Abstract:Automated driving vehicles~(ADV) promise to enhance driving efficiency and safety, yet they face intricate challenges in safety-critical scenarios. As a result, validating ADV within generated safety-critical scenarios is essential for both development and performance evaluations. This paper investigates the complexities of employing two major scenario-generation solutions: data-driven and knowledge-driven methods. Data-driven methods derive scenarios from recorded datasets, efficiently generating scenarios by altering the existing behavior or trajectories of traffic participants but often falling short in considering ADV perception; knowledge-driven methods provide effective coverage through expert-designed rules, but they may lead to inefficiency in generating safety-critical scenarios within that coverage. To overcome these challenges, we introduce BridgeGen, a safety-critical scenario generation framework, designed to bridge the benefits of both methodologies. Specifically, by utilizing ontology-based techniques, BridgeGen models the five scenario layers in the operational design domain (ODD) from knowledge-driven methods, ensuring broad coverage, and incorporating data-driven strategies to efficiently generate safety-critical scenarios. An optimized scenario generation toolkit is developed within BridgeGen. This expedites the crafting of safety-critical scenarios through a combination of traditional optimization and reinforcement learning schemes. Extensive experiments conducted using Carla simulator demonstrate the effectiveness of BridgeGen in generating diverse safety-critical scenarios.

A new SOTIF scenario hierarchy and its critical test case generation based on potential risk assessment.

Genetic Algorithm-Based SOTIF Scenario Construction for Complex Traffic Flow

SOTIF-Compliant Scenario Generation Using Semi-Concrete Scenarios and Parameter Sampling

Method to Automatic Testcase Generation toward Safety Critical Scenarios of Cyber-physical Systems

Is Scenario Generation Ready for SOTIF? A Systematic Literature Review

A Quantitative Approach of Generating Challenging Testing Scenarios Based on Functional Safety Standard

A Test Scenario Automatic Generation Strategy for Intelligent Driving Systems

Test Scenario Design for Intelligent Driving System Ensuring Coverage and Effectiveness

Method of Automatic Test Case Generation for Safety-Critical Scenarios in Train Control Systems

Adaptive Mining of Failure Scenarios for Autonomous Driving Systems Based on Multi-population Genetic Algorithm

Test Scenario Generation and Optimization Technology for Intelligent Driving Systems

SOTIF Entropy: Online SOTIF Risk Quantification and Mitigation for Autonomous Driving

Test scenario generation method for autonomous vehicles based on combinatorial testing and Bayesian network

Testing Scenario Library Generation for Connected and Automated Vehicles, Part I: Methodology

Diversified Critical-Scenario-Search for Defect Detection of Autonomous Driving System

A Test Case Design Method Based on Path Depth Coverage

Bridging Data-Driven and Knowledge-Driven Approaches for Safety-Critical Scenario Generation in Automated Vehicle Validation

A Critical Scenario Search Method for Intelligent Vehicle Testing Based on the Social Cognitive Optimization Algorithm

Risk Response for Critical Infrastructures with Multiple Interdependent Risks: A Scenario-Based Extended CBR Approach