Sequential Experimentation to Efficiently Test Automated Vehicles

TL;DR

This paper proposes a sequential learning method using kriging models and heuristic optimization to reduce the number of costly on-track tests for automated vehicle safety evaluation.

Contribution

It introduces a novel sequential experimentation framework that efficiently identifies critical test scenarios for automated vehicles, saving time and resources.

Findings

Reduces experimental runs in vehicle safety testing

Demonstrates effectiveness through numerical case studies

Offers a cost-efficient alternative to traditional testing

Abstract

Automated vehicles have been under heavy developments in major auto and tech companies and are expected to release into market in the foreseeable future. However, the road safety of these vehicles remains a concern. One approach to evaluate their safety is via on-track experimentation, but this requires gigantic costs and time investments. This paper discusses a sequential learning approach based on kriging models to reduce the experimental runs and economize on-track experimentation. The approach relies on a heuristic simulation-based gradient descent procedure to search for the best next test scenario. We demonstrate our approach with some numerical test cases.