基于神经网络的非线性系统的间接自适应控制

针对一类不确定非线性动态系统 ,利用多层神经网络系统MNNs的逼近能力 ,提出了一种间接鲁棒自适应神经网络控制器的设计方案。该方案不仅能够保证闭环系统的所有信号有界 ,而且理论分析证明了闭环系统的跟踪误差渐近收敛到零。仿真试验表明本控制算法是有效的。     

Decentralized direct adaptive neural network control for a class of interconnected systems

1 .INTRODUCTIONA number of interconnected systems found in theworld, such as electric power systems ,industrymanipulators and computer networks , are oftencomposed of a set of subsystems . A centralizedcontrol strategy for the requirement of a large a-mount of information exchange between the sub-systems . A decentralized control method, devel-oped based only on local measurements ,is oftenpreferable . At present ,there have been many re-search results for adaptive control of interconnec-…     

直接鲁棒自适应模糊控制

针对一类不确定非线性系统 ,基于监督控制方案并利用第二类模糊系统的逼近能力 ,提出了一种直接鲁棒自适应模糊控制器设计的新方案。该方案通过引入最优逼近误差的自适应补偿项来消除建模误差的影响 ,从而在稳定性分析中取消了要求逼近误差平方可积或逼近误差的上确界已知的条件。理论分析证明了闭环控制系统是全局稳定的 ,跟踪误差收敛到零。仿真结果表明了该方法的有效性。     

增益符号未知的鲁棒自适应模糊控制

针对一类具有函数控制增益符号未知的不确定非线性系统,根据滑模控制原理和模糊系统的逼近能力,提出了一种自适应模糊控制器设计新方案。该方案利用Nussbaum函数的性质,取消了常用的函数控制增益符号已知这一条件。并通过引入最优逼近误差的自适应补偿项来消除建模误差和参数估计误差的影响。理论分析证明了闭环系统的有界性和跟踪误差收敛到零。仿真结果表明了该方法的有效性。     

航天器姿态的神经网络动态逆控制

针对航天器姿态系统,提出了一种基于自适应神经网络动态逆的控制算法。该算法针对滚转、俯仰和偏航三个姿态子系统,设计了两组神经网络:第一组是BP网络,用来逼近三个姿态通道的非线性项,可获得姿态逆模型;第二组是非线性自适应神经网络,用于在线实时地补偿逆模型存在的误差和外加干扰。详细分析了非线性自适应神经网络的拓扑结构、学习规则和调整算法。给出了应用该算法的具体实例,通过仿真实验证明该算法的有效性。     

超机动飞行的非线性鲁棒自适应控制系统研究

针对飞机模型中存在气动参数不确定性以及外界干扰等影响因素,设计了一种超机动飞行的非线性鲁棒自适应控制系统。控制系统设计过程中,模型不确定性和外界干扰由RBF神经网络在线补偿,控制律及神经网络权值自适应律由反步法得到。解决了系统中控制增益矩阵未知,同时存在外界干扰情况下的鲁棒飞行控制系统设计,并证明了闭环系统所有信号有界,系统跟踪误差和神经网络权值估计误差指数收敛到有界紧集内。对所研究的飞行控制系统进行了过失速Herbst机动仿真,结果验证了该系统在过失速机动条件下具有良好的控制性能。     

含非线性输入的非线性系统自适应模糊控制

讨论一类含非线性输入的非线性系统的自适应模糊控制问题。首先运用隐函数存在定理证明系统的理想控制器的存在性,利用I型模糊逻辑系统对该理想控制器进行在线逼近,提出了一种具有监督器的自适应模糊滑模控制器设计的新方案。该方案通过直接自适应模糊控制器与监督控制器的适当切换,保证了闭环系统的稳定性,由此确定出建模的有界区域,而且通过引入最优逼近误差的自适应补偿项,保证跟踪误差收敛到零。仿真结果表明了该方法的有效性。     

一类非线性系统的模型参考自适应模糊控制

针对一类不确定非线性系统 ,基于一种修改的监督控制方案并利用广义多线性模糊逻辑系统的逼近能力 ,提出了一种模型参考自适应模糊控制器设计的新方案。该方案利用参考模型作为性能指标 ,自适应调节模糊控制器的规则库 ,并通过引入最优逼近误差的自适应补偿项来消除建模误差的影响 ,不但能保证闭环系统稳定 ,而且可使跟踪误差收敛到零。仿真结果表明了该方法的有效性     

