基于深度强化学习的主线优化和匝道合流的协同控制模型与测试数据

由于高速公路交织区频繁的合流和变道带来的交通干扰，合流区常被认为是高速公路研究中最具挑战性的交通场景之一。联网自动驾驶车辆 (CAV) 被研究用于减少合流区域的拥堵，它通过与交通管制员实时通信发出的控制命令。本数据集汇集了测试过程中训练较优的几组模型，可在SUMO中进行数据重建与控制观察。

Due to frequent traffic disturbances caused by merging and lane-changing maneuvers in highway weaving sections, merging zones are widely regarded as one of the most challenging traffic scenarios in highway research. Connected and Automated Vehicles (CAVs) have been studied to alleviate congestion in merging zones, as they can issue control commands through real-time communication with traffic controllers. This dataset compiles several optimally trained models from the testing phase, supporting data reconstruction and control observation within the SUMO traffic simulation environment.

搜集汇总

背景与挑战

背景概述

该数据集包含基于深度强化学习的高速公路主线优化和匝道合流协同控制模型及测试数据，旨在解决高速公路交织区的交通拥堵问题。数据集包含70个文件，总大小为20.63MB，适用于SUMO仿真环境中的数据重建与控制观察。