Alfred Ng

Actuator Placement Optimization and Adaptive Vibration Control of Plate Smart Structures

In this paper, a performance criterion is proposed for the optimization of piezoelectric patch actuator locations on flexible plate structures based on maximizing the controllability grammian. This is followed by the determination of parameters required for actuator location optimization through Structuring Analysis in ANSYS Finite Element Analysis Package. Genetic Algorithm is then used to implement the optimization. Finally, with the actuators bonded on optimized locations, a filtered-x LMS-based multichannel adaptive control is applied to suppress vibration response of the plate. Numerical simulations are performed in suppressing tri-sinusoidal response at three points of the plates. The results show that the developed actuator placement optimization methodology is very effective in searching for the optimal actuator locations that minimize the energy requirement of vibration control. The control algorithm is also demonstrated to be efficient and robust in the smart structure vibration control.

A Bang -Bang Control Approach to Maneuver Spacecraft in a Formation with Differential Drag

The idea of Formation keeping and maneuvering with diff erential drag was first introduced by C.L.Leonard in the late nineteen eighties. The method is based on simultaneously solving a double integrator and a harmonic oscillator that represent the relative states of motion. Although the results of the paper sho w that the method is feasible, it is only true for the Keplerian case of motion. In this paper, we elaborate on this method by considering the factors that were ignored in the original paper; natural perturbations, changing values of drag acceleration, err oneous measurements, attitude errors and large separation distances. Also the practical implementation issues of this controller for the upcoming JC2Sat -FF mission are discussed .

Testing of Membrane Space Structure Shape Control Using Genetic Algorithm

This paper investigates the application of the genetic algorithm in active control of inflatable structures. The algorithm is used to search for the optimal tensions, which minimize membrane wrinkles. A genetic algorithm-based control system is developed using MATLAB, LabView, and Automation Manager. The control system is tested on a 200 x 300 mm rectangular Kapton membrane pulled by three tensions along each edge. Different tension combinations are exerted onto the membrane through SMA actuators. All tensions are calculated through the strains of a thin aluminum strip installed in the tension links. A vision system is developed to measure the membrane flatness under different tension combinations. Testing results show that the genetic algorithm works very well in finding the optimal tensions.

Neural network modeling of cold-gas thrusters for a spacecraft formation flying test-bed

This work presents a neural network based modeling strategy to precisely identify the thrusts of cold-gas thrusters deployed in a nano-satellite experimental test-bed developed at the Canadian Space Agency (CSA). Eight thrusters are used to control the planar motion of an emulated free-floating spacecraft supported by air-bearing. Calibration experiments conducted on these thrusters revealed that the generated thrusts are highly nonlinear with respect to their inputs, the digital openings and the air pressure. Motivated by the learning and approximation capabilities of artificial neural networks (ANNs), an ANN is used to model the nonlinear thruster behavior using experimental data. The performance of the proposed strategy is satisfactory and clearly demonstrated by the resulting high precision model.

Application of SMA in Membrane Structure Shape Control

Shape memory alloy (SMA) actuators have found a wide range of applications due to their unique properties such as high force, long stroke, small size, light weight, and silent operation, etc. However, their strong nonlinear properties make them a challenge to achieve accurate actuations. A simple control strategy is presented based on the idea of adjusting the SMA wire temperature as rapidly as possible. This strategy is simple, stable, and requires no hysteresis model or thermal model. The strategy is tested first on tracking displacement outputs, and effects of updating rate and input current on control accuracy are also discussed. It is then used for active shape control of a membrane structure model by adjusting its boundary tensions. Results indicate that under the developed control strategy, SMA wire actuators can offer very good accuracy in tracking displacement outputs and tension outputs. For the membrane structure shape control, the structure shape precision is improved greatly.

Testing of Inflatable-Structure Shape Control Using Genetic Algorithms and Neural Networks

I NFLATABLE structures have attractedmuch interest in the space community due to their unique advantages in achieving lowmass and high packaging efficiency [1,2]. We are currently working on an in-house research and development project of a membrane synthetic aperture radar antenna and solar array (see Fig. 1). It is expected that the membrane will be subjected to flatness problems during its lifetime in orbit, due to the thermal variation in space. A purely passive control method may not be sufficient to keep the membrane flat. Hence, an active control system is proposed to adjust the tensions according to the thermal variation. Actuators are installed in series with the links, such that the tensions stretching the membrane can be adjusted. A genetic algorithm and neural network (GA–NN) scheme is proposed for modeling and tension optimization. The neural network model of the membrane is established as a mapping from the boundary stretching tensions and space environment to membrane flatness. After the neural network training is completed, the membrane flatness can be estimated by inputting the measured stretching tensions and space environment data to the neural network model. Based on the neural network model, the genetic algorithm is applied to search for the optimal tensions that minimize the membrane wrinkles [3]. Experimental results demonstrate the effectiveness of the proposed scheme.