不确定多时滞系统动态自适应神经网络控制

针对一类具有不确定多时滞的非线性系统,提出一种由时滞补偿器和动态结构自适应神经网络所构成的控制器.通过设计时滞补偿器中的参数自适应调节规律来消除多时滞对控制输入的影响,再引入动态自适应神经网络,利用其隐层神经元个数可以随着逼近误差的增大而在线增加的特点,获得满意的逼近精度,提高控制性能.最后,对时滞混沌系统进行仿真,表明该方法的有效性.     

Cooperative Neuro Adaptive Control of Leader Following Uncertain Multi-Agent Systems with Unknown Hysteresis and Dead-Zone

In this paper, a cooperative adaptive control of leader-following uncertain nonlinear multiagent systems is proposed. The communication network is weighted undirected graph with fixed topology. The uncertain nonlinear model for each agent is a higher-order integrator with unknown nonlinear functions, unknown disturbances and unknown input actuators. Meanwhile, the gains of input actuators are unknown nonlinear functions with unknown sign. Two most common behaviors of input actuators in practical applications are hysteresis and dead-zone. In this paper, backlash-like hysteresis and dead-zone are used to model the input actuators. Using universal approximation theorem proved for neural networks, the unknown nonlinear functions are tackled. The unknown weights of neural networks are derived by proposing appropriate adaptive laws. To cope with modeling errors and disturbances an adaptive robust structure is proposed. Considering Lyapunov synthesis approach not only all the adaptive laws are derived but also it is proved that the closed-loop network is cooperatively semi-globally uniformly ultimately bounded(CSUUB). In order to investigate the effectiveness of the proposed method, it is applied to agents modeled with highly nonlinear mathematical equations and inverted pendulums. Simulation results demonstrate the effectiveness and applicability of the proposed method in dealing with both numerical and practical multi-agent systems.     

基于回归神经网络的滑模跟踪器设计

基于回归神经网络的在线辨识 ,为实现非线性系统自适应控制提供了一种很有应有前景的方法。本文基于具有线性输入特性的神经网络 ,提出了使系统辨识误差具有指数收敛特性的学习算法。为了得到尽可能普遍的控制律 ,本文运用滑模技术设计出控制信号 ,用其补偿神经网络模型与非线性系统之间的匹配误差。以此为基础 ,提出一种新的基于回归神经网络的间接自适应控制方案。仿真结果表明 ,本控制方案具有较好的跟踪性能     

Neuromorphic Continuous-Time State Space Pole Placement Adaptive Control

Abstract: A neuromorphic continuous-time state space pole assignment adaptive controller is proposed, which is particularlyappropriate for controlling a large-scale time-variant state-space model due to the parallely distributed nature ofneurocomputing. In our approach, Hopfield neural network is exploited to identify the parameters of a continuous-timestate-space model, and a dedicated recurrent neural network is designed to compute pole placement feedback control law inreal time. Thus the identification and the control computation are incorporated in the closed-loop, adaptive, real-timecontrol system. The merit of this approach is that the neural networks converge to their solutions very quickly andsimultaneously.     

一种基于小波网络模型参考自适应控制系统的设计与应用

根据神经网络内模控制的思想,提出了一种基于小波网络的模型参考自适应控制方法。该方法选用两个小波神经网络,分别作为系统模型的辨识器和控制器。此外,为了减小系统误差,在控制器和辨识器之间加入一个小波神经网络,作为非线性自适应滤波器。仿真结果和应用实例表明了该方法的有效性和合理性。     

Adaptive functional link network control of near-space vehicles with dynamical uncertainties

The control law design for a near-space hypersonic vehicle (NHV) is highly challenging due to its inherent nonlinearity, plant uncertainties and sensitivity to disturbances. This paper presents a novel functional link network (FLN) control method for an NHV with dynamical thrust and parameter uncertainties. The approach devises a new partially-feedback-functional-link-network (PFFLN) adaptive law and combines it with the nonlinear generalized predictive control (NGPC) algorithm. The PFFLN is employed for approximating uncertainties in flight. Its weights are online tuned based on Lyapunov stability theorem for the first time. The learning process does not need any offline training phase. Additionally, a robust controller with an adaptive gain is designed to offset the approximation error. Finally, simulation results show a satisfactory performance for the NHV attitude tracking, and also illustrate the controller's robustness.     

