无线传感器网络中基于量化观测的粒子滤波状态估计

无线传感器网络环境下处理分布式状态估计问题,由于网络中的带宽限制,减少通信成本是非常重要的一个环节,需要将观测值量化后再传送.针对非线性系统的状态滤波问题,本文提出了一种基于量化观测的粒子滤波状态估计算法,并阐述了基于量化观测的状态估计过程.文中分别采用基于均匀量化(UQDPF)和非均匀量化(NUQDPF)观测的分布式粒子滤波算法进行状态估计,通过被动跟踪仿真实例,利用均方根误差(RMSE)比较了误差性能,并且比较了在不同量化级数下的非均匀量化算法的跟踪误差,仿真结果表明,基于非均匀量化观测的粒子滤波器具有更高的跟踪精度,是一种有效的非线性滤波算法.     

带丢包的一致性最小方差滤波

针对传感网路中存在的丢包问题,提出了一种新的基于一致性的滤波方法。对于节点的检测数据丢包,该滤波引入一致性协议提高了网络对于丢包的鲁棒性;对于传感器节点之间的通信丢包,该滤波引入了丢包补偿的方法。通过与已有丢包算法仿真对比,该方法具有更好的估计精度。     

考虑数据量化的改进无模型自适应迭代学习控制算法

针对一类存在数据量化的离散时间单输入单输出非线性系统,提出一种带有编码解码量化机制的无模型自适应迭代学习控制(MFAILC)算法.首先使用伪偏导数将受控非线性系统动态线性化,进而考虑系统输出数据经由均匀量化器进行量化处理的过程,并设计了一种编码解码量化机制,最后基于这种编码解码量化机制提出了一种改进的MFAILC算法.理论上给出了算法的收敛性分析,结果表明,当系统存在数据量化时,所提出的算法仍可保证系统收敛.与已有算法相比,所提算法仅利用较少的输入输出数据,就可以实现跟踪误差的零收敛.仿真进一步验证了算法的有效性.     

离散高阶分布式一致性算法

为了提高分布式一致性算法的收敛速度, 提出了一种离散高阶分布式一致性算法。该算法通过单跳通信, 利用二跳邻接节点的前多步信息来加速分布式一致性算法的收敛速度。对无向通信拓扑下该算法的收敛性能和收敛速度, 以及带通信延时的该算法的收敛性能进行了分析和仿真比较, 结果显示, 该算法在满足条件下能收敛到初始状态的平均值, 与同样利用二跳邻接节点信息的算法相比, 具有通信量小, 收敛速度更快的特点, 但是能容忍的通信延时变小。     

传感器网络中带有一致性策略的分布式H∞滤波:考虑一致性跟踪误差

现有的带有一致性策略的分布式 滤波方法包含两个步骤:与相邻传感器节点交互通信的一致性步骤, 以及本地滤波步骤. 本文分析了一致性跟踪误差对于本地估计误差的影响, 并针对此影响, 提出了新的分布式H∞滤波方法. 当采样周期中一致性迭代次数有限时, 本文提出的方法能够抑制一致性跟踪误差对本地估计误差的影响;当采样周期中一致性迭代次数趋于无穷, 即一致性跟踪误差收敛到零时, 本文提出的分布式算法中的本地滤波就等价于集中式滤波. 仿真表明了本文方法的有效性.     

移动传感器网络节点快速一致性滤波定位

针对移动传感器网络拓扑结构的动态特性,提出了一种快速卡尔曼一致性滤波定位算法。该算法依据Mc- tropolis准则,仅利用通信节点之间的RSSI值快速调整融合步长。在网络拓扑结构未知的情况下,利用卡尔曼一致性 滤波定位算法实现位置求精。仿真结果表明,与Saber算法相比,该方法能够在降低通信量的同时,提高节点的定位 精度,适合移动传感器网络。     

基于转换量测容积卡尔曼滤波器带多普勒量测的目标跟踪算法

针对带多普勒量测的目标跟踪问题,提出一种基于转换量测容积卡尔曼滤波器的序贯滤波目标跟踪算法.对具有量测误差相关性的距离和多普勒量测进行解相关处理,构造出新的解相关量测方程,进而基于贝叶斯方法提出带多普勒量测的序贯处理算法的统一理论框架,实现对位置量测和多普勒量测的序贯滤波.在该理论框架下,提出基于转换量测容积卡尔曼滤波器的序贯滤波目标跟踪算法.该算法先采用转换量测容积卡尔曼滤波器和位置量测对目标状态进行估计,再利用经典容积卡尔曼滤波器对新构造的伪多普勒量测进行量测更新以实现目标跟踪.通过对所提算法的性能分析验证该算法的一致性和收敛性.仿真结果表明,该算法与其他跟踪算法相比,具有更高的跟踪精度.     

