Yi Hsing Chien

Adaptive T-S fuzzy-neural modeling and control for general MIMO unknown nonaffine nonlinear systems using projection update laws

This paper describes a novel design of an on-line Takagi-Sugeno (T-S) fuzzy-neural controller for a class of general multiple input multiple output (MIMO) systems with unknown nonlinear functions and external disturbances. Instead of modeling the unknown systems directly, the T-S fuzzy-neural model approximates a virtual linearized system (VLS) of a real system with modeling errors and external disturbances. Compared with previous approaches, the main contribution of this paper is an investigation of more general MIMO unknown systems using on-line adaptive T-S fuzzy-neural controllers. In this paper, we also use projection update laws, which generalize the projection algorithm, to tune the adjustable parameters. This prevents parameter drift and ensures that the parameter matrix is bounded away from singularity. We prove that the closed-loop system controlled by the proposed controller is robust stable and the effect of all the modeling errors and external disturbances on the tracking error can be attenuated. Finally, two examples covering four cases are simulated in order to confirm the effectiveness and applicability of the proposed approach in this paper.

Robust Adaptive Controller Design for a Class of Uncertain Nonlinear Systems Using Online T–S Fuzzy-Neural Modeling Approach

This paper proposes a novel method of online modeling and control via the Takagi-Sugeno (T-S) fuzzy-neural model for a class of uncertain nonlinear systems with some kinds of outputs. Although studies about adaptive T-S fuzzy-neural controllers have been made on some nonaffine nonlinear systems, little is known about the more complicated uncertain nonlinear systems. Because the nonlinear functions of the systems are uncertain, traditional T-S fuzzy control methods can model and control them only with great difficulty, if at all. Instead of modeling these uncertain functions directly, we propose that a T-S fuzzy-neural model approximates a so-called virtual linearized system (VLS) of the system, which includes modeling errors and external disturbances. We also propose an online identification algorithm for the VLS and put significant emphasis on robust tracking controller design using an adaptive scheme for the uncertain systems. Moreover, the stability of the closed-loop systems is proven by using strictly positive real Lyapunov theory. The proposed overall scheme guarantees that the outputs of the closed-loop systems asymptotically track the desired output trajectories. To illustrate the effectiveness and applicability of the proposed method, simulation results are given in this paper.

Observer-Based T–S Fuzzy Control for a Class of General Nonaffine Nonlinear Systems Using Generalized Projection-Update Laws

In this paper, we propose an online observer-based Takagi-Sugeno (T-S) fuzzy-output tracking-control technique and an improved generalized projection-update law for a class of general nonaffine nonlinear systems with unknown functions and external disturbances. First, a T-S fuzzy model and a mean-value estimation technique are adopted to approximate a so-called virtual linearized system (VLS) of a real system and avoiding a high-order derivative problem, respectively. Second, a novel design concept combining the T-S fuzzy controller, observer, and tuning algorithm by neural networks is proposed to improve system performance. After that, we also use improved generalized projection-update laws, which prevent parameters drift and confine adjustable parameters to the specified regions, to tune adjustable parameters. As a result, both the stability guarantee based on strictly positive real (SPR) Lyapunov theory and Barbalats lemma and the better tracking performance are concluded. To illustrate the effectiveness of the proposed T-S fuzzy controller and observer-design methodology, numerical simulation results are given in this paper.

On-line Adaptive T-S Fuzzy-Neural Control for A Class of General Multi-Link Robot Manipulators

nonlinear systems. In [4-6], the authors proved that the T-S fuzzy system can approximate any continuous function to any precision. This paper proposes a novel method of on-line modeling and control through the Takagi-Sugeno (T-S) fuzzy-neural model for a class of general n-link robot manipulators. Compared with previous methods, this paper has two unique aspects: first, a more general n-link robot system using on-line adaptive T-S fuzzy-neural controller is investigated, and second, the complete proof of the controller is given. The general robot systems are linearized via the mean value theorem, and then the T-S fuzzy-neural model can approximate the linearized system. Also, we propose an on-line identification algorithm and put significant emphasis on robust tracking controller design using an adaptive scheme for the robot systems. Finally, an example including two cases is provided to demonstrate the feasibility and robustness of the proposed method.

