[论文解读] Platoon trajectories generation: A unidirectional interconnected LSTM-based car following model

Yangxin Lin, Ping Wang+4 more|arXiv (Cornell University)|Oct 25, 2019

Traffic control and management参考文献 26被引用 6

一句话总结

本文提出一种基于单向互联LSTM的车辆跟驰模型，通过解决时间与空间上的误差传播问题，提升车队轨迹生成的准确性。通过整合调度采样与一种新型车队感知架构，与传统LSTM方法相比，误差降低近40%，显著提升了交通振荡场景下的微观仿真保真度。

ABSTRACT

Car following models have been widely applied and made remarkable achievements in traffic engineering. However, the traffic micro-simulation accuracy of car following models in a platoon level, especially during traffic oscillations, still needs to be enhanced. Rather than using traditional individual car following models, we proposed a new trajectory generation approach to generate platoon level trajectories given the first leading vehicle's trajectory. In this paper, we discussed the temporal and spatial error propagation issue for the traditional approach by a car following block diagram representation. Based on the analysis, we pointed out that error comes from the training method and the model structure. In order to fix that, we adopt two improvements on the basis of the traditional LSTM based car following model. We utilized a scheduled sampling technique during the training process to solve the error propagation in the temporal dimension. Furthermore, we developed a unidirectional interconnected LSTM model structure to extract trajectories features from the perspective of the platoon. As indicated by the systematic empirical experiments, the proposed novel structure could efficiently reduce the temporal and spatial error propagation. Compared with the traditional LSTM based car following model, the proposed model has almost 40% less error. The findings will benefit the design and analysis of micro-simulation for platoon level car following models.

研究动机与目标

• 为解决传统车辆跟驰模型在车队级交通微观仿真中长期存在的时序与空间误差传播问题。
• 提升车队轨迹生成的准确性，特别是在交通振荡条件下。
• 开发一种深度学习架构，以捕捉车队整体的集体动态行为，而非孤立建模单个车辆。
• 通过修改LSTM-based车辆跟驰模型的训练过程与模型结构，减少误差传播。

提出的方法

• 在训练过程中采用调度采样技术，以减轻时序维度上的误差累积。
• 设计一种单向互联LSTM架构，从车队前端到后端建模车辆间的依赖关系。
• 端到端训练模型，根据领头车的轨迹预测后续车辆的轨迹。
• 将模型构架为序列到序列学习任务，输入为领头车的轨迹，输出为所有后续车辆的轨迹。
• 通过在车队中单向传播隐藏状态，实现时空特征的联合学习。
• 使用车辆跟驰系统的框图表示法，分析并诊断误差传播机制。

实验结果

研究问题

• RQ1 在车队运行过程中，传统LSTM-based车辆跟驰模型中误差如何在时序与空间上传播？
• RQ2 调度采样在多大程度上可减少序列轨迹预测中的误差累积？
• RQ3 与标准循环神经网络架构相比，单向互联LSTM结构是否能更准确地捕捉车队级动态行为？
• RQ4 在模拟交通振荡时，所提出的模型与传统LSTM-based车辆跟驰模型相比，准确性如何？

主要发现

• 与基线LSTM-based车辆跟驰模型相比，所提模型将轨迹预测误差降低了约40%。
• 调度采样在序列生成过程中有效缓解了时序维度上的误差传播。
• 单向互联LSTM结构通过将车队整体建模，更准确地捕捉了车辆间的依赖关系。
• 实证结果表明，在交通振荡场景下，时空准确性均显著提升。
• 在模拟真实车队行为方面，该模型优于标准LSTM模型，尤其在高密度与不稳定交通条件下表现更优。