algorithm to enable the learning. The riskiness of the scenario ismeasured by the well established RSS safety metric, euclidean distance, andinstance of a collision. We demonstrate that by using the proposed method, wecan more efficiently search for challenging scenarios which could cause thesystem to fail in order to satisfy the safety requirements.

Efficient falsification approach for autonomous vehicle validation using a parameter optimisation technique based on reinforcement learning