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

Ensuring the safety of autonomous vehicles (AVs) is of utmost importance andtesting them in simulated environments is a safer option than conductingin-field operational tests. However, generating an exhaustive test suite toidentify critical test scenarios is computationally expensive as therepresentation of each test is complex and contains various dynamic and staticfeatures, such as the AV under test, road participants (vehicles, pedestrians,and static obstacles), environmental factors (weather and light), and theroad's structural features (lanes, turns, road speed, etc.). In this paper, wepresent a systematic technique that uses Instance Space Analysis (ISA) toidentify the significant features of test scenarios that affect their abilityto reveal the unsafe behaviour of AVs. ISA identifies the features that bestdifferentiate safety-critical scenarios from normal driving and visualises theimpact of these features on test scenario outcomes (safe/unsafe) in 2D. Thisvisualization helps to identify untested regions of the instance space andprovides an indicator of the quality of the test suite in terms of thepercentage of feature space covered by testing. To test the predictive abilityof the identified features, we train five Machine Learning classifiers toclassify test scenarios as safe or unsafe. The high precision, recall, and F1scores indicate that our proposed approach is effective in predicting theoutcome of a test scenario without executing it and can be used for testgeneration, selection, and prioritization.

Identifying and Explaining Safety-critical Scenarios for Autonomous Vehicles via Key Features