Neural net robot controller: Structure and stability proofs

A multilayer neural net (NN) controller for a general serial-link robot arm is developed. The structure of the NN controller is derived using a filtered error approach. It is argued that standard backpropagation tuning, when used for real-time closed-loop control, can yield unbounded NN weights if: (1) the net can not exactly reconstruct a certain required control function, (2) there are bounded unknown disturbances in the robot dynamics, or (3) the robot arm has more than one link (i.e. nonlinear case). On-line weight tuning algorithms including correction terms to backpropagation, plus an added robustifying signal, guarantee tracking as well as bounded weights. The correction terms involve a second-orderforward-propagated wave in the backprop network.     

Multilayer neural-net robot controller with guaranteed trackingperformance

A multilayer neural-net (NN) controller for a general serial-link rigid robot arm is developed. The structure of the NN controller is derived using a filtered error/passivity approach. No off-line learning phase is needed for the proposed NN controller and the weights are easily initialized. The nonlinear nature of the NN, plus NN functional reconstruction inaccuracies and robot disturbances, mean that the standard delta rule using backpropagation tuning does not suffice for closed-loop dynamic control. Novel online weight tuning algorithms, including correction terms to the delta rule plus an added robust signal, guarantee bounded tracking errors as well as bounded NN weights. Specific bounds are determined, and the tracking error bound can be made arbitrarily small by increasing a certain feedback gain. The correction terms involve a second-order forward-propagated wave in the backpropagation network. New NN properties including the notions of a passive NN, a dissipative NN, and a robust NN are introduced.     

Multilayer discrete-time neural-net controller with guaranteedperformance

A family of novel multilayer discrete-time neural-net (NN) controllers is presented for the control of a class of multi-input multi-output (MIMO) dynamical systems. The neural net controller includes modified delta rule weight tuning and exhibits a learning while-functioning-features. The structure of the NN controller is derived using a filtered error/passivity approach. Linearity in the parameters is not required and certainty equivalence is not used. This overcomes several limitations of standard adaptive control. The notion of persistency of excitation (PE) for multilayer NN is defined and explored. New online improved tuning algorithms for discrete-time systems are derived, which are similar to sigma or epsilon-modification for the case of continuous-time systems, that include a modification to the learning rate parameter plus a correction term. These algorithms guarantee tracking as well as bounded NN weights in nonideal situations so that PE is not needed. An extension of these novel weight tuning updates to NN with an arbitrary number of hidden layers is discussed. The notions of discrete-time passive NN, dissipative NN, and robust NN are introduced. The NN makes the closed-loop system passive.     

电驱动刚性机器人的鲁棒神经网络复合控制

采用逐步逆向的设计思想，提出一种新的电驱动刚性机器人轨迹跟踪的鲁棒神经网络复合控制策略，该策略不仅能有效地消除模型不确定性的影响，而且可避免复杂的求导运算以及对关节角加速度可测的要求。给出了控制器的具体组成和神经网络连接权的在线学习算法，理论表明该算法能保证跟踪误差及神经网络连接权估计最终一致有界，仿真结果也验证了算法的有效性。     

Discrete-time neural net controller for a class of nonlineardynamical systems

A family of two-layer discrete-time neural net (NN) controllers is presented for the control of a class of mnth-order MIMO dynamical system. No initial learning phase is needed so that the control action is immediate; in other words, the neural network (NN) controller exhibits a learning-while-functioning-feature instead of a learning-then-control feature. A two-layer NN is used which is linear in the tunable weights. The structure of the neural net controller is derived using a filtered error approach. It is indicated that delta-rule-based tuning, when employed for closed-loop control, can yield unbounded NN weights if: 1) the net cannot exactly reconstruct a certain required function, or 2) there are bounded unknown disturbances acting on the dynamical system. Certainty equivalence is not used, overcoming a major problem in discrete-time adaptive control. In this paper, new online tuning algorithms for discrete-time systems are derived which are similar to ϵ-modification for the case of continuous-time systems that include a modification to the learning rate parameter and a correction term to the standard delta rule     

Robust backpropagation training algorithm for multilayered neuraltracking controller

A robust backpropagation training algorithm with a dead zone scheme is used for the online tuning of the neural-network (NN) tracking control system. This assures the convergence of the multilayer NN in the presence of disturbance. It is proved in this paper that the selection of a smaller range of the dead zone leads to a smaller estimate error of the NN, and hence a smaller tracking error of the NN tracking controller. The proposed algorithm is applied to a three-layered network with adjustable weights and a complete convergence proof is provided. The results can also be extended to the network with more hidden layers.     

