Horn-Yong Jan

GA-based multiobjective PID control for a linear brushless DC motor

This paper presents a robust output tracking control design method for a linear brushless DC motor with modeling uncertainties. Two frequency-domain specifications directly related to the mixed sensitivity function and control energy consumption are imposed to ensure stability and performance robustness. With regard to time-domain specifications, the rise time, maximum overshoot and steady-state error of the step response are considered. A generalized two-parameters proportional, integral, and derivative (PID) control framework is developed via a genetic searching approach ensuring the specifications imposed. The proposed design method is intuitive and practical that offers an effective way to implement simple but robust solutions covering a wide range of plant perturbation and, in addition, provides excellent tracking performance without resorting to excessive control. Extensive experimental and numerical results for a linear brushless motor confirm the proposed control design approach.

Self-organizing PID control design based on DNA computing method

A novel structure variable PID control design based on the DNA algorithm is proposed. The algorithm uses a coding method originated from the structure of biological DNA molecules to map parameters as well as the structure of PID controllers. A frameshift mutation operator is presented to modify the controllers structure while performing the molecular evolution process. Extensive computer simulations and experiments are presented to support the proposed design.

Evolutionarily multiobjective PID control for linear brushless DC motors

This paper presents a robust output tracking control design for a linear brushless DC motor with modeling uncertainties. Frequency-domain design specifications directly related to the mixed sensitivity function and control energy consumption are imposed to ensure stability and performance robustness. A generalized two-parameter PID control framework is developed via an evolution algorithm, which searches the available solutions over certain specified domain. The proposed design paradigm is intuitive and practical in the sense that it offers an effective way to implement simple but robust solutions covering a wide range of plant perturbation and, in addition, provides excellent tracking performance without resorting to excessive control. Experimental and numerical studies are discussed.

Self-organized PID control design using DNA computing approach

Abstract A new design and realization method solving for the constrained multi‐objective problem via a self‐organizing PID control design using the DNA computing algorithm is proposed. Requirements of stability robustness and optimality related to the H ∞ and H 2 performance specifications are imposed as the objectives. The algorithm uses a coding method originating from the structure of biological DNA molecules to map parameters as well as the structure of PID controllers into DNA strings. Structured mutation operators are proposed to modify the control structure during the computation process. Simulations and experiments are performed to verify the performance and applicability of our design.

System identification: DNA computing approach

A DNA computing algorithm (DNACA) with an electron-ion interaction potential (EIIP) decoding scheme is proposed to identify a class of transfer functions. The DNACA includes enzyme and virus operators which provide a highly modular, flexible, and accurate self-organizing structure environment. Simulation study based on De Jongs test functions show its superior performance when compared with the improved and standard genetic algorithms (GAs).

Singularity Characterization and Path Planning of a New 3 Links 6-DOFs Parallel Manipulator

This paper presents singularity characterization and optimal path planning for a new three links six degreesof- freedom (DOFs) parallel manipulator. To achieve high speed handling and machining, the base platform is equipped with three linear slideways each one actuated by a linear DC motor, and each extensible vertical link connecting the upper and base platforms is actuated by an inductive AC servo motor. Characterization of workspace, singularity, and development of an optimal path planner are presented. Our special emphases are on characterization of the platform singularity using genetic algorithms (GAs) and singularity avoidance of the path planning based on a DNA evolutionary computing algorithm. We propose a path planning scheme whose coding technique speeds up the execution of search for fast path generation within the feasible workspace.

Structure variable PID control design based on DNA coding method

A novel structure variable PID control design based on the DNA algorithm is proposed. The algorithm uses a coding method originated from the structure of biological DNA molecules to map parameters as well as the structure of PID controllers. A frameshift mutation operator is presented to modify the controllers structure while performing the molecular evolution process. Extensive computer simulations and experiments are presented to support the proposed design.

Singularity analysis and path planning for a MDOF manipulator

Abstract This paper presents singularity characterization and path planning design for a newly developed 3 leg 6 degree‐of‐freedom (DOFs) parallel manipulator. Special emphasis is placed on characterizing the platform singularity and singularity avoidance of path planning based on genetic algorithms. The path planning scheme proposed uses a real‐coded genetic algorithm including a modified crossover and a swap mutation for searching out the shortest moving path in the available workspace.

Optimal path planning on 3D space using a DNA computing algorithm

This paper proposes a new DNA computation-based optimal path planning algorithm on 3D space. In the proposed approach, the working space is converted into several slices by the DNA coding scheme during the evolutionary process, where less blocks of every slices correspond to the areas containing loosely dense obstacles and vice versa. The molecular programming algorithm imitates the biological evolution mechanism through artificial programming to enhance the opportunities for searching the shortest moving path while avoiding obstacles. Simulation results are presented to show the effectiveness of the proposed approach.

Control design for a mixed rotary and linear motors based manipulator

Modeling and control design for a six degree-of-freedom (DOF) platform manipulator is presented. The platform is setup by three extensible legs sliding on three linear slideways (forming the base platform) each actuated by a synchronous linear servo motor. The vertical legs are actuated by inductive AC servo motors. Combination of linear and AC servo motors contribute high-speed performance of the platform. Inverse kinematics describing the platform motion is derived. With regard to the control system, two types of controllers for motors (linear and AC servo motors) and platform are designed respectively. The controller proposed consists of two parts, one is a state feedback component, and the other one uses a learning feedback component constituted by a wavelet neural network. For platform motion control, a cerebellar model arithmetic controller is adopted to control the position and orientation of the moving platform. Extensive simulation studies are presented to verify effectiveness of the control strategy on the motors and the overall platform motion.