基于神经网络的模糊自适应PID控制方法

提出一种基于BP神经网络的模糊自适应PID控制器。该控制器综合模糊控制、神经网络与PID调节各自的优点，既具有模糊控制的简单和有效的非线性控制作用，又具有神经网络的学习和适应能力，同时具备PID控制的广泛适应性，仿真实验表明该控制器对模型、环境具有较好的适应能力和较强的鲁棒性。     

Dynamic structure adaptive neural fuzzy control for MIMO uncertain nonlinear systems

This paper proposes a novel dynamic structure neural fuzzy network (DSNFN) to address the adaptive tracking problems of multiple-input-multiple-output (MIMO) uncertain nonlinear systems. The proposed control scheme uses a four-layer neural fuzzy network (NFN) to estimate system uncertainties online. The main feature of this DSNFN is that it can either increase or decrease the number of fuzzy rules over time based on tracking errors. Projection-type adaptation laws for the network parameters are derived from the Lyapunov synthesis approach to ensure network convergence and stable control. A hybrid control scheme that combines the sliding-mode control and the adaptive bound estimation control with different weights improves system performance by suppressing the influence of external disturbances and approximation errors. As the employment of the DSNFN, high-quality tracking performance could be achieved in the system. Furthermore, the trained network avoids the problems of overfitting and underfitting. Simulations performed on a two-link robot manipulator demonstrate the effectiveness of the proposed control scheme.     

Adaptive control of nonlinear PID-based analog neural networks for a nonholonomic mobile robot

An adaptive controller of nonlinear PID-based analog neural networks is developed for the velocity- and orientation-tracking control of a nonholonomic mobile robot. A superb mixture of a conventional PID controller and a neural network, which has powerful capability of continuously online learning, adaptation and tackling nonlinearity, brings us the novel nonlinear PID-based analog neural network controller. It is appropriate for a kind of plant with nonlinearity uncertainties and disturbances. Computer simulation for a differentially driven nonholonomic mobile robot is carried out in the velocity- and orientation-tracking control of the nonholonomic mobile robot. The effectiveness of the proposed control algorithm is demonstrated through the simulation experiment, which shows its superior performance and disturbance rejection.     

Adaptive Fourier Series Neural Network PID Controller

In this work, an Adaptive Neural Networks PID controller structure, called Adaptive Fourier Series Neural Networks PID controller (AFSNNPID), is developed. The main objective is to obtain a simple controller for nonlinear systems that can be tuned online to reject perturbations effect and compensate the system parameters variation. Due to its simple architecture and very attractive proprieties, the Fourier Series Neural Network (FSNN) is used to online adjust the parameters of the PID controller. Furthermore, using the delta-rule algorithm, the adaptation dynamics of the FSNN is globally stable. The design procedure of the proposed controller and the stability analysis of the closed loop system using the small gain theorem are given. To assess the effectiveness of the proposed control scheme, the control of a 3-DOF robot arm manipulator is considered and a comparative study, using the adaptive neural network PID controller and the particle swarm optimization based PID controller, is carried out. The obtained results, through the experimental study, indicate that the AFSNNPID controller presents better control performance than the other controllers.

     

Fuzzy PID Control of a Five DOF Robot Arm

This paper studies the application of fuzzy logic control on a five degrees of freedom (DOF) robot arm, the Maker 100 of U.S. Robots. The elaboration of the fuzzy control laws is based on two structures of coupled rules fuzzy PID controllers. The fuzzy PID controllers are numerically simulated and the simulation results confirm the success of the fuzzy PID control in trajectory tracking problems. Seeking a performance optimization, a systematic study of the choice of tuning parameters of the controllers is done. The success of the proposed fuzzy control law is again affirmed by a comparative evaluation with respect to the computed torque control method and the direct adaptive control method, the last two controls being also numerically implemented using the same dynamic model of the robot arm.     

