Synthesizing and Identifying Noise Levels in Autonomous Vehicle Camera Radar Datasets
This work addresses sensor failure robustness for autonomous vehicles, but it is incremental as it builds on existing methods with a new data augmentation approach.
The paper tackled the problem of improving robustness in autonomous vehicle object detection pipelines by creating a synthetic data augmentation pipeline to simulate sensor failures and data deterioration, achieving a 54.4% accuracy in noise recognition on an augmented dataset.
Detecting and tracking objects is a crucial component of any autonomous navigation method. For the past decades, object detection has yielded promising results using neural networks on various datasets. While many methods focus on performance metrics, few projects focus on improving the robustness of these detection and tracking pipelines, notably to sensor failures. In this paper we attempt to address this issue by creating a realistic synthetic data augmentation pipeline for camera-radar Autonomous Vehicle (AV) datasets. Our goal is to accurately simulate sensor failures and data deterioration due to real-world interferences. We also present our results of a baseline lightweight Noise Recognition neural network trained and tested on our augmented dataset, reaching an overall recognition accuracy of 54.4\% on 11 categories across 10086 images and 2145 radar point-clouds.