Shiuh-Jer Huang

Control of an inverted pendulum using grey prediction model

The system with partial unknown structure, parameters and characteristics is called a grey system. The grey theory can be employed to improve the control performance of a system without sufficient information or with a highly nonlinear property. In this paper, the grey prediction model combined with a PD controller is proposed to balance an inverted pendulum which is a classic example of an inherently nonlinear unstable system. The control objective is to swing up the pendulum from the stable position to the unstable position and bring its slider back to the origin of the track. The overall control algorithm is decomposed into two separate grey model controllers for swinging up and balancing based upon the angular and velocity values of the pendulum. The actuator is a Nippon Seiko Co. (NSK) linear motor. The experimental results show that this grey model controller is able to swing up and balance the inverted pendulum and guide its slider to the center of the track. It also has the robustness to balance the inverted pendulum in suffering an external impact acting on the pendulum.<<ETX>>

Adaptive fuzzy controller with sliding surface for vehicle suspension control

Since the hydraulic actuating suspension system has nonlinear and time-varying behavior, it is difficult to establish an accurate model for designing a model-based controller. Here, an adaptive fuzzy sliding mode controller is proposed to suppress the sprung mass position oscillation due to road surface variation. This intelligent control strategy combines an adaptive rule with fuzzy and sliding mode control algorithms. It has online learning ability to deal with the system time-varying and nonlinear uncertainty behaviors, and adjust the control rules parameters. Only eleven fuzzy rules are required for this active suspension system and these fuzzy control rules can be established and modified continuously by online learning. The experimental results show that this intelligent control algorithm effectively suppresses the oscillation amplitude of the sprung mass with respect to various road surface disturbances.

Development and application of a novel radial basis function sliding mode controller

Generally, physical systems have certain non-linear and time-varying behaviours and various uncertainties. It is difficult to establish an appropriate model for controller design. Adaptive and sliding mode control schemes have been employed to solve some of these problems under certain model-based conditions and limitations. Here a novel adaptive radial basis functions sliding mode control is proposed by combining the advantages of the adaptive, neural network and sliding mode control strategies without precise system model information. It has on-line learning ability to deal with the system time-varying and non-linear uncertainties by adjusting the control parameters. The proposed scheme is implemented on a three degree-of-freedom dynamic absorber system. Only five radial basis functions are required for this control system and their weightings can be established and updated continuously by on-line learning. The experimental results show that this intelligent control approach effectively suppresses the vibration amplitude of the main mass due to external disturbances.

A stable self-organizing fuzzy controller for robotic motion control

It is well known that robotic manipulators are highly nonlinear coupling dynamic systems. It is difficult to establish an appropriate mathematical model for the design of a model-based controller. Although fuzzy logic control has a model-free feature, it still needs time-consuming work for the rules bank and fuzzy parameters adjustment. In this paper, a stable self-organizing fuzzy controller (SOFC) is proposed to manipulate the motion trajectory of a 5-degrees-of-freedom robot. This approach has a learning ability for responding to the time-varying characteristic of a robot. Its control rules bank can be established and modified continuously by online learning with zero initial fuzzy rules. In addition, this control strategy has effectively improved the stability problem of a previous SOFC. The experimental results show that this intelligent controller has a stable learning ability and good motion control capability.

A Self-Organizing Fuzzy Controller for an Active Suspension System

A quarter-car hydraulic suspension system has been constructed to evaluate the performance of active vehicle suspension. Since this hydraulic actuating suspension system has a nonlinear and complicated mathematical model, it is difficult to design a model-based controller. Hence, a self-organizing fuzzy controller (SOFC) is employed to control the position and acceleration oscillation amplitudes of the sprung mass due to the rough road variation. This approach has learning ability for responding to the time-varying characteristic of the oscillation coming from the tire. Its control rule bank can be established and modified continuously by on-line learning. E-modification and dead-zone concepts are introduced into the SOFC fuzzy adaptation rule to improve the oscillation feature of control law and the gradual divergence problem. The experimental results show that this intelligent controller effectively suppresses the vibration amplitude and reduces the acceleration of the sprung mass correlating to the road variation for improving the vehicle ride comfort.