Application of Neurocomputing in Adaptive Control of Large-Scale Aerospace Systems

We are engaged in solving two difficult problems in adaptive control of the large-scale time-variant aerospace system. One is parameter identification of time-variant continuous-time state-space modei; the other is how to solve algebraic Riccati equation (ARE) of large order efficiently. In our approach, two neural networks are employed to independently solve both the system identification problem and the ARE associated with the optimal control problem. Thus the identification and the control computation are combined in closed-loop, adaptive, real-time control system . The advantage of this approach is that the neural networks converge to their solutions very quickly and simultaneously.     

Adaptive integral dynamic surface control based on fully tuned radial basis function neural network

An adaptive integral dynamic surface control approach based on fully tuned radial basis function neural network (FTRBFNN) is presented for a general class of strict-feedback nonlinear systems, which may possess a wide class of uncertainties that are not linearly parameterized and do not have any prior knowledge of the bounding functions. FTRBFNN is employed to approximate the uncertainty online, and a systematic framework for adaptive controller design is given by dynamic surface control. The control algorithm has two outstanding features, namely, the neural network regulates the weights, width and center of Gaussian function simultaneously, which ensures the control system has perfect ability of restraining different unknown uncertainties and the integral term of tracking error introduced in the control law can eliminate the static error of the closed loop system effectively. As a result, high control precision can be achieved. All signals in the closed loop system can be guaranteed bounded by Lyapunov approach. Finally, simulation results demonstrate the validity of the control approach.     

基于神经网络的自由漂浮空间机器人鲁棒控制

针对带有模型误差及外界扰动的自由漂浮空间机器人轨迹跟踪问题,提出了一种基于神经网络的自适应鲁棒控制策略。采用对神经网络状态空间进行划分后与滑模变结构结合的控制器,对不确定非线性进行自适应学习,逼近误差作为外部干扰由鲁棒控制器消除。该方法从整个闭环系统的稳定性出发,利用H_∞理论设计的鲁棒控制器及神经网络权值的在线调整规则保证了系统的稳定性,并能使系统L2增益小于给定的指标,具有较好的控制精度及动态特性。仿真分析进一步证明了该自适应鲁棒控制算法的有效性。     

一种改进的自适应模糊滑模大包线飞行控制方法

提出了一种改进的自适应模糊滑模大包线飞行控制方法。该方法以经模拟退火粒子群算法优化的小波神经网络实现非线性模型的逆,能够更加细致地逼近非线性模型,并针对自适应控制的鲁棒性与瞬态性能差的缺点,将滑模控制与自适应控制相结合共同补偿逆误差,提高了自适应控制的鲁棒性与瞬态性。仿真结果表明:所设计的自适应模糊滑模大包线飞行控制器具有优良的控制性能。     

一类径向基神经网络干扰观测器轨迹线性化控制

利用径向基神经网络(RBFNN)的逼近能力,研究了基于径向基神经网络干扰观测器(RDO)的鲁棒自适应轨迹线性化控制(TLC)策略,以解决空天飞行器复杂飞行条件下系统不确定及干扰的控制问题。分析了系统存在不确定性时轨迹线性化控制方法性能降低甚至失效的原因,设计了自适应调节律,并采用Lyapunov方法严格证明了在该自适应调节律作用下闭环系统所有误差信号最终有界。仿真结果表明,较当前TLC方法的控制性能,新方案在空天飞行器系统上具有更优异的控制性能和鲁棒性。     

多变量非线性系统的直接自适应模糊预测控制

对一类多变量非线性系统提出了直接自适应模糊预测控制方法,此方法首先对被控对象提出了线性时变子模型加非线性子模型的预测模型,然后直接利用模糊系统设计预测控制器,并基于时变增益自适应律对控制器中的未知向量和逼近误差估计值进行自适应调整。证明了此方法可使跟踪误差收敛到原点的一个邻域内,仿真结果验证了此方法的有效性。