In this paper, the lattice model is presented, incorporating not only site information about preceding cars but also relative currents in front. We derive the stability condition of the extended model by considering a...In this paper, the lattice model is presented, incorporating not only site information about preceding cars but also relative currents in front. We derive the stability condition of the extended model by considering a small perturbation around the homogeneous flow solution and find that the improvement in the stability of traffic flow is obtained by taking into account preceding mixture traffic information. Direct simulations also confirm that the traffic jam can be suppressed efficiently by considering the relative currents ahead, just like incorporating site information in front. Moreover, from the nonlinear analysis of the extended models, the preceding mixture traffic information dependence of the propagating kink solutions for traffic jams is obtained by deriving the modified KdV equation near the critical point using the reductive perturbation method.展开更多
高斯混合模型(Gaussian mixture model,GMM)可以描述遥感数据的概率密度函数,通过估计各高斯分布的参数,计算后验概率,实现信息提取。为了提高利用GMM进行遥感信息提取的准确度,首先在GMM中使用马尔科夫随机场(Markov random field,MRF...高斯混合模型(Gaussian mixture model,GMM)可以描述遥感数据的概率密度函数,通过估计各高斯分布的参数,计算后验概率,实现信息提取。为了提高利用GMM进行遥感信息提取的准确度,首先在GMM中使用马尔科夫随机场(Markov random field,MRF)计算各像元邻域内各类地物的先验概率,代替各类地物的混合概率,使其反映出各类地物的空间相关性;然后在参数估计过程中利用模拟退火(simulated annealing,SA)思想获得全局最优的参数估计值;最后利用该参数估计值求出每个像元对于各类地物的后验概率,获得各类地物的空间分布。通过对遥感实验场的图像数据进行信息提取,发现所述新方法取得了更好的效果,证明了上述改进的有效性。展开更多
基金Project supported by the National Natural Science Foundation of China (Grant Nos. 60904068,10902076,11072117,and 61004113)
文摘In this paper, the lattice model is presented, incorporating not only site information about preceding cars but also relative currents in front. We derive the stability condition of the extended model by considering a small perturbation around the homogeneous flow solution and find that the improvement in the stability of traffic flow is obtained by taking into account preceding mixture traffic information. Direct simulations also confirm that the traffic jam can be suppressed efficiently by considering the relative currents ahead, just like incorporating site information in front. Moreover, from the nonlinear analysis of the extended models, the preceding mixture traffic information dependence of the propagating kink solutions for traffic jams is obtained by deriving the modified KdV equation near the critical point using the reductive perturbation method.