Network-based Feedback Research Articles

Recent advances of ECG monitoring and webserver health monitoring applications: A review

In response to recent improvements in mobile technology and cloud computing-based real-time monitoring systems, numerous cloud-based healthcare services and devices have been developed. In the cloud system, it is feasible to collect and distribute medical data immediately from anywhere to medical specialists, and patients can receive network-based feedback. This paper reviews the most recent developments and trends in wearable technology and biomedical equipment (mostly ECG), as well as the development of a cloud-based solution for mobile device and web browser users. The system contains ECG enhancement, ECG quality grading, and ECG parameter extraction methods. A use case was utilized to showcase the system, in which a web server processed ECG data from a mobile phone in real-time. The system has been proved to be functional, accurate, and efficient. We review the primary principles underlying the operation of the ECG and its fundamental operation, as well as the recording and transmission paradigm for ECG signals. Detailed processes were also provided for the system components, their configuration, and the employed technologies. In conclusion, we examine the complete study and the project's future scope.

Offline supervised learning v.s. online direct policy optimization: A comparative study and a unified training paradigm for neural network-based optimal feedback control

This work is concerned with solving neural network-based feedback controllers efficiently for optimal control problems. We first conduct a comparative study of two prevalent approaches: offline supervised learning and online direct policy optimization. Albeit the training part of the supervised learning approach is relatively easy, the success of the method heavily depends on the optimal control dataset generated by open-loop optimal control solvers. In contrast, direct policy optimization turns the optimal control problem into an optimization problem directly without any requirement of pre-computing, but the dynamics-related objective can be hard to optimize when the problem is complicated. Our results underscore the superiority of offline supervised learning in terms of both optimality and training time. To overcome the main challenges, dataset and optimization, in the two approaches respectively, we complement them and propose the Pre-train and Fine-tune strategy as a unified training paradigm for optimal feedback control, which further improves the performance and robustness significantly. Our code is accessible at https://github.com/yzhao98/DeepOptimalControl.

To Use or Not to Use: Exploring Therapists' Experiences with Pre-Treatment EMA-Based Personalized Feedback in the TheraNet Project.

Using idiographic network models in psychotherapy has been a growing area of interest. However, little is known about the perceived clinical utility of network models. The present study aims to explore therapists' experiences with network model-based feedback within the context of the TheraNet Project. In total, 18 therapists who had received network-based feedback for at least 1 patient at least 2 months prior were invited to retrospective focus groups. The focus group questions related to how participation in the study influenced the therapeutic relationship, how the networks were used, and what might improve their clinical utility. The transcribed focus groups were analyzed descriptively using qualitative content analysis. Most therapists mentioned using the feedback to support their existingtheir case concept, while fewer therapists discussed the feedback directly with the patients. Several barriers to using the feedback were discussed, as well as various suggestions for how to make it more clinically useful. Many therapists reported skepticism with regards to research in the outpatient training center in general, though they were also all pleasantly surprised by being involved, having their opinions heard, and showing a readiness to adapt research to their needs/abilities. This study highlights the gap between researchers' and therapists' perceptions about what useful feedback should look like. The TheraNet therapists' interest in adapting the feedback and building more informative feedback systems signals a general openness to the implementation of clinically relevant research. We provide suggestions for future implementations of network-based feedback systems in the outpatient clinical training center setting.

571Effect of a computer network-based feedback program on antibiotic prescription rates of primary care physicians

Abstract Background Antibiotic overuse is one of the major prescription problems in rural China and a major risk factor for antibiotic resistance. Low antibiotic prescription rates can effectively reduce the risk of antibiotic resistance. Methods A cluster randomized crossover open controlled trial was conducted in 31 hospitals. These hospitals were randomly allocated to two groups to receive the intervention for three months followed by no intervention for three months in a random sequence. The feedback intervention information, which displayed the physicians’ antibiotic prescription rates and ranking, was updated every 10 days. The primary outcome was the 10-day antibiotic prescription rate of the physicians. Results There were 82 physicians in group 1 (intervention first followed by control) and 81 in group 2 (control first followed by intervention). Baseline comparison showed no significant difference in antibiotic prescription rate between the two groups (30.8% vs 35.2%, P-value = 0.07). At the crossover point, the relative reduction in antibiotic prescription rate was significantly higher among physicians in the intervention group than in the control group (33.1% vs 20.3%, P-value < 0.001). After a further 3 months, the rate of decline in antibiotic prescriptions was also significantly greater in the intervention group compared to the control group (14.2% vs 4.6%, P-value < 0.001). Conclusions A computer network-based feedback intervention can significantly reduce the antibiotic prescription rates of primary care outpatient physicians. Key messages The feedback intervention continuously affected their prescription behavior for up to six months.

