Stability Margin Improvement of Vehicular Platoon Considering The platooning of autonomous vehicles has the potential to significantly improve traffic capacity, enhance highway safety, and reduce fuel consumption
The structure of decoupled non linear systems Decouplingoflineartime-invariantsystemsbystatefeedbackandprecompensationhasbeentreatedinseveralpapers(FalbandWolovich1967,Gilbert1969,WonhamandMorse19
Sampleddata vehicular platoon control with communication delay Thisarticleinvestigatessampled-datavehicularplatooncontrolwithcommunicationdelay.Anewsampled-datacontrolmethodisestablished,inwhichtheeffectofthecommu
Robust control of heterogeneous vehicular platoon with uncertain dynamics Platoonformationofhighwayvehicleshasthepotentialtosignificantlyenhanceroadsafety,improvehighwayutility,andincreasetrafficefficiency.However,variousunc
Dynamical Modeling and Distributed Control of Connected and Automated Vehicles Theplatooningofconnectedandautomatedvehicles(CAVs)isexpectedtohaveatransformativeimpactonroadtransportation,e.g.,enhancinghighwaysafety,improvingtraff
卷积神经网络研究综述 作为一个十余年来快速发展的崭新领域,深度学习受到了越来越多研究者的关注,它在特征提取和建模上都有着相较于浅层模型显然的优势.深度学习善于从原始输入数据中挖掘越来越抽象的特征表示,而这些表示具有良好的泛化能力.它克服了过去人工智能中被认为难以解决的一些问题.且随着训练数据集数量的显著增长以及芯片处理能
改进的基于卷积神经网络的图像超分辨率算法 针对现有的基于卷积神经网络的图像超分辨率算法参数较多、计算量较大、训练时间较长、图像纹理模糊等问题,结合现有的图像分类网络模型和视觉识别算法对其提出了改进。在原有的三层卷积神经网络中,调整卷积核大小,减少参数;加入池化层,降低维度,减少计算复杂度;提高学习率和输入子块的尺寸,减少训练消耗的时间;扩大
基于卷积神经网络的道路车辆检测方法 提出了一种基于卷积神经网络的前方车辆检测方法。首先,根据车底阴影特征,运用基于边缘增强的路面检测算法以及车底阴影自适应分割算法来分割并形成车底候选区域,以解决路面灰度分布不均及光照条件变化问题;其次,运用针对道路交通环境的卷积神经网络结构,建立图像样本库进行网络训练;在此基础上,采用基于卷积神经网络
滑模变结构控制理论及其算法研究与进展 针对近年来滑模变结构控制的发展状况,将滑模变结构控制分为18个研究方向,即滑模控制的消除抖振问题、准滑动模态控制、基于趋近律的滑模控制、离散系统滑模控制、自适应滑模控制、非匹配不确定性系统滑模控制、时滞系统滑模控制、非线性系统滑模控制、Terminal滑模控制、全鲁棒滑模控制、滑模观测器、神经网络滑
FlowTopologyonClosed_loopStabilityofVehiclePlatoon Besides automated controllers, the information flow among vehicles can significantly affect the dynamics of a platoon. This paper studies the influenc