Intelligent corner synthesis via cycle-consistent generative adversarial networks for efficient validation of autonomous driving systems
Intelligent corner synthesis via cycle-consistent generative adversarial networks for efficient validation of autonomous driving systems
复制标题
DOI：
10.1109/aspdac.2018.8297275
复制
发表时间：
2018-01
期刊：
影响因子：
--
通讯作者：
Handi Yu;Xin Li
中科院分区：
文献类型：
--
作者：
Handi Yu;Xin Li
当今的机动车辆通常配备有用于驾驶员辅助和/或自动驾驶的强大的数据处理系统。为了满足严格的安全标准，一个关键任务是确保在所有可能的操作条件下的故障率极小。这样的验证任务需要大量的道路测试数据来覆盖所有可能的角落。在本文中，我们描述了一种新的通用方法，综合和有效地产生一个广泛的角落的情况下进行验证。我们提出的方法是基于周期一致的生成对抗网络（CycleGANs），该网络由一小组图像样本训练，以数学方式将标称情况映射到其他角落情况。以STOP符号检测为例，我们的数值实验表明，所提出的方法能够减少高达100倍的验证误差给定有限的数据集的角落的情况下。
Today's automotive vehicles are often equipped with powerful data processing systems for driver assistance and/or autonomous driving. To meet the rigorous safety standard, one critical task is to ensure extremely small failure rate over all possible operation conditions. Such a validation task requires a large amount of on-road testing data to cover all possible corners. In this paper, we describe a novel general-purpose methodology to synthetically and efficiently generate a broad spectrum of corner cases for validation purpose. Our proposed method is based upon cycle-consistent generative adversarial networks (CycleGANs) trained by a small set of image samples to mathematically map a nominal case to other corner cases. By taking STOP sign detection as an example, our numerical experiments demonstrate that the proposed approach is able to reduce the validation error by up to 100× given a limited data set for corner cases.