用知识蒸馏网络预测混合交通中车辆跟驰行为，兼顾精度与效率。
Knowledge Distillation Neural Network for Predicting Car-following Behaviour of Human-driven and Autonomous Vehicles
- 基于真实轨迹数据，构建知识蒸馏神经网络预测跟驰速度。
- 模型精度接近LSTM，且最小时间到碰撞值更优，防撞性能更强。
- 计算开销低，适合自动驾驶系统实时应用。
随着自动驾驶车辆（AV）与人工驾驶车辆（HDV）共存的混合交通场景推进，理解跟驰行为对提升交通效率与道路安全至关重要。基于真实轨迹数据，本研究通过描述性与统计分析，探究了三类车对：HDV-AV、AV-HDV 和 HDV-HDV 的跟驰行为差异。方差分析（ANOVA）显示，不同车对间的跟驰行为具有统计显著性（p < 0.05）。我们提出一种数据驱动的知识蒸馏神经网络（KDNN）模型，用于预测跟驰速度。该模型在预测精度上与教师网络（长短期记忆网络，LSTM）相当，优于独立学生网络（多层感知机，MLP）及传统物理模型（如Gipps模型）。值得注意的是，KDNN 模型在最小时间到碰撞（TTC）指标上表现更优，有效降低碰撞风险，同时计算资源消耗更低，适用于对计算效率要求高的自动驾驶系统或驾驶模拟器。
原文摘要 · Abstract (English)
As we move towards a mixed-traffic scenario of Autonomous vehicles (AVs) and Human-driven vehicles (HDVs), understanding the car-following behaviour is important to improve traffic efficiency and road safety. Using a real-world trajectory dataset, this study uses descriptive and statistical analysis to investigate the car-following behaviours of three vehicle pairs: HDV-AV, AV-HDV and HDV-HDV in mixed traffic. The ANOVA test showed that car-following behaviours across different vehicle pairs are statistically significant (p-value < 0.05). We also introduce a data-driven Knowledge Distillation Neural Network (KDNN) model for predicting car-following behaviour in terms of speed. The KDNN model demonstrates comparable predictive accuracy to its teacher network, a Long Short-Term Memory (LSTM) network, and outperforms both the standalone student network, a Multilayer Perceptron (MLP), and traditional physics-based models like the Gipps model. Notably, the KDNN model better prevents collisions, measured by minimum Time-to-Collision (TTC), and operates with lower computational power, making it ideal for AVs or driving simulators requiring efficient computing.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。