A new method for the robust control of flexible-joint (FJ) robots with model uncertainties in both robot dynamics and actuator is proposed. The proposed system a combination adaptive dynamic surface (DSC) technique self-recurrent wavelet neural network (SRWNN). DSC provides ability to overcome "explosion complexity" problem backstepping controllers. SRWNNs are used observe arbitrary FJ robots, all their weights trained online. From Lyapunov stability analysis, adaptation laws induced,...

In this paper, a generalized predictive control (GPC) method based on self-recurrent wavelet neural network (SRWNN) is proposed for stable path tracking of mobile robots. Since the SRWNN has mother layer, it can well attract complex nonlinear system although less nodes than network. Thus, used as model identifier approximating on-line states robot. our system, since inputs, parameters are trained by gradient descent with adaptive learning rates (ALRs), optimal which suitable time-varying...

In this paper, we propose a high-resolution time-frequency-domain reflectometry technique as methodology of detection and estimation faults on wire. This method adopts the time-frequency cross-correlation characteristics observed signal in both time frequency domains simultaneously. The accuracy proposed is verified with experiments using radio-guide-type coaxial cable comparing it traditional time-domain well frequency-domain methods. It clearly shown here that algorithm produces excellent...

This paper presents the decentralized sampled-data fuzzy observer design techniques for nonlinear interconnected systems, which are assumed to be composed of subsystems and unknown interconnections. To observer, estimation error dynamics is obtained, performance function defined. Based on function, continuous-time first proposed minimize interconnection bound attenuation degree ratio. In addition, observers designed by using approximate discretization exact discrete-time approaches,...

A precise range difference (RD)-based passive target localization algorithm is proposed for mobile robot applications. To effectively solve the real-time issue, nonlinear relation between RD information and location removed by introducing as an auxiliary variable to be estimated. Then, RD-based problem formulated in setting of linear estimation. The resultant measurement model contains stochastic parametric uncertainty which causes severe performance degradation conventional least squares...

In addition to diagnosing a wiring of the vehicle in operation, it is also very important find wire mismatches during assembly process. this paper, we propose new method combining time-frequency-domain reflectometry and deep learning verify that connected proper port underhood electrical center. Considering frequency characteristics each (black, blue, red, yellow), develop an optimization signal design algorithm. Using time-frequency cross correlation (TFCC), reflected generated at impedance...

This paper proposes a decentralized H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> sampled-data fuzzy filter design for nonlinear interconnected systems composed of multiple subsystems with uncertain interconnections and variable sampling intervals. The system the are modeled using Takagi-Sugeno systems. To derive filter-design conditions, estimation error dynamics modeled, performance inequality is defined. aimed at minimizing ratio...

The maintenance of control and instrumentation (C&I) cables is crucial to safety operating nuclear power plants. Therefore, when an accident occurs, there a need for accurate assessment the impact on cable's integrity. Unfortunately, most cable diagnostic methods are destructive real-time effect accidents not possible. Thus, in this paper, we present analysis specific type accident, loss coolant (LOCA), C&I real-time, based time-frequency domain reflectometry (TFDR). Because TFDR sensitive...

This study presents a stability condition for guaranteed cost sampled-data fuzzy control problem of non-linear systems. A Takagi–Sugeno (T–S) model is adopted the system and controller designed T–S system. To develop control, new closed-loop in continuous-time Lyapunov sense, its sufficient conditions are formulated terms linear matrix inequalities. Based on condition, considered minimised from function. Finally, numerical examples provided to verify effectiveness proposed technique.

In this study, an intelligent digital redesign (IDR) technique is proposed for observer‐based sampled‐data fuzzy controller of non‐linear systems. By using a Takagi–Sugeno model, the pre‐designed analog and controllers are supposed, these discretised closed‐loop systems obtained, respectively. Based on IDR problem, authors guarantee both stability state‐matching conditions. Unlike previous techniques, not only improves performance error cost function, but also derived in strict linear matrix...

In this paper, a decentralized sampled-data tracking controller design technique for Takagi-Sugeno fuzzy interconnected systems is proposed. The overall large-scale system assumed to consist of subsystems with unknown but norm-bounded interconnections. This paper focuses mainly on the model reference based exact discretization approach, by which resulting error dynamics becomes asymptotically stabilized predefined H-infinity performance criterion. To end, we first continuous-time proposed...

A linear chirp reflectometry with stretching processing is used to detect and locate low-voltage control instrumentation cable splices fault. Time delay information in the reflected signal transformed instantaneous beat angular frequency by process estimated Kalman smoother order obtain high resolution time-frequency spectrum of nonstationary signal. Based on frequency, magnitude phase difference reflection coefficient are filtering. To verify performance proposed method, comparative...

This paper proposes a direct adaptive control method for stable path tracking of mobile robots using self recurrent wavelet neural network (SRWNN). As the proposed SRWNN is modified model (WNN), includes basic ability WNN such as fast convergence. Besides has property, unlike WNN, that can store past information because mother layer composed self-feedback neurons. Accordingly, easily cope with unexpected change system. For problem, two SRWNNs are used each controller generating inputs,...

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> An improved pinch detection algorithm is proposed for low-cost antipinch window control systems. Apart from previous works, the makes use of torque rate information to sense pinched conditions and perform safety precautions. The motivation this approach comes idea that less sensitive motor parameter uncertainty than or angular velocity. estimator designed by applying steady-state Kalman filter...