Fuzzy and Recurrent Neural Network Motion Control among Dynamic Obstacles for Robot Manipulators

An integration of fuzzy controller and modified Elman neural networks (NN) approximation-based computed-torque controller is proposed for motion control of autonomous manipulators in dynamic and partially known environments containing moving obstacles. The fuzzy controller is based on artificial potential fields using analytic harmonic functions, a navigation technique common used in robot control. The NN controller can deal with unmodeled bounded disturbances and/or unstructured unmodeled dynamics of the robot arm. The NN weights are tuned on-line, with no off-line learning phase required. The stability of the closed-loop system is guaranteed by the Lyapunov theory. The purpose of the controller, which is designed as a neuro-fuzzy controller, is to generate the commands for the servo-systems of the robot so it may choose its way to its goal autonomously, while reacting in real-time to unexpected events. The proposed scheme has been successfully tested. The controller also demonstrates remarkable performance in adaptation to changes in manipulator dynamics. Sensor-based motion control is an essential feature for dealing with model uncertainties and unexpected obstacles in real-time world systems.     

Robust neural-network control of rigid-link electrically drivenrobots

A robust neural-network (NN) controller is proposed for the motion control of rigid-link electrically driven (RLED) robots. Two-layer NN's are used to approximate two very complicated nonlinear functions. The main advantage of our approach is that the NN weights are tuned online, with no off-line learning phase required. Most importantly, we can guarantee the uniformly ultimately bounded (UUB) stability of tracking errors and NN weights. When compared with standard adaptive robot controllers, we do not require lengthy and tedious preliminary analysis to determine a regression matrix. The controller can be regarded as a universal reusable controller because the same controller can be applied to any type of RLED robots without any modifications.     

Direct adaptive controller for nonaffine nonlinear systems using self-structuring neural networks

A direct adaptive state-feedback controller is proposed for highly nonlinear systems. We consider uncertain or ill-defined nonaffine nonlinear systems and employ a neural network (NN) with flexible structure, i.e., an online variation of the number of neurons. The NN approximates and adaptively cancels an unknown plant nonlinearity. A control law and adaptive laws for the weights in the hidden layer and output layer of the NN are established so that the whole closed-loop system is stable in the sense of Lyapunov. Moreover, the tracking error is guaranteed to be uniformly asymptotically stable (UAS) rather than uniformly ultimately bounded (UUB) with the aid of an additional robustifying control term. The proposed control algorithm is relatively simple and requires no restrictive conditions on the design constants for the stability. The efficiency of the proposed scheme is shown through the simulation of a simple nonaffine nonlinear system.     

Control of Nonaffine Nonlinear Discrete-Time Systems Using Reinforcement-Learning-Based Linearly Parameterized Neural Networks

A nonaffine discrete-time system represented by the nonlinear autoregressive moving average with eXogenous input (NARMAX) representation with unknown nonlinear system dynamics is considered. An equivalent affinelike representation in terms of the tracking error dynamics is first obtained from the original nonaffine nonlinear discrete-time system so that reinforcement-learning-based near-optimal neural network (NN) controller can be developed. The control scheme consists of two linearly parameterized NNs. One NN is designated as the critic NN, which approximates a predefined long-term cost function, and an action NN is employed to derive a near-optimal control signal for the system to track a desired trajectory while minimizing the cost function simultaneously. The NN weights are tuned online. By using the standard Lyapunov approach, the stability of the closed-loop system is shown. The net result is a supervised actor-critic NN controller scheme which can be applied to a general nonaffine nonlinear discrete-time system without needing the affinelike representation. Simulation results demonstrate satisfactory performance of the controller.     

Hybrid trigonometric compound function neural networks for tracking control of a nonholonomic mobile robot

The purpose of this paper is to propose a hybrid trigonometric compound function neural network (NN) to improve the NN-based tracking control performance of a nonholonomic mobile robot with nonlinear disturbances. In the mobile robot control system, two NN controllers embedded in the closed-loop control system have the simple continuous learning and rapid convergence capability without the dynamics information of the mobile robot to realize the tracking control of the mobile robot. The neuron functions of the hidden layer in the three-layer feedforward network structure consist of the compound cosine function and the compound sine function combining a cosine or a sine function with a unipolar sigmoid function. The main advantages of this NN-based mobile robot control system are better real-time control capability and control accuracy by use of the proposed NN controllers for a nonholonomic mobile robot with nonlinear disturbances. Through simulation experiments applied to the nonholonomic mobile robot with the nonlinear disturbances of dynamics uncertainty and external disturbances, the simulation results show that the proposed NN control system of a nonholonomic mobile robot has better real-time control capability and control accuracy than the compound cosine function NN control system of a nonholonomic mobile robot and then verify the effectiveness of the proposed hybrid trigonometric compound function NN controller for improving the tracking control performance of a nonholonomic mobile robot with nonlinear disturbances.     