Online deep neural network-based feedback control of a Lutein bioprocess

An online adaptive deep neural network (DNN) scheme has been introduced for the tracking control of a nonlinear bioprocess with uncertain internal dynamics. First, a detailed controllability analysis is conducted for the Lutein bioprocess to represent the bioprocess as a nonlinear system in affine form. Next, a controller consisting of a DNN-based function approximator is designed for the nonlinear Lutein production bioprocess. It is demonstrated that closed-loop tracking control of a bioprocess for a desired yield profile is possible only with two inputs. The set point trajectory to yield maximum Lutein production is shown by the proposed online adaptive deep NN controller. The proposed controller exhibits self-learning capability under closed loop condition, due to the online learning phase. In other words, no explicit offline learning phase is required and online learning is preferred due to lack of a priori training data for approximating complex nonlinear functions. Simulation results are provided to confirm the performance of the proposed approach.

Effect of a computer network-based feedback program on antibiotic prescription rates of primary care physicians: A cluster randomized crossover-controlled trial

ObjectiveAntibiotic overuse is one of the major prescription problems in rural China and a major risk factor for antibiotic resistance. Low antibiotic prescription rates can effectively reduce the risk of antibiotic resistance. We hypothesized that under a paperless, computer-based feedback system the rates of antibiotic prescriptions among primary care physicians can be reduced. MethodsA cluster randomized crossover open controlled trial was conducted in 31 hospitals. These hospitals were randomly allocated to two groups to receive the intervention for three months followed by no intervention for three months in a random sequence. The feedback intervention information, which displayed the physicians’ antibiotic prescription rates and ranking, was updated every 10 days. The primary outcome was the 10-day antibiotic prescription rate of the physicians. ResultsThere were 82 physicians in group 1 (intervention first followed by control) and 81 in group 2 (control first followed by intervention). Baseline comparison showed no significant difference in antibiotic prescription rate between the two groups (30.8% vs 35.2%, P-value=0.07). At the crossover point, the relative reduction in antibiotic prescription rate was significantly higher among physicians in the intervention group than in the control group (33.1% vs 20.3%, P-value<0.001). After a further 3 months, the rate of decline in antibiotic prescriptions was also significantly greater in the intervention group compared to the control group (14.2% vs 4.6%, P-value<0.001). The characteristics of physicians did not significantly determine the change in rate of antibiotic prescriptions. ConclusionA computer network-based feedback intervention can significantly reduce the antibiotic prescription rates of primary care outpatient physicians and continuously affected their prescription behavior for up to six months. Trial registrationChiCTR1900021823

Network Feedback Passivation of Passivity-Short Multi-Agent Systems

In this paper, we propose a network-optimization framework for the analysis of multi-agent systems with passive-short agents. We consider the known connection between diffusively-coupled maximally equilibrium-independent passive systems, and network optimization, culminating in a pair of dual convex network optimization problems, whose minimizers are exactly the steady-states of the closed-loop system. We propose a network-based regularization term to the network optimization problem and show that it results in a network-based feedback using only relative outputs. We prove that if the average of the passivity indices is positive, then we convexify the problem, passivize the agents, and that steady-states of the augmented system correspond to the minimizers of the regularized network optimization problem. We also suggest a hybrid approach, in which only a subset of agents sense their own output, and show that if the set is nonempty, then we can always achieve the same correspondence as above, regardless of the passivity indices. We demonstrate our results on a traffic model with non-passive agents and limited GNSS reception.

Deep Neural Network-Based Feedback Control for Dynamic Soaring of Unpowered Aircraft

Abstract Dynamic soaring is a bio-inspired maneuver to harvest energy from the wind gradient, which allows albatrosses to fly across the ocean without flapping their wings. Although the underlying dynamics is well-known, which can be represented as a fixed-wing aircraft, the mechanism or the control law that successively extracts energy from the unforeseen wind gradient remains in question. In this study, a deep neural network architecture and a feedback control law for the dynamic soaring maneuver are proposed based on the investigation of the mechanical energy extraction mechanism. To train the neural network, a bunch of data composed of state and control pairs is generated via trajectory optimization, which is slightly modified to deal with the problem considered in this study. Numerical result shows that the trained network-based feedback control law can perform the dynamic soaring maneuver in various wind profiles.

