Bayesian Optimization applied for accelerated Virtual Validation of the Autonomous Driving Function

2025-07-30

AI Summary

Critical scenario discovery in autonomous driving simulation verification suffers from low efficiency and prohibitively high computational cost due to exhaustive exploration of high-dimensional parameter spaces. Method: This paper proposes a Bayesian optimization (BO)-based intelligent scenario sampling framework, the first to deeply integrate BO with a model predictive control (MPC) motion planner for efficient identification of diverse critical scenarios (e.g., lane departure). Contribution/Results: Compared to brute-force design-of-experiments (DoE) approaches, the framework reduces required simulations by several orders of magnitude while maintaining high recall. It significantly improves verification efficiency without compromising detection capability. The framework exhibits strong scalability and establishes a novel paradigm for efficient virtual validation of autonomous driving functionalities.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Rigorous Verification and Validation (V&V) of Autonomous Driving Functions (ADFs) is paramount for ensuring the safety and public acceptance of Autonomous Vehicles (AVs). Current validation relies heavily on simulation to achieve sufficient test coverage within the Operational Design Domain (ODD) of a vehicle, but exhaustively exploring the vast parameter space of possible scenarios is computationally expensive and time-consuming. This work introduces a framework based on Bayesian Optimization (BO) to accelerate the discovery of critical scenarios. We demonstrate the effectiveness of the framework on an Model Predictive Controller (MPC)-based motion planner, showing that it identifies hazardous situations, such as off-road events, using orders of magnitude fewer simulations than brute-force Design of Experiments (DoE) methods. Furthermore, this study investigates the scalability of the framework in higher-dimensional parameter spaces and its ability to identify multiple, distinct critical regions within the ODD of the motion planner used as the case study .

Problem

• Accelerate discovery of critical scenarios in autonomous driving validation
• Reduce computational cost in exploring vast parameter spaces
• Identify multiple hazardous situations efficiently in high-dimensional spaces

Innovation

• Bayesian Optimization accelerates critical scenario discovery
• Framework reduces simulations versus brute-force methods
• Scalable to high-dimensional parameter spaces