非线性离散系统的自适应PID控制器

针对非线性离散系统设计了利用TSK(Takagi Sugeno Kang)模糊模型的自适应PID控制器。利用模糊模型预测控制信号误差,通过控制信号误差自适应PID控制器参数。比较系统输出和模糊模型输出自适应模糊模型的参数。该方法可以弥补系统参数的模糊性、数学模型的模型误差和系统参数的变化。非线性离散系统的仿真实验验证了所设计的自适应PID控制器对非线性离散系统控制的有效性。     

Adaptive Fuzzy Neural Network Control Design via a T–S Fuzzy Model for a Robot Manipulator Including Actuator Dynamics

This paper focuses on the development of adaptive fuzzy neural network control (AFNNC), including indirect and direct frameworks for an $n$-link robot manipulator, to achieve high-precision position tracking. In general, it is difficult to adopt a model-based design to achieve this control objective due to the uncertainties in practical applications, such as friction forces, external disturbances, and parameter variations. In order to cope with this problem, an indirect AFNNC (IAFNNC) scheme and a direct AFNNC (DAFNNC) strategy are investigated without the requirement of prior system information. In these model-free control topologies, a continuous-time Takagi–Sugeno (T–S) dynamic fuzzy model with online learning ability is constructed to represent the system dynamics of an $n$-link robot manipulator. In the IAFNNC, an FNN estimator is designed to tune the nonlinear dynamic function vector in fuzzy local models, and then, the estimative vector is used to indirectly develop a stable IAFNNC law. In the DAFNNC, an FNN controller is directly designed to imitate a predetermined model-based stabilizing control law, and then, the stable control performance can be achieved by only using joint position information. All the IAFNNC and DAFNNC laws and the corresponding adaptive tuning algorithms for FNN weights are established in the sense of Lyapunov stability analyses to ensure the stable control performance. Numerical simulations and experimental results of a two-link robot manipulator actuated by dc servomotors are given to verify the effectiveness and robustness of the proposed methodologies. In addition, the superiority of the proposed control schemes is indicated in comparison with proportional–differential control, fuzzy-model-based control, T–S-type FNN control, and robust neural fuzzy network control systems.      

Stable fuzzy neural tracking control of a class of unknown nonlinear systems based on fuzzy hierarchy error approach

In this paper, a stable fuzzy neural tracking control of a class of unknown nonlinear systems based on the fuzzy hierarchy approach is proposed. The adaptive fuzzy neural controller is constructed from the fuzzy neural network with a set of fuzzy rules. The corresponding network parameters are adjusted online according to the control law and update law for the purpose of controlling the plant to track a given trajectory. A stability analysis of the unknown nonlinear system is discussed based on the Lyapunov principle. In order to improve the convergence of the nonlinear dynamical systems, a fuzzy hierarchy error approach (FHEA) algorithm is incorporated into the adaptive update and control scheme. The simulation results for an unstable nonlinear plant demonstrate the control effectiveness of the proposed adaptive fuzzy neural controller and are consistent with the theoretical analysis.     

Adaptive robust fuzzy control for path tracking of a wheeled mobile robot

This paper proposes an adaptive robust fuzzy control scheme for path tracking of a wheeled mobile robot with uncertainties. The robot dynamics including the actuator dynamics is considered in this work. The presented controller is composed of a fuzzy basis function network (FBFN) to approximate an unknown nonlinear function of the robot complete dynamics, an adaptive robust input to overcome the uncertainties, and a stabilizing control input. The stability and the convergence of the tracking errors are guaranteed using the Lyapunov stability theory. When the controller is designed, the different parameters for two actuator models in the dynamic equation are taken into account. The proposed control scheme does not require the accurate parameter values for the actuator parameters as well as the robot parameters. The validity and robustness of the proposed control scheme are demonstrated through computer simulations. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

基于自适应神经网络的不确定非线性系统的模糊跟踪控制

提出了一种基于模糊模型和自适应神经网络的跟踪控制方法.在系统具有未知不确定非线性特性的情况下,首先利用T＿S模糊模型对系统的已知特性进行近似建模,对基于模糊模型的模糊H∞跟踪控制律进行输出跟踪控制.并在此基础上,进一步采用RBF神经网络完全自适应控制,通过在线自适应调整RBF神经网络的权重、函数中心和宽度,从而有效地消除系统的未知不确定性和模糊建模误差的影响,保证了非线性闭环系统的稳定性和系统的H∞跟踪性能,而不要求系统的不确定项和模糊建模误差满足任何匹配条件或约束.最后,将所提出的方法应用到一非线性混沌系统,仿真结果表明了所提出的方案不仅能够有效地稳定该混沌系统,而且能使系统输出跟踪期望输出.     

