Towards an Ontology for Scenario Definition for the Assessment of Automated Vehicles: An Object-Oriented Framework

Abstract

The development of new assessment methods for the performance of automated vehicles is essential to enable the deployment of automated driving technologies, due to the complex operational domain of automated vehicles. One contributing method is scenario-based assessment in which test cases are derived from real-world road traffic scenarios obtained from driving data. Given the complexity of the reality that is being modeled in these scenarios, it is a challenge to define a structure for capturing these scenarios. An intensional definition that provides a set of characteristics that are deemed to be both necessary and sufficient to qualify as a scenario assures that the scenarios constructed are both complete and intercomparable.In this article, we develop a comprehensive and operable definition of the notion of scenario while considering existing definitions in the literature. This is achieved by proposing an object-oriented framework in which scenarios and their building blocks are defined as classes of objects having attributes, methods, and relationships with other objects. The object-oriented approach promotes clarity, modularity, reusability, and encapsulation of the objects. We provide definitions and justifications of each of the terms. Furthermore, the framework is used to translate the terms in a coding language that is publicly available.

Scenario Parameter Generation Method and Scenario Representativeness Metric for Scenario-Based Assessment of Automated Vehicles

“…There is no single best way to parameterize the scenarios considering the wide variety of scenarios. To deal with this variety, this work distinguishes quantitative scenarios from qualitative scenarios, using the definitions of scenario and scenario category of [34]: Definition 1 (Scenario). A scenario is a quantitative description of the relevant characteristics and activities and/or goals of the ego vehicle(s), the static environment, the dynamic environment and all events that are relevant to the ego vehicle(s) within the time interval between the first and last relevant event.…”

“…A scenario category is an abstraction of a scenario and, therefore, a scenario category comprises multiple scenarios [34]. For example, the scenario category "cut-in" comprises all possible cut-in scenarios.…”

Plane and Sample: Maximizing Information about Autonomous Vehicle Performance using Submodular Optimization

“…To structure the scenario space, we take advantage of the geometric representation that uses features, or building blocks, as coordinates of a scenario vector [18], [19]. This modular view captures the combinatorial size of the scenario space.…”

Safety of the Intended Driving Behavior Using Rulebooks

“…Challenges with scenario-based testing include (i) the difficulty of extrapolating the results from staged tests to realworld safety performance, (ii) the absence of consensus on the number of scenarios required, and (iii) the absence of general criteria to assess whether the AV behavior in a scenario is acceptable [13]. While research to date offers approaches to generate scenarios and define appropriate completeness metrics, none offer a universal method to define the desired AV behavior in different scenarios [23], [24], [25], [26]. Consistent with current guidance, [5], [13], we believe that addressing the SOTIF requires a complimentary approach that combines statistical and scenario-based testing to address the limitations of each individual method.…”

“…The scenarios can come from various sources, including System-Theoretic Process Analysis (STPA) [36], adaptive testing [37], orthogonal arrays [38], or human driving situations [24], [39], [23].…”