# PEM: Perception Error Model for Virtual Testing of Autonomous Vehicles

## TL;DR

This paper introduces Perception Error Models (PEM) for virtual testing of autonomous vehicles, enabling analysis of perception errors' impact on safety without detailed sensor modeling, using data-driven methods and open-source tools.

## Contribution

The paper presents a generalized data-driven approach to model perception errors and integrates PEMs into simulators for improved AV safety assessment.

## Key findings

- PEMs can effectively simulate perception errors in virtual environments.
- The approach reveals limitations in current AV safety evaluation metrics.
- PEMs enable analysis of sensor setup impacts on perception accuracy.

## Abstract

Even though virtual testing of Autonomous Vehicles (AVs) has been well recognized as essential for safety assessment, AV simulators are still undergoing active development. One particularly challenging question is to effectively include the Sensing and Perception (S&P) subsystem into the simulation loop. In this article, we define Perception Error Models (PEM), a virtual simulation component that can enable the analysis of the impact of perception errors on AV safety, without the need to model the sensors themselves. We propose a generalized data-driven procedure towards parametric modeling and evaluate it using Apollo, an open-source driving software, and nuScenes, a public AV dataset. Additionally, we implement PEMs in SVL, an open-source vehicle simulator. Furthermore, we demonstrate the usefulness of PEM-based virtual tests, by evaluating camera, LiDAR, and camera-LiDAR setups. Our virtual tests highlight limitations in the current evaluation metrics, and the proposed approach can help study the impact of perception errors on AV safety.