Fuzzy logic controller for a vehicle active suspension system

Abstract An active suspension system has been proposed to improve the ride comfort. Its dynamic performance was investigated by using computer simulation results. A quarter-car 2 degree-of-freedom (DOF) system is designed and constructed on the basis of the concept of a four-wheel independent suspension to simulate the actions of an active vehicle suspension system. Since the mathematical model of this hydraulic actuating suspension system is non-linear and complicated, it is difficult to derive an accurate system model for designing a model-based controller. Therefore, a model-free fuzzy control algorithm is employed to design a controller for achieving vibration isolation. The experimental results show that the tyre deformation influences significantly the control performance of the active suspension system. Hence, a grey predictor is introduced to predict the tyre deformation and filter it from the feedback error signal. The control performance of this fuzzy control strategy with grey predictor is significantly improved.

A Combination of Fuzzy Logic And Neural Network Algorithms for Active Vibration Control

The construction of a dynamic absorber incorporating active vibration control is described. The absorber is a 2 degree of freedom spring-lumped mass system sliding on a guide pillar, with two internal vibration disturbance sources. Both the main mass and the secondary absorber mass were acted on by direct current (d.c.) servo motors, respectively, to suppress the vibration amplitude. In this paper, a new control approach is proposed by combining fuzzy logic and neural network algorithms to control the multi-input/multi-output (MIMO) system. Firstly, the fuzzy logic controller was designed for controlling the main influence part of the MIMO system. Secondly, the coupling neural network controller was employed to take care of the coupling effect and refine the control performance of the MIMO system. The experimental results show that the control system effectively suppresses the vibration amplitude and with good position tracking accuracy.

An adaptive fuzzy sliding-mode controller for servomechanism disturbance rejection

A two-level spring-lumped mass servomechanism system was constructed for disturbance rejection control investigation. This dynamic absorber is similar to a model of the serial-type vehicle suspension system. The lower level is actuated by two DC servo motors, to provide the specified internal and external disturbances to the vibration control system. The upper level has another DC servo motor to control the main body balancing position. In order to tackle the systems nonlinear and time-varying characteristics, an adaptive fuzzy sliding-mode controller is proposed to suppress the main mass position variation due to external disturbance. This intelligent control strategy combines an adaptive rule with fuzzy and sliding-mode control technologies. It has online learning ability for responding to the systems time-varying and nonlinear uncertainty behaviors, and for adjusting the control rules and parameters. Only seven rules are required for this control system, and its control rules can be established and modified continuously by online learning. The experimental results show that this intelligent control approach effectively suppresses the vibration amplitude of the body, with respect to the external disturbance.

Fuzzy logic for constant force control of end milling

The machining condition usually has significant variation resulting from the change of cutting depth and the intrinsic property of the workpiece. In order to maintain the performance of a classical proportional integral derivative control system, the tool life and machining quality, conservative feedrate, and cutting depth change are prespecified as the limitations of computer numerically controlled operators. Therefore, constant cutting force control is proposed as a useful approach for increasing the metal removal rate and the tool life. However, the model-based controller cannot handle the nonlinearity of a force control system due to cutting condition variations. Here, a fuzzy controller with learning ability was employed to improve both the system performance and the adaptability. This control approach vias implemented on a retrofit old-fashioned milling machine for the end milling process. The experimental results show that this control strategy has smooth feedrate and good cutting force dynamic responses.

A hybrid fuzzy logic and neural network algorithm for robot motion control

Robotic manipulators are multivariable nonlinear coupling dynamic systems. Industrial robots were controlled by using a traditional controller, the control performance of which may change with respect to operating conditions. Since the robotic manipulators have complicated nonlinear mathematical models, control systems based on the system model are difficult to design. In this paper, a model-free hybrid fuzzy logic and neural network algorithm was proposed to control this multi-input/multi-output (MIMO) robotic system. First, a fuzzy logic controller was designed to control individual joints of this 4-degree-of-freedom (DOF) robot. Secondly, a coupling neural network controller was introduced to take care of the coupling effect among joints and refine the control performance of this robotic system. The experimental results showed that the application of this control strategy effectively improved the trajectory tracking precision.