Ching-Huei Huang

Structure-specified IIR filter and control design using real structured genetic algorithm

This paper develops an innovative optimization method, real structured genetic algorithm (RSGA), which combines the advantages of traditional real genetic algorithm (RGA) with structured genetic algorithm (SGA), and applies it for digital filter and control design optimization problems. For infinite impulse response (IIR) filter designs, the proposed approach fulfills all types of filters by minimizing the order of the filter and the absolute error of both passband and stopband. Both system structure and parametric variables are simultaneously optimized via the proposed chromosome scheme. The approach has also been extended to deal with robust control design problems. The approach offers an effective method for designing an optimal controller with robust stability. Simulation and experimental results conveys the excellence of the proposed algorithm over traditional approaches in convergence speed, performance, cost effectiveness, and attains simpler structure.

Microbrushless DC Motor Control Design Based on Real-Coded Structural Genetic Algorithm

This paper presents the realization of a microbrushless dc motor (MBDCM) feedback system based on a real-coded structural genetic algorithm (RSGA), which combines the advantages of conventional real genetic algorithms and structured genetic algorithms for optimal control design. In the RSGA, a dynamic crossover and mutation probability adjusting method mimicking the characteristics of Butterworth filters is proposed to enhance the search performance. A SinCos encoder with a line drive of 128 sin/cos signals per revolution is implemented to achieve precise positioning. The SinCos encoder possesses the advantage of high resolution via signal interpolation. The method inherited is simple yet effective, based on logic devices. To verify effectiveness of the proposed methodology, simulations are conducted and an experimental platform with a digital signal processing unit, a motor driver, a MBDCM, and a SinCos encoder is built to verify applicability of the proposed method. The experimental results demonstrating the aforementioned method work properties correlate well with the expectation.

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).

Flight path planning for mini rotor UAVs

Optimized path planning algorithm design is an important function of unmanned aerial vehicles (UAVs). Since the path obtained by the (rapidly exploring random tree) RRT is not necessarily optimal, excessive waypoints need to be refined. Heading of the quadrotor UAV is considered here as a moving cost of the improved A-star algorithm. Since the secondary path generated by the improved A-star algorithm is piecewise linear and may exhibit rugged curvature, its necessary to refine the flight path making it to be applicable for quadrotor UAV flight. The proposed path planning method has been well verified via a variety of scenarios.

Evolutionary neural networks and DNA computing algorithms for dual-axis motion control

A new method is proposed to deal with the dual-axis control of a multi-variables system with two induction motors. Investigation of resolving the cross-coupling problem of dual-axis platform is addressed by a neural net-based decoupling compensator and a sufficient condition ensuring closed-loop stability is derived. An evolutionary algorithm processing the universal seeking capability is proposed for finding the optimal connecting weights of the neural decoupling compensator and the gains of PID controllers. Extensive numerical studies verify the performance and applicability of the proposed design under a variety of operating conditions.

Identification of Flight Vehicle Models Using Fuzzified Eigensystem Realization Algorithm

This paper presents a new approach with a fuzzified eigensystem realization algorithm for identification of flight vehicle models in low-speed wind tunnel (LSWT) and high-speed wind tunnel (HSWT). A variety of variables in model types and testing environment (such as angle-of-attack, sideslip angle, tunnel wind speed) and profile, elevator, and power system (motor and propeller) of mini unmanned aerial vehicle (mini-UAV) model are considered in a power-on mini-UAV testing system in LSWT and an Advisory Group for Aerospace Research and Development (AGARD) standard calibration model in HSWT. The method based on the fuzzy logic inference structure is simple and effective. The results obtained are compared to those obtained by the conventional wind tunnel testing method. To verify the effectiveness of the proposed methodology, simulations are conducted using real-world experimental results that demonstrate that the working performance of the proposed method correlates well as expected.

Unmanned Aerial Vehicles Evolutional Flight Route Planner Using the Potential Field Approach

This study proposes a newflight route-planning technique for autonomous navigation of unmanned aerial vehicles (UAVs) based on the combination of evolutionary algorithms with virtual potential fields. By combining a radial forcefieldwith a swirling forcefield, three-dimensional virtual potentialfields are constructed for repelling infeasible UAV flight routes from threatening zones. To ensure feasibility, major flight constraints are considered when searching for the optimal flight route. This study examines both singleand multiple-obstacle cases to determine the efficiency of the proposed flight route planner. The UAV navigation method uses an offline planner in known environments and an online planner for flight route replanning when popup threats emerge. Both planners were tested under various scenarios. The results show that the proposed planner can efficiently enable the safe navigation of UAVs.

Flight path planning simulation for VTUAV

This paper develops a path planning simulation method in AGI software platform for multirotor UAVs in a 3D environment with avoiding the obstacles is presented. A multi-RRT (rapidly exploring random tree, RRT) algorithm is applied to compute the preliminary flight paths.

Point stabilization for autonomous lawnmower using backstepping adaptive control

This paper studied the point stabilization problem for a constrained autonomous lawnmower. A constrained kinetic model is established first. The authors proposed a backstepping adaptive controller to solve the stabilizing control design problem. The proposed approach has been numerically verified.