机构地区: 华南理工大学土木与交通学院
出 处: 《暨南大学学报(自然科学与医学版)》 2002年第5期31-34,共4页
摘 要: 利用递阶结构和神经网络来进行高速公路入口匝道控制,其基本思想是:把高速公路作为一个大系统问题,子系统为高速公路的路段,协调控制层负责计算各路段的期望轨线,应用神经网络对各路段交通状态进行预测,并根据预测结果实施控制.给出了控制器的构造方法并进行了仿真实验,实验结果表明,该方法能够有效地消除交通拥挤和维持主线车流稳定. The hierarchical structure and neural network are need to realize the onramp control of freeway. The freeway is treated as a large scale system and segments of freeway as subsystems. The coordinated controller is responsible for computing the expected state of each segment. Meanwhile, neural networks are applied to forecast the traffic state of each segment and corresponding control is taken according to the forecast. In the article the controller's constructing methods are also presented and simulation are carried out. Simulation results show that the control method can effectively eliminate traffic jam and maintain the stability of the main traffic flow.