Timothy J. Gale

Direct adaptive fuzzy control with a self-structuring algorithm

This paper presents a direct self-structuring adaptive fuzzy control (DSAFC) scheme for affine nonlinear single-input-single-output systems. We show that the only restriction on the control gain is that it be positive. No upper bound on this gain nor its derivative needs to be known. From an initial fuzzy system with a small number of rules, the self-structuring algorithm adds membership functions and rules when needed. To limit the size of the fuzzy system from growing indefinitely, the self-structuring algorithm replaces old membership functions by new ones instead of adding more membership functions so that the number of rules never exceeds a predefined upper bound. The stability of the closed loop system is guaranteed using the Lyapunov synthesis approach. The proposed control scheme is demonstrated by application to an inverted pendulum and a magnetic levitation system.

Two-Mode Adaptive Fuzzy Control With Approximation Error Estimator

In this paper, we propose a two-mode adaptive fuzzy controller with approximation error estimator. In the learning mode, the controller employs some modified adaptive laws to tune the fuzzy system parameters and an approximation error estimator to compensate for the inherent approximation error. In the operating mode, the fuzzy system parameters are fixed, only the estimator is updated online. Mathematically, we show that the closed-loop system is stable in the sense that all the variables are bounded in both modes. We also establish mathematical bounds on the tracking error, state vector, control signal and the RMS error. Using these bounds, we show that controllers design parameters can be chosen to achieve desired control performance. After that, an algorithm to automatically switch the controller between two modes is presented. Finally, simulation studies of an inverted pendulum system and a Chuas chaotic circuit demonstrate the usefulness of the proposed controller.

Development of a Sailing Dinghy Simulator

A sailing dinghy simulator was developed that recreated the task of sailing in the laboratory. The simulator comprised the deck of a sailing dinghy, a pivoting support frame, a control system, in strumentation, a mathematical model of sailing dinghy dynamics and a computer simulation with a three-dimensional graphics display. Pneumatic rams were used to apply motion feedback. Esti mated dinghy performance data were calibrated using on-water recordings of boat speed. A sailors ability in sailing to windward was assessed using the simulator. The sailor performed typical sailing tasks, including tacking, hiking and judging lay- lines, and their activities were recorded for subse quent analysis. It was found that the simulator successfully simulated dinghy sailing and that the recorded data were useful for analysis of sailor performance. The simulator may be used in the training and assessment af sailors, ranging from the raw beginner through the experienced sailor.

Direct Adaptive Fuzzy Control with Less Restrictions on the Control Gain

In the adaptive fuzzy and neural network control field, there are two basic configurations: direct and indirect. It is well known that the direct configuration needs more restrictions on the control gain than the indirect configuration. In this paper, we propose a direct adaptive fuzzy controller with less restrictions on the control gain. Using an extension of the universal approximation theorem, we show that the only required constraint on the control gain is that its sign is known. We also show that using the approximation error estimator enhances performance. Finally, application to an inverted pendulum demonstrates the effectiveness of the proposed controller.

Analytical and Numerical Studies of Volume Measurement using Conductance Catheter Techniques

Calculation of intra-ventricular blood volume using the impedance electrode technique suffers from inaccuracy arising from irregularity in ventricular geometry and non-uniformity in imposed electric fields. This is particularly evident in the right ventricle. To investigate this problem, we present analytical and numerical solutions for the electric field in a conductor with asymmetrical source electrodes. The results show the effects of field non-uniformity, and a potential improvement in using dual-excitation electrodes. We plan to extend these techniques to develop improved electrode configurations for use in blood volume measurements in the right ventricle

Clinical evaluation of a novel adaptive algorithm for automated control of oxygen therapy in preterm infants on non-invasive respiratory support

