Abstract:To meet the requirements of scenario-based testing for Autonomous Vehicles (AVs), driving scenario characterization has become a critical issue. Existing studies have concluded that complexity is a necessary criticality measure for supporting critical AV testing scenario identification. However, the existing scenario complexity quantification studies mainly have two limitations, namely, subjective quantification methods highly rely on human participations and are difficult to apply to big data, and existing objective methods lack consideration of human driver characteristics and hence the performance cannot be guaranteed. To bridge these research gaps, a general objective quantification framework is proposed to quantify human drivers' judgement on driving environment complexity, by describing the vehicle-vehicle spatial-temporal interactions from the perspectives of quantity, variety, and relations. The model mainly contains three parts. First, to describe quantity information, a dynamic influencing area was set to identify the surrounding vehicles that contribute to driving environment complexity based on ResponsibilitySensitive Safety (RSS) theory. Second, considering the various surrounding vehicles' driving statuses, behaviors, and intentions, a basic vehicle-pair complexity quantification model was constructed based on encounter angles. Then, nonlinear relationships based upon information entropy theory were introduced to capture the heterogeneous longitudinal and lateral complexities. Third, a vehicle-pair complexity aggregation and smoothing step was conducted to reflect the characteristics of human driver's cognition. To demonstrate the abovementioned model, empirical Field Operational Test (FOT) data from Shanghai urban roadways were used to conduct case studies, and it can be concluded that this model can accurately describe the timing and extent of the complexity change, and reveal the complexity differences due to scenario type and spatial-temporal heterogeneity. Besides, Inter-Rater Reliability (IRR) index was calculated to validate the consistency of scenario complexity judgement between the proposed model and human drivers, and for performance comparison with the existing models. Finally, the applications of the proposed model and its further investigations have been discussed.

Complexity of Driving Scenarios Based on Traffic Accident Data

Driving Behaviour Style Study with a Hybrid Deep Learning Framework Based on GPS Data

An entropy‐based model for quantifying multi‐dimensional traffic scenario complexity

Dynamic Driving Environment Complexity Quantification Method and Its Verification

A system of vision sensor based deep neural networks for complex driving scene analysis in support of crash risk assessment and prevention

A Multimodal Data-Driven Approach for Driving Risk Assessment

Intersection Two-Vehicle Crash Scenario Specification for Automated Vehicle Safety Evaluation Using Sequence Analysis and Bayesian Networks

Complexity Quantification of Driving Scenarios with Dynamic Evolution Characteristics

Traffic Sensory Data Classification by Quantifying Scenario Complexity.

Unified Quantitative Method of Driving Risk by Comprehensive Considering Driver-Vehicle-Road Factors

Analyzing Traffic Crash Severity With Combination of Information Entropy and Bayesian Network

Analysis of Spatiotemporal Factors Affecting Traffic Safety Based on Multisource Data Fusion

Driving Risk Assessment Using Cluster Analysis Based on Naturalistic Driving Data

Causative Analysis of Freight Railway Accident in Specific Scenes Using a Data-Driven Bayesian Network

Dynamic Bayesian Network Approach to Evaluate Vehicle Driving Risk Based on On-Road Experiment Driving Data

Accident Diagnosis and Evaluation System in Parking Lots Using Multisource Data Based on Bayesian Networks

Driving risk assessment based on naturalistic driving study and driver attitude questionnaire analysis

Driving risk assessment using near-crash database through data mining of tree-based model

Modeling crash severity by considering risk indicators of driver and roadway: A Bayesian network approach

A Method for Vehicle Collision Risk Assessment through Inferring Driver's Braking Actions in Near-Crash Situations

Comprehensive Evaluation of Freeway Driving Risks Based on Fuzzy Logic