Dynamic neural network-based feedback linearization control of full-car suspensions using PSO

This paper proposes a nonlinear control approach using dynamic neural network-based input–output feedback linearization to resolve the inherent conflicting performance criteria for a full-car nonlinear electrohydraulic active vehicle suspension system. Particle swarm optimization is applied both for the dynamic neural network models’ trainings and the computation of the controllers’ parameters. The intelligent control scheme outperformed the passive vehicle suspension system and the benchmark particle swarm-optimized proportional+integral+derivative controller. Effectiveness and robustness of the proposed controller are demonstrated through simulations both in time- and frequency-domains.

Optimal and robust control of a class of nonlinear systems using dynamically re-optimised single network adaptive critic design

ABSTRACTFollowing the philosophy of adaptive optimal control, a neural network-based state feedback optimal control synthesis approach is presented in this paper. First, accounting for a nominal system model, a single network adaptive critic (SNAC) based multi-layered neural network (called as NN1) is synthesised offline. However, another linear-in-weight neural network (called as NN2) is trained online and augmented to NN1 in such a manner that their combined output represent the desired optimal costate for the actual plant. To do this, the nominal model needs to be updated online to adapt to the actual plant, which is done by synthesising yet another linear-in-weight neural network (called as NN3) online. Training of NN3 is done by utilising the error information between the nominal and actual states and carrying out the necessary Lyapunov stability analysis using a Sobolev norm based Lyapunov function. This helps in training NN2 successfully to capture the required optimal relationship. The overall architecture is named as ‘Dynamically Re-optimised single network adaptive critic (DR-SNAC)’. Numerical results for two motivating illustrative problems are presented, including comparison studies with closed form solution for one problem, which clearly demonstrate the effectiveness and benefit of the proposed approach.

"온라인 토론학습에서 사회연결망분석 기반 피드백이 상호작용 및 성취도에 미치는 영향"

"온라인 토론학습에서 사회연결망분석 기반 피드백이 상호작용 및 성취도에 미치는 영향"

Network-based modelling and active control for offshore steel jacket platform with TMD mechanisms

The network-based modelling and active control for an offshore steel jacket platform with an active tuned mass damper mechanism is investigated. A network-based dynamic model of the offshore platform is first established. A network-based state feedback control scheme is developed. Under this scheme, the corresponding closed-loop system is modelled by a system with an artificial interval time-varying delay. Then, a delay-dependent stability criterion for the corresponding closed-loop system is derived. Based on this stability criterion, a sufficient condition on the existence of the network-based controller is obtained. It is found through simulation results that (i) both the oscillation amplitudes of the offshore platform and the required control force under the network-based state feedback controller are smaller than those under the nonlinear controller and the dynamic output feedback controller; (ii) the oscillation amplitudes of the offshore steel jacket platform under the network-based feedback controller are almost the same as the ones under the integral sliding mode controller, while the required control force by the former is smaller than the one by the latter.

NARMA-L2 Control of a Nonlinear Half-Car Servo-Hydraulic Vehicle Suspension System

In this paper, the performance of a nonlinear, 4 degrees-of-freedom, servo- hydraulic half-car active vehicle suspension system is compared with that of a passive vehicle suspension system with similar model parameters. The active vehicle suspension system is controlled by an indirect adaptive Neural Network-based Feedback Linearization controller (NARMA-L2). Hydraulic actuator force tracking is guaranteed by an inner Proportional+Integral+Derivative-based force feedback control loop. The output responses of the vehicles are presented and analyzed in the frequency and time domains, in the presence of model uncertainties in the form of variation in vehicle sprung mass loading. The results show that the NARMA-L2-based active vehicle suspension system performed better than the passive vehicle suspension system within the constraints.

Direct adaptive neural control of antilock braking systems incorporated with passive suspension dynamics

A direct adaptive neural network-based feedback linearization (NNFBL) slip control scheme for an antilock braking system (ABS) is presented. The NNFBL slip controller is developed to minimise the vehicle braking distance and to simultaneously improve its overall ride comfort and road handling. The comprehensive vehicle model incorporates the passive suspension dynamics, the dynamics of the electro-mechanical based braking system and air drag and wheel bearing friction. A feedforward, multilayer perceptron (MLP) neural network (NN) model that is well suited for control by discrete input-output linearization (NNIOL) is selected to represent the ABS with passive suspension. The NN model was trained using Levenberg-Marquardt optimization algorithm. The controlled signal was further boosted using a genetic algorithm generated gain. The effectiveness of the proposed controller is demonstrated by simulation results, in the presence of deterministic road disturbance input to the suspension and varying road conditions. The results are superior with respect to braking distance minimization and also to reference slip tracking, especially on the dry asphalt road.