Objective To evaluate the performance of a novel rapidly responsive proportional-integral-derivative (PID) algorithm for automated oxygen control in preterm infants with respiratory insufficiency. Design Interventional study of a 4-hour period of automated oxygen control compared with combined data from two flanking periods of manual control (4 hours each). Setting Neonatal intensive care unit. Participants Preterm infants (n=20) on non-invasive respiratory support and supplemental oxygen, with oxygen saturation (SpO2) target range 90%–94% (manual control) and 91%–95% (automated control). Median gestation at birth 27.5 weeks (IQR 26–30 weeks), postnatal age 8.0 (1.8–34) days. Intervention Automated oxygen control using a standalone device, receiving SpO2 input from a standard oximeter and computing alterations to oxygen concentration that were actuated with a modified blender. The PID algorithm was enhanced to avoid iatrogenic hyperoxaemia and adapt to the severity of lung dysfunction. Main outcome measure Proportion of time in the SpO2 target range, or above target range when in air. Results Automated oxygen control resulted in more time in the target range or above in air (manual 56 (48–63)% vs automated 81 (76–90)%, p<0.001) and less time at both extremes of oxygenation. Prolonged episodes of hypoxaemia and hyperoxaemia were virtually eliminated. The control algorithm showed benefit in every infant. Manual changes to oxygen therapy were infrequent during automated control (0.24/hour vs 2.3/hour during manual control), and oxygen requirements were unchanged (automated control period 27%, manual 27% and 26%, p>0.05). Conclusions The novel PID algorithm was very effective for automated oxygen control in preterm infants, and deserves further investigation.

Automated control of inspired oxygen for preterm infants: What we have and what we need

This review provides the first comprehensive technically focused image of algorithms developed for automation of inspired oxygen control in preterm infants. The paper has three main parts; the first provides an overview of the existing algorithms, the second presents the major design challenges of automation and the third proposes directions for future research and development of improved controllers. In the first section, the algorithms are classified in four categories, namely rule-based, proportional-integral-derivative, adaptive, and robust. The second section discusses variability in oxygenation, technologic shortcomings of infant monitoring and safety considerations as the three major challenges for designing automated controllers. The paper finally proceeds to suggest some future directions for improving automated oxygen control in the preterm infant. It suggests that based on the nature of the physiological system, an optimal algorithm must be capable of making continuous adjustments and it should be adaptive, including to alterations in severity of lung dysfunction and position on the oxygen saturation curve. It is also suggested that future controllers must utilise additional inputs for tasks such as oximeter signal validation and real-time prediction of hypoxic events.

A Relative Density Ratio-Based Framework for Detection of Land Cover Changes in MODIS NDVI Time Series

To improve statistical approaches for near real-time land cover change detection in nonGaussian time-series data, we propose a supervised land cover change detection framework in which a MODIS NDVI time series is modeled as a triply modulated cosine function using the extended Kalman filter and the trend parameter of the triply modulated cosine function is used to derive repeated sequential probability ratio test (RSPRT) statistics. The statistics are based on relative density ratios estimated directly from the training set by a relative unconstrained least squares importance Fitting (RULSIF) algorithm, unlike traditional likelihood ratio-based test statistics. We test the framework on simulated, synthetic, and real-world beetle infestation datasets, and show that using estimated relative density ratios, instead of assuming the individual density functions to be Gaussian or approximating them with Gaussian Kernels, in the RSPRT statistics achieves better performance in terms of accuracy and detection delay. We verify the efficiency of the proposed approach by comparing its performance with three existing methods on all the three datasets under consideration in this study. We also propose a simple heuristic technique that tunes the threshold efficiently in difficult cases of near real-time change detection, when we need to take three performance indices, namely, false positives, false negatives, and mean detection delay, into account simultaneously.

Assessment of validity and predictability of the FiO2–SpO2 transfer-function in preterm infants

In this paper an investigation of the gain, delay, and time-constant parameters of the transfer function describing the relation between fraction of inspired oxygen (FiO2) and oxygen saturation in the blood (SpO2) in preterm infants is presented. The parameters were estimated following FiO2 adjustments and goodness of fit was used to assess the validity of the model when using an assumed first-order transfer function. For responses identified to be first-order, the estimated parameters were then clustered to identify areas where they tended to be concentrated. Each group described an operating region of the transfer function; thus, predicting the right operating region could potentially assist a range-based robust inspired oxygen controller to provide more optimal control by adapting itself to different clusters. Accordingly, the samples were assigned labels based on their cluster associations and 14 features available at the time of each adjustment were used as inputs to an artificial neural network to classify the clustered samples. The validity study suggested that 37% of the adjustments were followed by first-order responses. Prediction studies on the first-order responses indicated that the clusters could be predicted with an average accuracy of 64% when the parameters were divided into two groups.

Implantable Blood Flow Measurement Techniques for Humans

A wide range of blood flow measurement techniques were reviewed in regard to development of an implantable blood flow sensor. These include indicator methods and techniques to measure velocity and stroke volume. Most techniques were found to be unsuitable for implantable use in humans. The intracardiac impedance method was favored, though significant methodological obstacles remain