Wavelet neural adaptive proportional plus conventional integral-derivative controller design of SSSC for transient stability improvement

Although the PI or PID (PI/PID) controllers have many advantages, their control performance may be degraded when the controlled object is highly nonlinear and uncertain; the main problem is related to static nature of fixed-gain PI/PID controllers. This work aims to propose a wavelet neural adaptive proportional plus conventional integral-derivative (WNAP+ID) controller to solve the PI/PID controller problems. To create an adaptive nature for PI/PID controller and for online processing of the error signal, this work subtly employs a one to one offline trained self-recurrent wavelet neural network as a processing unit (SRWNN-PU) in series connection with the fixed-proportional gain of conventional PI/PID controller. Offline training of the SRWNN-PU can be performed with any virtual training samples, independent of plant data, and it is thus possible to use a generalized SRWNN-PU for any systems. Employing a SRWNN-identifier (SRWNNI), the SRWNN-PU parameters are then updated online to process the error signal and minimize a control cost function in real-time operation. Although the proposed WNAP+ID is not limited to power system applications, it is used as supplementary damping controller of static synchronous series compensator (SSSC) of two SSSC-aided power systems to enhance the transient stability. The nonlinear time-domain simulation and system performance characteristics in terms of ITAE revealed that the WNAP+ID has more control proficiency in comparison to PID controller. As additional simulations, the features of the proposed controller are compared to those of the literature while some of its promising features like its fast noise-rejection ability and its high online adapting ability are also highlighted.     

模糊小波基神经网络的机器人轨迹跟踪控制

提出一种模糊神经网络控制器并用于机器人轨迹跟踪控制.这种模糊神经网络利用了小波基函数作为隶属函数,可在线根据误差调整隶属函数的形状,使模糊神经网络具有更强的学习和适应能力.仿真与实验结果表明这种网络能很好的用于机器人的轨迹跟踪控制,具有很好的性能.     

基于模糊模型和神经网络的多时滞不确定非线性系统的鲁棒H∞控制

针对一类具有多时滞的不确定非线性系统,提出了一种基于模糊模型和神经网络的组 合控制方法.利用具有多时滞的模糊T-S模型对系统进行近似建模并给出基于线性矩阵不等式 (LMI)的模糊H∞控制律.提出完全自适应RBF神经网络控制方法,通过在线自适应调整RBF 神经网络的权重、函数中心和宽度,来对消系统的未知不确定性和模糊建模误差的影响,不要求 系统的不确定项和模糊建模误差满足任何匹配条件或约束,并证明了闭环系统的稳定性.最后, 将所提出的方法应用到一具有多时滞的非线性混沌系统,仿真结果表明了该方法的有效性.     

Fuzzy logic based adaptive admittance control of a redundantly actuated ankle rehabilitation robot

Ankle rehabilitation robots have recently attracted great attention since they provide various advantages in terms of rehabilitation process from the viewpoints of patients and therapists. This paper presents development and evaluation of a fuzzy logic based adaptive admittance control scheme for a developed 2-DOF redundantly actuated parallel ankle rehabilitation robot. The proposed adaptive admittance control scheme provides the robot to adapt resistance/assistance level according to patients' disability level. In addition, a fuzzy logic controller (FLC) is developed to improve the trajectory tracking ability of the rehabilitation robot subject to external disturbances which possibly occur due to human-robot interaction. The boundary scales of membership functions of the FLC are tuned using cuckoo search algorithm (CSA). A classical proportional-integral-derivative (PID) controller is also tuned using the CSA to examine the performance of the FLC. The effectiveness of the adaptive admittance control scheme is observed in the experimental results. Furthermore, the experimental results demonstrate that the optimized FLC significantly improves the tracking performance of the ankle rehabilitation robot and decreases the steady-state tracking errors about 50% compared to the optimized PID controller. The performances of the developed controllers are evaluated using common error based performance indices indicating that the FLC has roughly 50% better performance than the PID controller.     

Adaptive explicit force control of position-controlled manipulators

This article presents a new adaptive outer-loop approach for explicit force regulation of position-controlled robot manipulators. The strategy is computationally simple and does not require knowledge of the manipulator dynamic model, the inner-loop position controller parameters, or the environment. It is shown that the control strategy guarantees global uniform boundedness of all signals and convergence of the position/force regulation errors to zero when applied to the full nonlinear robot dynamic model. If bounded external disturbances are present, a slight modification to the control scheme ensures that global uniform boundedness of all signals is retained and that arbitrarily accurate stabilization of the regulation errors can be achieved. Additionally, it is shown that the adaptive controller is also applicable to robotic systems with PID inner-loop position controllers. Computer simulation results are given for a Robotics Research Corporation (RRC) Model K-1207 redundant arm and demonstrate that accurate and robust force control is achievable with the proposed controller. Experimental results are presented for the RRC Model K-1207 robot and confirm that the control scheme provides a simple and effective means of obtaining high-performance force control. © 1996 John Wiley & Sons, Inc.     

