Handling Noise in Search-Based Scenario Generation for Autonomous Driving Systems

TL;DR

This paper evaluates k-nearest neighbours-Averaging (kNN-Avg), a novel noise-handling technique for search-based scenario generation in autonomous driving, demonstrating its advantages over traditional repetition methods in real-world case studies.

Contribution

It introduces kNN-Avg as an effective noise mitigation method for multi-objective optimization in autonomous driving scenario generation, avoiding repeated costly evaluations.

Findings

kNN-Avg outperforms noisy baseline in real-world case study

kNN-Avg is more efficient than repetition methods

Guidelines for choosing between kNN-Avg and repetition methods

Abstract

This paper presents the first evaluation of k-nearest neighbours-Averaging (kNN-Avg) on a real-world case study. kNN-Avg is a novel technique that tackles the challenges of noisy multi-objective optimisation (MOO). Existing studies suggest the use of repetition to overcome noise. In contrast, kNN-Avg approximates these repetitions and exploits previous executions, thereby avoiding the cost of re-running. We use kNN-Avg for the scenario generation of a real-world autonomous driving system (ADS) and show that it is better than the noisy baseline. Furthermore, we compare it to the repetition-method and outline indicators as to which approach to choose in which situations.