Flocking of mobile agents with fuzzy controller

Based on individual local observed information of the swarm, a fuzzy logical controller （FLC） of the attraction/repulsion function is built. Combining with the rate consensus algorithm, the swarming of the separation, cohesion and alignment is achieved. Based on the optimal controller theories, a capability target is used to measure the states of the system. In the computer simulation, it is shown that the aggregation of the multi-agent systems can get consensus following a mobile leader from the individual local perceived information.     

二阶多智能体系统量化蜂拥控制

针对在多智能体系统的通信网络中需要对交换信息进行量化的客观情况,研究基于量化信息的二阶多智能体系统蜂拥控制问题。首先,假设多智能体系统采用一致量化器对速度和位置信息进行量化,并且有一个虚拟领导者沿着固定方向匀速运动;然后,设计基于量化信息的多智能体蜂拥控制输入,并利用非平滑系统的Lyapunov稳定性判据和不变集原理证明系统的稳定性;最后,利用Matlab对多智能体系统在二维平面上的蜂拥运动进行仿真实验,仿真结果验证了理论分析的正确性。     

多智能体系统编队控制相关问题研究综述

首先梳理了编队控制研究的脉络, 介绍了3 种经典的编队控制方法, 即跟随领航者法、基于行为法和虚拟结构法的研究思想; 接着综述了近年来发展的, 包容了上述3 种方法且基于图论的编队控制理论的研究成果, 包括多智能体系统图论的建模, 基于代数图论和基于刚性图论的多智能体编队控制律设计、编队构型变换等方面的研究成果; 然后从图论结果出发, 回顾了与编队控制密切相关的一致性、聚集/同向、群集/蜂拥和包络控制的最新进展; 最后, 为了促进多智能体系统在实际中的应用, 指出了多智能体编队控制研究中有待解决的若干问题.

     

Connectivity Preserving Flocking without Velocity Measurement

In this paper, we address the design of a decentralized controller for connectivity‐preserving flocking, where each agent only can access to the position information of the agents within its sensing zone. An output vector, based on the position information alone, is constructed to replace the role of velocity, and some bounded attractive and repulsive forces are integrated together to design the controller. We prove that the controller not only synchronizes all agents in a stable formation, but also enables collision avoidance and connectivity preserving all of the time, when the initial condition meets certain requirements. Moreover, a leader‐follower method is used to guide the group to a desired direction, where the followers can sense the leader only if the distance between them is less than the communication radius.     

活性炭纤维静电植绒工艺研究

利用静电植绒技术,制成具有空气净化性能的活性炭纤维静电植绒制品．将植入绒毛量作为统一的评价标准,通过改变极板电压、极板距离以及植绒时间等3个工艺参数,分析其对单位面积植入绒毛量的影响．实验结果表明,单位面积植入绒毛量随着极板电压和植绒时间的增加而增加,随着极板距离的增加而减少．     

群体系统蜂拥控制理论及应用研究进展*

群体系统蜂拥控制研究近年来受到了国内外众多领域高度重视,目前主要借鉴生物学领域关于群体蜂拥行为的研究成果,集中在个体之间交互形成的网络拓扑结构已知条件下的控制问题,分析了群体系统蜂拥控制研究的几个发展阶段,总结了群体蜂拥行为的理论研究趋势及其在工程领域中的应用研究现状,并指出了今后的发展方向。     

Flocking Algorithms in Networks with Directed Switching Velocity Interaction Topologies

This paper studies flocking algorithms for multi‐agent systems with directed switching velocity interaction topologies. It is assumed that the position information of each agent is available for agents within its neighborhood region of radius r, however, they communicate the velocity information between each other through unidirectional links modeled by a particular class of directed topologies. A new energy function is defined to analyze the global stability and a sufficient condition is derived for asymptotic flocking in the face of switching topologies. The proposed control strategy guarantees the achievement of desired formations, while avoiding collisions among agents. It also ensures velocity agreement under suitable conditions in a variety of real networks with greatly reduced velocity data exchange. Moreover, a leader‐follower framework is formulated for the described class of interaction topologies and it is shown that a more relaxed condition is required to achieve the desired performance. Finally, several simulations are performed to illustrate and confirm the theoretical results obtained.     

基于有向网络的智能群体群集运动控制

对具有二次积分动态的智能群体(flock/swarm),在有向网络取得群集运动(flocking/swarming)进行了研究．提出了一个分散控制方法对智能群体进行分散控制．用有向图模型表示智能体之间的相互作用及通信关系．对固定的网络拓扑,控制互连拓扑是固定的,时不变的,运用传统的LaSalle不变集原理,代数图论的有关技巧进行了稳定性分析,并得到以下主要结论:i)智能群体速度方向渐进收敛,并保持方向一致;ii)智能群体速度大小渐进收敛,并保持大小相等;iii)有邻接关系的智能体(Agent)之间没有碰撞发生;iv)智能群体的势场函数被最小化．理论分析显示,有向图的弱连通性及一种称为平衡图的有向图在系统的稳定性分析中扮演着关键角色．最后,给出了一个仿真例子对理论结果进行了验证．     

大规模群体系统的现状研究

大规模群体系统源于自然界中广泛存在的群体现象，其重要性不仅在于它可以描述和解释大多数生物群体的群体协调行为和自组织现象，而且大规模群体系统具有明显的工程应用背景。因此，大规模群体系统的理论与应用正受到众多学科的高度重视。系统地综述了大规模群体系统的发展历程、理论研究和应用研究，并进行了分析和评论。     

基于新型RBF神经网络的中速磨煤机建模

提出了一种结合免疫聚类和免疫进化规划的径向基函数(RBF)神经网络设计方法,该方法采用人工免疫聚类机制根据输入数据集合自适应确定RBF神经网络隐节点的数量和数据中心位置,并通过提取核函数宽度作为疫苗和免疫选择操作确定网络输出层的权值。利用RBF神经网络设计方法建立了中速磨煤机出口温度和入口空气流量的RBF神经网络模型,仿真结果表明该模型具有较高的精度。