威尼斯人

会议报告摘要

Constructing the evolution models of biological networks via simple rules

刘曾荣上海大学

It has been widely recognized that networks is a good tool for describing various artificial and natural systems. Thus, a natural problem arises: how to construct networks and further properly study its features. At present, it is still lack of basic laws for describing the quantitative relationships in biological networks, social networks, etc. Therefore, the construction method of networks should be different from that used to build up the mathematical models in the past based on the quantitative relationships in various scientific fields, such as physics and mechanics. Here, we provide a method to construct protein-protein interaction networks via simple rules and further analyze its features. The main feature of the method is that the basic rules for constructing networks are firstly proposed based on the point of view of natural selection and biological evolution, and further the rationality of the rules is evaluated by comparing the topological characteristics of the constructed networks with those measured from real biological interaction networks.

CHARACTERISTIC POLYNOMIAL METHOD FOR ANALYZING DYNAMICS OF BOOLEAN NETWORKS

陈芳跃杭州电子科技大学

In our work, an efficient method for analyzing the dynamics of Boolean networks based on the characteristic polynomial of the linearization matrix of Boolean network is proposed. The results presented in this work may provide a new basis for the elucidation of the function of complexity in the living cells, and therefore are significant and meaningful for systems biology and mathematical biosciences research.

多机器人系统网络化协调模式的动力学与控制

周进上海大学

本报告主要介绍从现代网络科学和工程的角度来研究多机械臂系统协调模式的动力学与控制问题，以及由这种多机器人系统网络化的研究模式所引发一些值得关注科学问题的思考与展望，同时介绍我们近年来在这一领域所取得的一些研究进展。

State Estimation Based-on the Outputs of Partial Nodes for Discrete-Time Stochastic Complex Networks

刘玉荣扬州大学

In this study, the state estimation problem is studied for a class of discrete-time stochastic complex networks with switched topology. In the network under consideration, we assume that measurement outputs can be got from only partial nodes, besides, the switching rule of this network is characterized by a sequence of Bernoulli random variables. The aim of the presented estimation problem is to develop a recursive estimator based on the framework of extended Kalman filter (EKF), such that the upper bound for the filtering error convariance is optimized. In order to address the nonlinear functions, the Taylor series expansion is utilized and the high-order terms of linearization errors are expressed in an exact way. Furthermore, by solving two Ricatti-like difference equations, the gain matrix can be acquired at each time instant. It is shown that the filtering error is bounded in mean square under some conditions with the aid of stochastic analysis techniques. A numerical example is given to demonstrate the validity of the proposed estimator.