New methodology for analytical and optimal design of fuzzy PIDcontrollers

Describes a methodology for the systematic design of fuzzy PID controllers based on theoretical fuzzy analysis and, genetic-based optimization. An important feature of the proposed controller is its simple structure. It uses a one-input fuzzy inference with three rules and at most six tuning parameters. A closed-form solution for the control action is defined in terms of the nonlinear tuning parameters. The nonlinear proportional gain is explicitly derived in the error domain. A conservative design strategy is proposed for realizing a guaranteed-PID-performance (GPP) fuzzy controller. This strategy suggests that a fuzzy PID controller should be able to produce a linear function from its nonlinearity tuning of the system. The proposed PID system is able to produce a close approximation of a linear function for approximating the GPP system. This GPP system, incorporated with a genetic solver for the optimization, will provide the performance no worse than the corresponding linear controller with respect to the specific performance criteria. Two indexes, linearity approximation index (LAI) and nonlinearity variation index (NVI), are suggested for evaluating the nonlinear design of fuzzy controllers. The proposed control system has been applied to several first-order, second-order, and fifth-order processes. Simulation results show that the proposed fuzzy PID controller produces superior control performance to the conventional PID controllers, particularly in handling nonlinearities due to time delay and saturation     

基于模糊小波神经网络的提升机恒减速制动系统的研究

针对提升机恒减速制动系统采用常规PID控制方式、模糊控制方式存在控制效果差的问题,提出了一种基于模糊小波神经网络的提升机恒减速制动系统的设计方案。该系统采用小波基函数作为模糊隶属函数,利用神经网络的自学习能力和小波基良好的局部特性来增强模糊控制的自适应能力,并采用遗传算法对小波基函数的平移、伸缩因子以及控制器的连接权值进行训练,使网络参数达到全局最优。Matlab仿真结果表明,该系统具有良好的动态特性和较高的控制精度。     

模糊自适应在机器人行走平衡控制中的应用

针对多变量、非线性的两轮机器人系统的行走平衡控制问题,提出一种基于Backstepping（反推）方法和PID的控制策略。该策略在Backstepping控制器中加入模糊自适应部分,利用模糊系统逼近Backstepping设计过程中的未知非线性函数,模糊系统中的参数基于自适应律调整,解决了Backstepping控制器中因含有未知参数难以实现的困难,避免了两轮机器人系统不满足严格三角结构的问题。针对两轮机器人的仿真实验结果表明：采用设计的控制策略,可以实现两轮机器人的行走平衡控制任务。     

Wavelet based control for a class of delayed nonlinear systems with input constraints

In this paper, a delay independent adaptive control strategy is presented for a class of uncertain, delayed nonlinear system subjected to actuator saturation. In proposed control scheme wavelet networks are used for approximation of unknown system dynamics as well as a wavelet based compensator is designed to deal with actuator saturation. Delayed wavelet networks are used for identification of unknown system dynamics having state delayed terms, thereby the approximation capabilities of delayed wavelet network are utilized. Adaptation laws are developed for the online tuning of wavelet parameters. Adaptation singularity problem is solved by employing a switching scheme. The stability of closed loop system and ultimate upper boundedness all closed loop signals is proved by constructing a Lyapunov–Krasovskii functional.     

New tuning conditions for a class of nonlinear PID global regulators of robot manipulators

Several nonlinear proportional-integral-derivative (PID) controllers for robot manipulators that ensure global asymptotic stability have been proposed in the literature. However, the tuning criteria obtained are expressed in terms of conditions so restrictive that they have avoided, until now, carrying out experimental tests with such controllers. Tuning criteria of some PID controllers for robot manipulators with conditions more relaxed than those presented previously in the literature have been proposed in two recent works by the authors. This was achieved by setting the tuning conditions individually for each joint instead of general conditions for the whole robot. In this paper we extend these results to a class of nonlinear PID global regulators for robot manipulators. The obtained tuning criteria are given in terms of conditions so relaxed that they have allowed to carry out, for the first time, experimental essays with these controllers. Such experiments are presented in this paper using a two-degrees-of-freedom robot manipulator.