[论文解读] Driving Anomaly Detection Using Conditional Generative Adversarial Network

Yuning Qiu, Teruhisa Misu+1 more|arXiv (Cornell University)|Mar 15, 2022

Anomaly Detection Techniques and Applications被引用 4

一句话总结

本论文提出了一种基于条件生成对抗网络（cGAN）的无监督驾驶异常检测方法，通过过去观测数据预测未来的CAN-Bus信号和生理信号。异常分数基于判别器对预测信号与实际信号输出差异计算得出，无需预设规则或标注异常，即可实现对危险且低频事件的高检测率。

ABSTRACT

Anomaly driving detection is an important problem in advanced driver assistance systems (ADAS). It is important to identify potential hazard scenarios as early as possible to avoid potential accidents. This study proposes an unsupervised method to quantify driving anomalies using a conditional generative adversarial network (GAN). The approach predicts upcoming driving scenarios by conditioning the models on the previously observed signals. The system uses the difference of the output from the discriminator between the predicted and actual signals as a metric to quantify the anomaly degree of a driving segment. We take a driver-centric approach, considering physiological signals from the driver and controller area network-Bus (CAN-Bus) signals from the vehicle. The approach is implemented with convolutional neural networks (CNNs) to extract discriminative feature representations, and with long short-term memory (LSTM) cells to capture temporal information. The study is implemented and evaluated with the driving anomaly dataset (DAD), which includes 250 hours of naturalistic recordings manually annotated with driving events. The experimental results reveal that recordings annotated with events that are likely to be anomalous, such as avoiding on-road pedestrians and traffic rule violations, have higher anomaly scores than recordings without any event annotation. The results are validated with perceptual evaluations, where annotators are asked to assess the risk and familiarity of the videos detected with high anomaly scores. The results indicate that the driving segments with higher anomaly scores are more risky and less regularly seen on the road than other driving segments, validating the proposed unsupervised approach.

研究动机与目标

• 开发一种无监督驾驶异常检测系统，能够在事故发生前识别出危险偏离行为。
• 克服基于规则或阈值的系统仅在错误发生后才作出反应的局限性。
• 利用多模态数据（CAN-Bus信号与驾驶员生理信号（ECG、EDA、BR））实现全面的异常检测。
• 通过驾驶异常数据集（DAD）的感知评估与人工标注，验证模型的有效性。

提出的方法

• 采用条件生成对抗网络框架，其中生成器基于过去观测信号预测未来的信号（CAN-Bus与生理信号）。
• 使用卷积神经网络（CNN）从原始传感器信号中提取空间与时间特征。
• 集成长短期记忆（LSTM）网络以建模驾驶序列中的序列依赖关系与时间动态。
• 将异常分数计算为真实信号与生成信号在判别器输出上的差异，差异越大表示偏离越严重。
• 采用端到端无监督方式训练模型，仅使用正常驾驶数据进行学习。
• 在包含250小时自然驾驶数据及人工标注危险事件的DAD数据集上评估模型。

实验结果

研究问题

• RQ1 条件生成对抗网络框架是否能在不依赖预设规则或标注异常的情况下有效检测驾驶异常？
• RQ2 该模型在检测客观上具有危险性但难以被传统基于规则的系统捕捉的事件方面表现如何？
• RQ3 与单模态输入相比，多模态输入（CAN-Bus与生理信号）在多大程度上提升了异常检测性能？
• RQ4 高异常分数片段的感知评估结果与随机片段相比，在风险感知与熟悉度方面有何差异？

主要发现

• 被标注为高风险事件（如避让行人、交通违规）的驾驶片段，其异常分数始终高于未标注片段。
• 基于CNN+LSTM的cGAN模型将33%的前100个高异常分数片段识别为“危险”或“非常危险”，而仅LSTM模型为27.7%，仅CNN模型为22.7%。
• CNN+LSTM模型选出的前100个片段中，33%被评价为“从未”或“几乎从未”在道路上发生，表明其对罕见异常事件具有强检测能力。
• 感知评估结果证实，高异常分数片段被感知为显著更危险（64.7%评为危险或非常危险），且比随机选择片段更不熟悉。
• CNN+LSTM模型在识别危险且罕见驾驶状况方面，优于仅使用CNN或仅使用LSTM的变体。
• 模型在DAD数据集上的性能得到验证，表明仅使用正常驾驶数据与多模态信号，即可实现无监督异常检测。