On-Line Hybrid Intelligent Tracking Control for a Class of Nonaffine Multivariable Systems

A novel B-spline neural backstepping controller design with mean-value approximation and first-order filters is proposed for a class of uncertain multiple-input–multiple-output nonaffine nonlinear systems. By combining the proposed systematic backstepping design technique with B-spline neural network structure, one not only has the improved tracking performance but also reduces the computation time. Moreover, using the proposed control scheme, the problems of higher-order derivative and complexity explosion can be solved. According to the stability analysis, it is proven that the tracking errors can be made small by tuning adjustable parameters appropriately. Finally, simulation results are provided to confirm the effectiveness and applicability of the proposed control scheme.

An image-based area measurement system

A novel image-based area measurement system is proposed in this paper, we can convert the pixels of object into real area by image processes. No matter the system is vertical to the measuring plane or not and the height of system, it could calculates arbitrary two points on the image convert into the actual distance. This paper proposed a new structure for achieve our goal, there are four laser projects and camera fixed on same base, then generate four laser spots on the object or measurement plant, and assume connect each two laser spots as parallel lines, finally it could simulated a ruler in the image. Without consider photography angle for this system, it can measure area of floor or height of building.

Control of uncertain active suspension system with anti-lock braking system using fuzzy neural controllers

This paper proposes anti-lock braking system to integrate with active suspensions system applied in a quarter vehicles model, and can use a road estimate to get the road condition. This estimate is based on the LuGre friction model with a road condition parameter, and can transmit a reference slip ration to slip ratio controller through a mapping function considering the effect of road characteristics. In the controller design, an observer-based direct adaptive fuzzy-neural controller (DAFC) for an ABS is developed. After, this paper will discuss that active suspension system influence on ABS. Active suspension systems are not ideal, unchanging, and certain, as many control systems assume. If parts of the suspension system fail, it becomes an uncertain system. In such cases, we need an approximator to remodel this uncertain system to maintain good control. We propose a new method to on-line identify the uncertain active suspension system and design a T-S fuzzy-neural controller to control it. Finally, integrating algorithm is constructed to coordinate these two subsystems. Simulation results of the ABS with active suspension system, and is shown to provide good effectiveness under varying conditions.

T-S fuzzy-neural control for robot manipulators

This paper proposes a novel method of on-line modeling and control through the Takagi-Sugeno (T-S) fuzzy-neural model for a class of general n-link robot manipulators. Compared with the previous method, the main contribution of this paper is an investigation of the more general robot systems using on-line adaptive T-S fuzzy-neural controller. Specifically, the general robot systems are exactly formed a linearized system via the mean value theorem, and then the T-S fuzzy-neural model can approximate the linearized system. Also, we propose an on-line identification algorithm and put significant emphasis on robust tracking controller design using an adaptive scheme for the robot systems. Finally, an example including two cases is provided to demonstrate feasibility and robustness of the proposed method.

FPGA implementation of improved ant colony optimization algorithm for path planning

An improved ant colony optimization (ACO) algorithm is proposed in this paper for improving the accuracy of path planning. The main idea of this paper is to avoid local minima by continuously tuning a setting parameter and the establishment of novel mechanisms for updating partial pheromone and opposite pheromone. As a result, the global search of the proposed ACO algorithm can be significantly enhanced in terms of calculating optimal path compared to the conventional ACO algorithm. Simulation results of the proposed approach show better performances in terms of the shortest distance, mean distance, and success rate towards optimal paths. To further reduce the computation time, the proposed ACO algorithm for path planning is realized on a FPGA chip to verify its practicalities. Experimental results indicate that the efficiency of the path planning is significantly improved by the hardware design of embedded applications.

Mean-Based Fuzzy Control for a Class of MIMO Robotic Systems

This paper presents mean-based fuzzy controllers for trajectory tracking for a class of multiple-input multiple-output robotic systems with nonaffine-like form and parametric uncertainties, in which direct adaptive controllers with state estimators are developed via a mean-based fuzzy identifier without prior knowledge of the membership functions. By using the proposed adaptive technique, unfavorable influence from the initial design of membership functions can be effectively diminished. Moreover, the computation burden of the adaptive laws can be successfully alleviated because the derivative of the fuzzy systems is not required. A Lyapunov-based stability analysis is utilized to guarantee successful system control and desired tracking performance of the closed-loop system. Finally, two examples are provided to demonstrate the feasibility of the proposed control method.