CC-SGG: Corner Case Scenario Generation using Learned Scene Graphs
CC-SGG: Corner Case Scenario Generation using Learned Scene Graphs
复制标题
DOI：
10.48550/arxiv.2309.09844
复制
发表时间：
2023-09
期刊：
影响因子：
--
通讯作者：
George Drayson;Efimia Panagiotaki;Daniel Omeiza;Lars Kunze
中科院分区：
文献类型：
--
作者：
George Drayson;Efimia Panagiotaki;Daniel Omeiza;Lars Kunze
角落场景是测试和验证自动驾驶汽车（AV）安全性的重要工具。由于这些场景通常在自然驾驶数据集中不充分存在，因此使用合成角落案例来增强数据可以极大地增强AV在独特情况下的安全操作。然而，生成合成的，但现实的，角落的情况下提出了一个重大挑战。在这项工作中，我们介绍了一种新的方法，基于异构图神经网络（HGNNs），将常规驾驶场景转换为角落的情况。为了实现这一点，我们首先生成简洁的表示，经常驾驶场景的场景图，最小限度地操纵它们的结构和属性。然后，我们的模型学习扰动这些图，以使用注意力和三重嵌入来生成角点情况。然后将输入和扰动图导入回模拟中以生成极端情况场景。我们的模型成功地学会了从输入场景图中生成角点情况，在我们的测试数据集上实现了89.9%的预测准确率。我们进一步验证了基线自动驾驶方法生成的场景，证明了我们的模型能够有效地为基线创建关键情况。
Corner case scenarios are an essential tool for testing and validating the safety of autonomous vehicles (AVs). As these scenarios are often insufficiently present in naturalistic driving datasets, augmenting the data with synthetic corner cases greatly enhances the safe operation of AVs in unique situations. However, the generation of synthetic, yet realistic, corner cases poses a significant challenge. In this work, we introduce a novel approach based on Heterogeneous Graph Neural Networks (HGNNs) to transform regular driving scenarios into corner cases. To achieve this, we first generate concise representations of regular driving scenes as scene graphs, minimally manipulating their structure and properties. Our model then learns to perturb those graphs to generate corner cases using attention and triple embeddings. The input and perturbed graphs are then imported back into the simulation to generate corner case scenarios. Our model successfully learned to produce corner cases from input scene graphs, achieving 89.9% prediction accuracy on our testing dataset. We further validate the generated scenarios on baseline autonomous driving methods, demonstrating our model's ability to effectively create critical situations for the baselines.