Reinforcement learning neural-network-based controller for nonlinear discrete-time systems with input constraints.

A novel adaptive-critic-based neural network (NN) controller in discrete time is designed to deliver a desired tracking performance for a class of nonlinear systems in the presence of actuator constraints. The constraints of the actuator are treated in the controller design as the saturation nonlinearity. The adaptive critic NN controller architecture based on state feedback includes two NNs: the critic NN is used to approximate the "strategic" utility function, whereas the action NN is employed to minimize both the strategic utility function and the unknown nonlinear dynamic estimation errors. The critic and action NN weight updates are derived by minimizing certain quadratic performance indexes. Using the Lyapunov approach and with novel weight updates, the uniformly ultimate boundedness of the closed-loop tracking error and weight estimates is shown in the presence of NN approximation errors and bounded unknown disturbances. The proposed NN controller works in the presence of multiple nonlinearities, unlike other schemes that normally approximate one nonlinearity. Moreover, the adaptive critic NN controller does not require an explicit offline training phase, and the NN weights can be initialized at zero or random. Simulation results justify the theoretical analysis.     

A direct NN‐approximation control for uncertain nonaffine systems with unknown control direction

In this paper, we present an adaptive neural network (NN) controller for uncertain nonaffine systems with unknown control direction. Most of the previous NN‐based controllers included a damping term in the adaptive law of NN weights to ensure the closed‐loop stability. The estimated error of the NN weights as well as the tracking error were therefore increased, relying not only on the size of the NN approximation error but also on the ideal NN weights. Compared with those, the proposed controller evades using the damping term through combining a novel adaptive algorithm and a switching mechanism to update the weights. The NN thus can directly approach a target controller with satisfactory accuracy even if the control direction is unknown. Stability analysis shows that the tracking error and the estimated error of NN weights both converge to small neighbors of 0 which solely depend on the NN approximation error. At last, simulations on a Duffing‐Holmes chaotic model show the effectiveness of the proposed controller in comparison to another NN‐based method.     

Control of a nonholomic mobile robot: Backstepping kinematics into dynamics

A dynamical extension that makes possible the integration of a kinematic controller and a torque controller for nonholonomic mobile robots is presented. A combined kinematic/torque control law is developed using backstepping, and asymptotic stability is guaranteed by Lyapunov theory. Moreover, this control algorithm can be applied to the three basic nonholonomic navigation problems: tracking a reference trajectory, path following, and stabilization about a desired posture. The result is a general structure for controlling a mobile robot that can accommodate different control techniques, ranging from a conventional computed-torque controller, when all dynamics are known, to robust-adaptive controllers if this is not the case. A robust-adaptive controller based on neural networks (NNs) is proposed in this work. The NN controller can deal with unmodeled bounded disturbances and/or unstructured unmodeled dynamics in the vehicle. On-line NN weight tuning algorithms that do not require off-line learning yet guarantee small tracking errors and bounded control signals are utilized. © 1997 John Wiley & Sons, Inc.     

Neural-Network-Based State Feedback Control of a Nonlinear Discrete-Time System in Nonstrict Feedback Form

In this paper, a suite of adaptive neural network (NN) controllers is designed to deliver a desired tracking performance for the control of an unknown, second-order, nonlinear discrete-time system expressed in nonstrict feedback form. In the first approach, two feedforward NNs are employed in the controller with tracking error as the feedback variable whereas in the adaptive critic NN architecture, three feedforward NNs are used. In the adaptive critic architecture, two action NNs produce virtual and actual control inputs, respectively, whereas the third critic NN approximates certain strategic utility function and its output is employed for tuning action NN weights in order to attain the near-optimal control action. Both the NN control methods present a well-defined controller design and the noncausal problem in discrete-time backstepping design is avoided via NN approximation. A comparison between the controller methodologies is highlighted. The stability analysis of the closed-loop control schemes is demonstrated. The NN controller schemes do not require an offline learning phase and the NN weights can be initialized at zero or random. Results show that the performance of the proposed controller schemes is highly satisfactory while meeting the closed-loop stability.      

