Abstract:Evaluating the decision-making system is indispensable in developing autonomous vehicles, while realistic and challenging safety-critical test scenarios play a crucial role. Obtaining these scenarios is non-trivial, thanks to the long-tailed distribution, sparsity, and rarity in real-world data sets. To tackle this problem, in this paper, we introduce a natural adversarial scenario generation solution using naturalistic human driving priors and reinforcement learning techniques. By doing this, we can obtain large-scale test scenarios that are both diverse and realistic. Specifically, we build a simulation environment that mimics natural traffic interaction scenarios. Informed by this environment, we implement a two-stage procedure. The first stage incorporates conventional rule-based models, e.g., IDM~(Intelligent Driver Model) and MOBIL~(Minimizing Overall Braking Induced by Lane changes) model, to coarsely and discretely capture and calibrate key control parameters from the real-world dataset. Next, we leverage GAIL~(Generative Adversarial Imitation Learning) to represent driver behaviors continuously. The derived GAIL can be further used to design a PPO~(Proximal Policy Optimization)-based actor-critic network framework to fine-tune the reward function, and then optimizes our natural adversarial scenario generation solution. Extensive experiments have been conducted in the NGSIM dataset including the trajectory of 3,000 vehicles. Essential traffic parameters were measured in comparison with the baseline model, e.g., the collision rate, accelerations, steering, and the number of lane changes. Our findings demonstrate that the proposed model can generate realistic safety-critical test scenarios covering both naturalness and adversariality, which can be a cornerstone for the development of autonomous vehicles.

LMM-enhanced Safety-Critical Scenario Generation for Autonomous Driving System Testing From Non-Accident Traffic Videos

LeGEND: A Top-Down Approach to Scenario Generation of Autonomous Driving Systems Assisted by Large Language Models

Adversarial Safety-Critical Scenario Generation using Naturalistic Human Driving Priors

Adversarial Generation of Safety-Critical Lane-Change Scenarios for Autonomous Vehicles

Critical Scenario Generation for Realistic Testing of Autonomous Driving System Via Multi-vehicle Cooperative Control

Accelerated Verification of Autonomous Driving Systems Based on Adaptive Subset Simulation

A Hazardous Lane-Changing Scenario Search Approach Based on Adaptive Random Testing for Automated Vehicles

Interactive Critical Scenario Generation for Autonomous Vehicles Testing Based on In-depth Crash Data Using Reinforcement Learning

Generating Critical Scenarios for Testing Automated Driving Systems

AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehicles

MOSAT: Finding Safety Violations of Autonomous Driving Systems Using Multi-Objective Genetic Algorithm.

EvoScenario: Integrating Road Structures into Critical Scenario Generation for Autonomous Driving System Testing

ADEPT: A Testing Platform for Simulated Autonomous Driving.

Restricted Natural Language and Model-based Adaptive Test Generation for Autonomous Driving

Realistic Safety-critical Scenarios Search for Autonomous Driving System via Behavior Tree

Multimodal Large Language Model Driven Scenario Testing for Autonomous Vehicles

A Risk Level Assessment Method for Traffic Scenarios Based on BEV Perception.

DeepMF: Deep Motion Factorization for Closed-Loop Safety-Critical Driving Scenario Simulation

From Collision to Verdict: Responsibility Attribution for Autonomous Driving Systems Testing

A Survey on Safety-Critical Driving Scenario Generation -- A Methodological Perspective

From Dashcam Videos to Driving Simulations: Stress Testing Automated Vehicles against Rare Events