Differential Drag as a Means of Spacecraft Formation Control

This paper investigates the feasibility of using differential drag as a means of nano-satellite formation control. Differential drag is caused when the ballistic coefficients of the spacecraft in a formation are not equal. The magnitude of differential drag depends on the difference in ballistic coefficients and also the altitude of the spacecraft formation. AGIs Satellite Tool Kit is used initially to assess the magnitude of drifts caused due to differential drag for different altitudes. This information is then used to show that it is feasible to use differential drag for spacecraft formation control. A simple PID controller is then implemented that adjusts the cross sectional areas of the satellites such that the energies of the orbits remain equal. Results are presented that show that the control law can maintain the formation separation with reasonable accuracy.

Application of Shape Memory Alloy Actuators in Active Shape Control of Inflatable Space Structures

SMA (shape memory alloy) actuators have found a wide range of applications due to their unique properties such as high force, long stroke, small size, light weight, and silent operation, etc. However, their strong nonlinear properties make them a challenge to achieve accurate actuation. This paper presents a simple control strategy based on the idea of adjusting the SMA wire temperature as fast as possible. This strategy is simple, stable, and requires no hysteresis model or thermal model. This strategy is tested with displacement output, and effects of updating rate and input current on control accuracy are also discussed. This control strategy is then used for active shape control of inflatable space structures. Results indicate that under this control strategy, shape memory alloy wire actuators can offer very good accuracy, and for inflatable structure shape control, great improvement can be achieved

Development of GA-based control system for active shape control of inflatable space structures

This paper describes the development of a control system used for the active shape control of inflatable space structures. The genetic algorithm is utilized for the optimization of control variables. A vision system is implemented for the measurement of the structure shape. Shape memory alloy wire actuators are used to exert the obtained optimal tensions. The developed control system is then tested on a 200mm times 300mm rectangular Kapton membrane structure. The membrane is pulled by three tensions along each edge. Different combinations of the tensions produce various wrinkles on the membrane. Test results indicate that the developed control system works very well in improving the structure shape precision

Time-Optimal Low-Thrust Formation Maneuvering Using a Hybrid Linear/Nonlinear Controller

M AINTAINING two spacecraft in a formation with a conventional fuel-based propulsion system could be challenging in its own right. But this problem becomes more difficult when the propulsion system has limited capabilities or control actuation is available in only one or two axes and timeoptimal and fuel-optimal control is desired. Fortunately, for the spacecraft relative motion problem, because the relative equations of motion are coupled, motion in more than one direction can be controlled by control actuation in only one direction. In this Note, we formulate a hybrid linear/nonlinear controller that can efficiently maneuver a spacecraft formation by applying control only in the along-track direction. We also assume that the available control is very small in magnitude. This type of problem could be anticipated when dealing with low-thrust systems such as plasma thrusters, control with differential drag or solar radiation pressure, or any other system with very-low-thrust capability. One of the very early papers on underactuated control of a spacecraft formation with low-thrust capability by Leonard [1] presented an elegant algorithm to control a formation with only differential drag (assumed to be a constant and acting only in the along-track direction). Although the derived control was proven to be time-optimal control, it was not fuel-optimal. This was because differential drag was assumed to be a free resource with no need to economize it. However, for a spacecraft relying on expendables for control (plasmic thrusters), fuel economy would be highly desired. Some of the recent papers on this topic [2,3] also address the problem of spacecraft formation control with limited resources. The idea is to use linear or nonlinear controllers with a saturation function. The resultant controllers are robust and globally stable but not time-optimal. In the applied control field, the problem to control less frequently andmore efficiently ismore often an issue and elegant solutions exist for the same problem [4,5]. In this Note, we borrow some ideas discussed from [1,5] and apply them to the spacecraft formation-control problem. The result is a time-optimal controller that is easy to implement for underactuated formation control with limited resources. Although the controller is not proven to be fueloptimal, it is shown to consume less fuel than a conventional timeoptimal or a linear robust controller.