基于CKF 的分布式滤波算法及其在目标跟踪中的应用

针对已有基于Sigma点信息滤波的分布式滤波算法,其性能易受参数影响而导致应用范围受限的问题,以容积卡尔曼滤波(CKF)为基础,利用信息滤波和平均一致性理论提出一种分布式CKF算法。该算法在保持分布式滤波优良特性(即可扩展性和对节点故障强鲁棒性)的同时,兼具CKF的高滤波精度和强稳定性。仿真结果表明了所提出算法的有效性,与分布式Unscented卡尔曼滤波(UKF)算法相比,该算法显著提高了目标跟踪的精度和稳定性。     

基于辅助模型的量化控制系统辨识方法

针对具有通信约束的量化控制系统模型, 在采用随机重复性试验测量信息的技术上, 提出了基于辅助模型的量化系统参数辨识方法. 首先分析了在随机重复性试验方法下量化系统的模型特征并给出了分两步辨识的策略.分析表明, 在上述模型里系统具有时变的估计误差, 推导了进行参数辨识所满足的持续激励条件, 并给出了基于辅助模型的多新息量化辨识递推算法. 接着研究了所给出辨识算法的收敛性分析, 得到了系统参数估计误差上界的计算式,最后将方法推广到一类Hammerstein非线性系统量化辨识问题上. 数字仿真验证了该算法及结论     

基于一致性度量属性约简的决策树构建优化研究

针对样本中有无关的、冗余的属性会降低决策树算法的分类精度,本文提出基于一致性度量属性约简后构建决策树的方法。对UCI机器学习数据库中5个两类分类样本离散化后,分别基于粗糙集和一致性度量的属性约简来构建C45和CART决策树,实验表明基于一致性度量属性约简构建的决策树有较高的精度和可行性。     

Discrete-time multi-agent consensus with quantization and communication delays

In this paper, discrete-time multi-agent consensus problem with quantization and communication delays is investigated. A new discrete-time multi-agent consensus model is considered in which each agent can only receive the delayed quantized information from its neighbors. In the presence of quantization and communication delays, it is shown that the multi-agent network can achieve consensus under the connectivity network topology. Simulation examples are also provided to demonstrate the correctness of the theoretical results.     

Quantized Consensus of Multi‐Agent Systems Via Broadcast Gossip Algorithms

Distributed consensus problems of multi‐agent systems on directed networks are studied in this paper. For the communication of agents, it is assumed that only one agent can be selected with a prescribed probability, and it broadcasts its own state to neighbors via quantized communication (for any arbitrary quantization) at each time step. For this kind of communication, the fundamental questions are how to design distributed algorithms and what kinds of network topology together lead to the quantized consensus. A class of broadcast gossip algorithms is proposed, and a necessary and sufficient graphical condition is given to ensure the quantized consensus. In particular, the obtained graphical condition does not require a symmetric network topology, which is weaker than those in some other literature. Several numerical simulations are given to show the effectiveness of the proposed algorithms.     

Reaching consensus in unbalanced networks with coarse information communication

This paper is concerned with the problem of seeking consensus for a network of agents under a fixed or switching directed communication topology. Each agent is modeled as discrete‐time first‐order dynamics and interacts with its neighbors via logarithmically quantized information. We assume that the digraph is not necessarily balanced and, thus, avoiding the double stochasticity requirement for the adjacency matrix. For the case of a fixed topology that is strongly connected, it is shown that the proposed protocol is admissible for arbitrarily coarse logarithmic quantization and the β‐asymptotic weighted‐average consensus is achieved. For the case of a switching topology that is periodically strongly connected, it is shown that the proposed protocol is admissible for arbitrarily coarse quantization and the β‐asymptotic consensus is achieved. Furthermore, for both cases, not only are the convergence rates for consensus specified but also the bounds on the consensus error that highlight their dependence on the sector bound β of the logarithmic quantizer are also provided. Copyright © 2015 John Wiley & Sons, Ltd.     

Adaptive motion coordination of passive systems under quantization effect

In this paper, we consider the quantization effect on an adaptive motion coordination control law for passive multi‐agent systems when the time‐varying reference velocity is only available to the leader. Both logarithmic quantizers and uniform quantizers are investigated. Using tools from nonsmooth analysis and the Barbalat lemma, we prove that with undirected connected communication graphs, coordination problem is still solvable when logarithmic quantizers are used and the parameter convergence is also guaranteed when the basis functions satisfy certain persistently exciting property. When uniform quantizers are used, we prove that practical consensus can be achieved, and both the bounds of the consensus error and the velocity tracking error are proportional to the parameter of the quantizer. Simulation examples are provided to illustrate the results. Copyright © 2014 John Wiley & Sons, Ltd.     