Evaluation of a Neural Network-Based Control Strategy for a Cost-Effective Externally-Powered Prosthesis

ABSTRACT This paper presents a control strategy that compensates for the nonlinearity in the inexpensive sensors and hardware of a cost effective prosthetic hand. The control strategy uses neural network-based force control and sensory feedback to detect disturbance induced by slippage. The neural network approach is chosen over other nonlinear models because it is easy to implement and it offered the additional advantage of having its parameters easily adjusted over the life span of the device. The proposed strategy was evaluated on a functional multi-digit underactuated prosthetic hand. The initial and incremental forces exerted from each finger were adjusted to balance the amount of disturbance and the deformation of the objects. Experiments were conducted to test the performance of the protocol in situations encountered in activities of daily living. The displacement of each object under three grasping configurations was measured as a performance criterion while the object's mass was changed. The results showed that with the adjusted parameters for each grasping configuration, the control strategy was able to detect the dynamic changes in mass of the object and was also able to successfully adjust the grasping force before the object drops from the hand.

Design of a robust neural network-based tracking controller for a class of electrically driven nonholonomic mechanical systems

This paper addresses the problem of designing robust tracking controls for a class of uncertain nonholonomic systems actuated by brushed direct current (DC) motors. This class of electrically driven nonholonomic mechanical systems can be perturbed by plant uncertainties, unmodeled time-varying perturbations, and external disturbances. An adaptive neural network-based dynamic feedback tracking controller will be developed such that all the states and signals of the closed-loop system are bounded and the trajectory tracking error is as small as possible. Consequently, for practical applications, the intelligent robust tracking control scheme developed here can be employed to handle a broader class of electrically driven nonholonomic systems in the presence of high-degree time-varying uncertainties. Finally, simulation examples are presented to demonstrate the effectiveness of the proposed control algorithms.

Network-based feedback control for systems with mixed delays based on quantization and dropout compensation

This paper deals with the problem of feedback control for networked systems with discrete and distributed delays subject to quantization and packet dropout. Both a state feedback controller and an observer-based output feedback controller are designed. The infinite distributed delay is introduced in the discrete networked domain for the first time. Also, it is assumed that system state or output signal is quantized before being communicated. Moreover, a compensation scheme is proposed to deal with the effect of random packet dropout through communication network. Sufficient conditions for the existence of an admissible controller are established to ensure the asymptotical stability of the resulting closed-loop system. Finally, a numerical example is given to illustrate the proposed design method in this paper.

Neuro-fuzzy control of antilock braking system using sliding mode incremental learning algorithm

A neuro-fuzzy adaptive control approach for nonlinear dynamical systems, coupled with unknown dynamics, modeling errors, and various sorts of disturbances, is proposed and used to design a wheel slip regulating controller. The implemented control structure consists of a conventional controller and a neuro-fuzzy network-based feedback controller. The former is provided both to guarantee global asymptotic stability in compact space and as an inverse reference model of the response of the controlled system. Its output is used as an error signal by an incremental learning algorithm to update the parameters of the neuro-fuzzy controller. In this way the latter is able to gradually replace the conventional controller from the control of the system. The proposed new learning algorithm makes direct use of the variable structure systems theory and establishes a sliding motion in terms of the neuro-fuzzy controller parameters, leading the learning error toward zero. In the simulations and in the experimental studies, it has been tested on the control of antilock breaking system model and the analytical claims have been justified under the existence of uncertainty and large nonzero initial errors.

SCFNN-Based Decision Feedback Equalization Robust to Frequency Offset and Phase Noise

This paper proposes a self-constructing fuzzy neural network-based decision feedback equalizer (SCFNN DFE). An online learning algorithm containing the structure and parameter learning phases is employed in training the SCFNN DFE. Specifically, the feedforward input vector classification and a gradient-descent method are both used in this online learning algorithm. We show by simulations that the proposed SCFNN DFE offers improvement compared to the traditional DFE methods in the presence of frequency offset and phase noise.

Fast self-constructing fuzzy neural network-based decision feedback equaliser in time-invariant and time-varying channels

A fast self-constructing fuzzy neural network-based decision feedback equaliser (FSCFNN DFE) is proposed. Without estimating the channel, a fast learning algorithm containing the structure and parameter learning phases is employed to the FSCFNN DFE. Both the partition of the feedforward input space and the gradient descent method are used simultaneously with the aid of decision feedback inputs in this fast learning procedure. The performance of FSCFNN DFE is compared with traditional non-linear equalisers in both time-invariant and time-varying channels. The reduced complexity and excellent performance of the FSCFNN DFE make it suitable for severely distorted channel equalisation.