Development of a Framework for Generating Driving Safety Assessment Scenarios for Automated Vehicles

Abstract

Despite the technological advances in automated driving systems, traffic accidents involving automated vehicles (AVs) continue to occur, raising concerns over the safety and reliability of automated driving. For the smooth commercialization of AVs, it is necessary to systematically assess the driving safety of AVs under the various situations that they potentially face. In this context, these various situations are mostly implemented by using systematically developed scenarios. In accordance with this need, a scenario generation framework for the assessment of the driving safety of AVs is proposed by this study. The proposed framework provides a unified form of assessment with key components for each scenario stage to facilitate systematization and objectivity. The performance of the driving safety assessment scenarios generated within the proposed framework was verified. Traffic accident report data were used for verification, and the usefulness of the proposed framework was confirmed by generating a set of scenarios, ranging from functional scenarios to test cases. The scenario generation framework proposed in this study can be used to provide sustainable scenarios. In addition, from this, it is possible to create assessment scenarios for all road types and various assessment spaces, such as simulations, proving grounds, and real roads.

The typical AV accident scenarios in the urban area obtained by clustering and association rule mining of real-world accident reports

“…10 -(c). The framework is proposed by Ko et al (2022) [ 53 ] and Kang et al (2022a) [ 39 ] based on the PEGASUS project. The logical scenario provides a range of factors.…”

Categorizing Data-Driven Methods for Test Scenario Generation to Assess Automated Driving Systems

“…Eight methods were used to evaluate the scenarios generated in terms of their criticality using, e.g., risk metrics [16], [23], [31], [32]. Other criticality metrics rely on collision detections/systems that fail [34], maximum vehicle yaw rate [43], or are probability- [53] or expert-based [47]. The methods used to assess the achieved coverage of the targeted scenario space can be explorative using histograms [34].…”

“…A rarely used data source is a dataset of pre-crafted scenarios or trajectories [20], [23], [68], [73]. Accident databases, as data sources, share approximately 26% of the identified methods, which, in most cases, originate from a set of police accident reports [19], [24], [28]- [30], [33], [35], [43], [46], [47], [55], [61], [67], [69], [70], [75], [76]. In some cases, only accidents involving vehicles with an ADS were considered and filtered from existing accident databases [34], [37], [38], [64].…”

“…The categorization of representativity is performed between RA and RB, with RA depicting representativity within each scenario set or space, which is the case in 25 generation approaches [16], [23], [24], [29]- [32], [40]- [45], [47], [48], [53], [55], [60], [65]- [69], [72], [74]. To achieve this goal, filtering of the generated scenario set to reduce the number of scenarios is frequently performed.…”

Implications of As-built Highway Horizontal Curves on Vehicle Dynamic/Kinematic Characteristics Under Adaptive Cruise Control

“…Concrete scenarios (∑ [(adopted RC range) / interval + 1] = 4 + 4 + 4 + 4 + 4 + 8 + 13 = 41) are designed through the pairs sampling approach [43], which involves all feature combinations within the defined range in the logical scenario.…”