Robust force/motion control of constrained robots using neural network

A robust neural network (NN) controller is proposed for the simultaneous force/motion control of constrained rigid robots. The NN weights here are tuned on‐line, with no off‐line learning phase required. Most importantly, we can guarantee the boundedness of constraint force errors, joint position tracking errors, and NN weights. When compared with adaptive controllers, we do not require linearity in the unknown parameters, and the tedious computation of the regression matrix. Novel passivity properties of the NN controller are stated and proven. ©1999 John Wiley & Sons, Inc.     

Asymptotic Adaptive Neural Network Tracking Control of Nonholonomic Mobile Robot Formations

In this paper, asymptotically stable control laws are developed for leader–follower based formation control using backstepping in order to accommodate the dynamics of the robots and the formation. First, a kinematic controller is developed around control strategies for single mobile robots and the idea of virtual leaders. The virtual leader is replaced with a physical mobile robot leader, and an auxiliary velocity control law is developed in order to prove the global asymptotic stability of the followers which in turn allows the local asymptotic stability of the entire formation. A novel approach is taken in the development of the dynamical controller such that the torque control inputs for the follower robots include the dynamics of the follower robot as well as the dynamics of its leader, and two cases are considered—the case when the robot dynamics are known and the case when they are unknown. In the first case, a robust adaptive control term is utilized to account for unmodeled dynamics. For the latter, a robust adaptive term is augmented with a NN control law to achieve asymptotic tracking performance in contrast with most NN controllers where a bounded tracking error result is shown. Additionally, the NN approximation error is assumed to be a function of tracking errors instead of a constant upper bound, which is commonly found in the literature. The stability of the follower robots as well as the entire formation is demonstrated in each case using Lyapunov methods and numerical results are provided.     

Neural network-based adaptive controller design of roboticmanipulators with an observer

A neural network (NN)-based adaptive controller with an observer is proposed for the trajectory tracking of robotic manipulators with unknown dynamics nonlinearities. It is assumed that the robotic manipulator has only joint angle position measurements. A linear observer is used to estimate the robot joint angle velocity, while NNs are employed to further improve the control performance of the controlled system through approximating the modified robot dynamics function. The adaptive controller for robots with an observer can guarantee the uniform ultimate bounds of the tracking errors and the observer errors as well as the bounds of the NN weights. For performance comparisons, the conventional adaptive algorithm with an observer using linearity in parameters of the robot dynamics is also developed in the same control framework as the NN approach for online approximating unknown nonlinearities of the robot dynamics. Main theoretical results for designing such an observer-based adaptive controller with the NN approach using multilayer NNs with sigmoidal activation functions, as well as with the conventional adaptive approach using linearity in parameters of the robot dynamics are given. The performance comparisons between the NN approach and the conventional adaptation approach with an observer is carried out to show the advantages of the proposed control approaches through simulation studies     

Projection-Based Adaptive Neurocontrol With Switching Logic Deadzone Tuning

In this brief, an adaptive neural network (NN) controller is proposed for multiple-input-multiple-output (MIMO) nonlinear systems with triangular control structure and unknown control directions. Deadzones are employed in the projection-based NN weight learning laws and the Nussbaum parameter update laws with levels tuned by an innovative switching logic tuning mechanism. Detailed analysis using a family of Lyapunov-like integral functions and the function approximation capability of NNs proves that all the tracking errors are semiglobal uniform ultimate bounded in small neighborhoods of the origin while the closed-loop system variables (state vector, NN weights, Nussbaum parameters) and the control law remain bounded. A simulation study confirms the theoretical results and verifies the effectiveness of the proposed design.     

Identification of nonlinear dynamical systems using multilayered neural networks

A novel multilayer discrete-time neural net paradigm is presented for the identification of multi-input multi-output (MIMO) nonlinear dynamical systems. The major novelty of this approach is a rigorous proof of identification error convergence that reveals a requirement for a new identifier structure and nonstandard weight tuning algorithms. The NN identifier includes modified delta rule weight tuning and exhibits a learning-while-functioning feature instead of learning-then-functioning, so that the identification is on-line with no explicit off-line learning phase needed. The structure of the neural net (NN) identifier is derived using a passivity aproach. Linearity in the parameters is not required and certainty equivalence is not used. The notion of persistency of excitation (PE) and passivity properties of the multilayer NN are defined and used in the convergence analysis of both the identification error and the weight estimates.