Distributed Average Consensus in Multi-agent Networks with Limited Bandwidth and Time-delays

This paper studies the distributed average consensus problem of multi-agent digital networks with time-delays. For the network, each agent can only exchange symbolic data with its neighbours. We provide a distributed average consensus protocol which uses the quantized and time-delay information. We consider two types of dynamic encoders/decoders in our protocol. One uses inverse proportion function as scaling function, while the other uses exponential function as scaling function. All agents can reach average consensus asymptotically with our protocol. Furthermore, we show that the average consensus protocol is robust to finite symmetric time-delays, but is sensitive to asymmetric time-delays that will destroy the average consensus of the networks. Finally, simulations are presented to illustrate validity of our theoretical results.      

An event‐triggered approach to quantized resilient consensus

We consider an event‐triggered update scheme for the problem of multiagent consensus in the presence of faulty and malicious agents within the network. In particular, we focus on the case where the agents take integer (or quantized) values. To keep the regular agents from being affected by the behavior of faulty agents, algorithms of the mean subsequence reduced type are employed, where neighbors taking extreme values are ignored in the updates. Different from the real‐valued case, the quantized version requires the update rule to be randomized. We characterize the error bound on the achievable level of consensus among the agents as well as the necessary structure for the network in terms of the notion of robust graphs. We verify via a numerical example the effectiveness of the proposed algorithms.     

Distributed edge event‐triggered consensus protocol of multi‐agent systems with communication buffer

This paper proposes a distributed edge event‐triggered (DEET) scheme of multi‐agent systems via a communication buffer to reduce unnecessary update of controllers induced by fast information transmission. This edge scheme avoids a synchronous phenomenon in node event‐triggered mechanism, in which the triggering of one agent activates information transmission of all edges linked with this agent. Hence, the node event‐triggered scheme leads to unnecessary update of control protocols while the DEET provides a new approach without constrains on synchronous phenomenon of edge information exchange. That is, the communication on each edge is independent with other edges. In addition, we investigate another case where edge information transmission is subject to quantization and a quantized edge event‐triggered control protocol is proposed. Note that such a quantized protocol guarantees asymptotical consensus instead of bounded consensus in most of the existing literature. Meanwhile, both DEET and quantized edge event‐triggered schemes have nontrivial properties of excluding Zeno behavior. Furthermore, an algorithm is provided to avoid continuous event detection; hence, the communication traffic of the whole network is reduced significantly. Copyright © 2016 John Wiley & Sons, Ltd.     

考虑量化的多智能体系统数据驱动双向一致性控制

针对未知动力学模型非线性离散时间多智能体系统,在信息传递过程中的数据量化问题,以及智能体之间的合作与竞争关系,提出了一种数据驱动控制算法,实现了多智能体系统的双向一致性跟踪控制.首先,利用紧凑形动态线性化(CFDL)方法,将未知动力学模型的非线性智能体转化为含有时变参数的数据模型,并通过设计性能指标函数获得时变参数的估...     

Output‐feedback adaptive consensus tracking control for a class of high‐order nonlinear multi‐agent systems

In this paper, an output‐feedback adaptive consensus tracking control scheme is proposed for a class of high‐order nonlinear multi‐agent systems. The agents are allowed to have unknown parameters, unknown nonlinearities, and input quantization simultaneously. The desired trajectory to be tracked is available for only a subset of agents, and only the relative outputs and the quantized inputs need to be measured or transmitted as signal exchange among neighbors regardless of the system order. By introducing a kind of high‐gain K‐filters and a smooth function, the effect among agents caused by the unknown nonlinearities is successfully counteracted, and all closed‐loop signals are proved to be globally uniformly bounded. Moreover, it is shown that the tracking errors converge to a residual set that can be made arbitrarily small. Simulation results on robot manipulators are presented to illustrate the effectiveness of the proposed scheme. Copyright © 2017 John Wiley & Sons, Ltd.     

一种基于信噪比加权共识的合作频谱感知算法

以基于平均共识的分布式合作频谱感知算法为基础,提出了一种基于信噪比加权共识的合作频谱感知算法.该算法在平均共识收敛过程中引入能够反映用户的平均信噪比的权重,并以平均共识收敛值为参考确定相应的权重.在不需要信噪比信息的条件下,算法实现了认知Ad Hoc网络中基于信噪比的频谱感知信息加权共识.分析和仿真表明,在用户具有不同的平均信噪比的情况下,该算法克服了基于平均共识的分布式合作频谱感知算法无法体现用户信噪比差异的缺陷,提高了频谱感知的性能.