Neural Control Research Articles

Adaptive Neural Network Tracking Control of Robotic Manipulators Based on Disturbance Observer

This article presents an adaptive neural network (ANN) control scheme based on a disturbance observer that can achieve trajectory tracking control of robotic manipulators under external disturbances and dynamic model uncertainties. Firstly, an ANN controller based on full-state feedback is derived using the backstepping technique to achieve an online approximation of uncertainty. The integral sliding mode surface with a position error is introduced into the controller, which reduces the steady-state error of the system and enhances robustness. Then, a novel disturbance observer is designed to estimate both the approximation errors of the ANN and external disturbances, and to provide compensation for the controller, effectively suppressing the trajectory tracking errors caused by approximation errors and disturbances. Subsequently, the Lyapunov stability theory is utilized to demonstrate the stability of the developed control strategy and the boundedness of all closed-loop signals. Finally, numerical simulations are used to confirm the efficacy of the proposed control method.

Neural network control of fractional-order chaotic systems with unknown control direction

In this paper, neural network control of fractional order chaotic systems (FOCSs) with input saturation and unknown sign of the controller gain is addressed by employing the Nussbaum function, where neural networks are utilized to model system uncertainties. To get rid of the limitation that reaching phase should be active before sliding motion in the traditional sliding mode control, a stable sliding surface is constructed. Then, by using the integer order Nussbaum gain control method, a novel controller with neural network sliding mode variable structure is designed. Finally, the practicality of the designed method is confirmed by a simulation experiment.

Machine Learning-Assisted Speech Analysis for Early Detection of Parkinson's Disease: A Study on Speaker Diarization and Classification Techniques.

Parkinson's disease (PD) is a neurodegenerative disorder characterized by a range of motor and non-motor symptoms. One of the notable non-motor symptoms of PD is the presence of vocal disorders, attributed to the underlying pathophysiological changes in the neural control of the laryngeal and vocal tract musculature. From this perspective, the integration of machine learning (ML) techniques in the analysis of speech signals has significantly contributed to the detection and diagnosis of PD. Particularly, MEL Frequency Cepstral Coefficients (MFCCs) and Gammatone Frequency Cepstral Coefficients (GTCCs) are both feature extraction techniques commonly used in the field of speech and audio signal processing that could exhibit great potential for vocal disorder identification. This study presents a novel approach to the early detection of PD through ML applied to speech analysis, leveraging both MFCCs and GTCCs. The recordings contained in the Mobile Device Voice Recordings at King's College London (MDVR-KCL) dataset were used. These recordings were collected from healthy individuals and PD patients while they read a passage and during a spontaneous conversation on the phone. Particularly, the speech data regarding the spontaneous dialogue task were processed through speaker diarization, a technique that partitions an audio stream into homogeneous segments according to speaker identity. The ML applied to MFCCS and GTCCs allowed us to classify PD patients with a test accuracy of 92.3%. This research further demonstrates the potential to employ mobile phones as a non-invasive, cost-effective tool for the early detection of PD, significantly improving patient prognosis and quality of life.

Multi-Parameter Fuzzy-Based Neural Network Sensorless PMSM Iterative Learning Control Algorithm for Vibration Suppression of Ship Rim-Driven Thruster

Aiming to reduce motor speed estimation and torque vibration present in the permanent magnet synchronous motors (PMSMs) of rim-driven thrusters (RDTs), a position-sensorless control algorithm using an adaptive second-order sliding mode observer (SMO) based on the super-twisting algorithm (STA) is proposed. In which the sliding mode coefficients can be adaptively tuned. Similarly, an iterative learning control (ILC) algorithm is presented to enhance the robustness of the velocity adjustment loop. By continuously learning and adjusting the difference between the actual speed and given speed of RDT motor through ILC algorithm, online compensation for the q-axis given current of RDT motor is achieved, thereby suppressing periodic speed fluctuations during motor running. Fuzzy neural network (FNN) training can be used to optimize the STA-SMO and ILC parameters of RDT control system, while improving speed tracking accuracy. Finally, simulation and experimental verifications have been conducted on the vector control system based on the conventional PI-STA and modified ILC-STA. The results show that the modified algorithm can effectively suppress the estimated speed and torque ripple of RDT motor, which greatly improves the speed tracking accuracy.

Model-free based adaptive BackStepping-Super Twisting-RBF neural network control with α-variable for 10 DOF lower limb exoskeleton

Model-free based adaptive BackStepping-Super Twisting-RBF neural network control with α-variable for 10 DOF lower limb exoskeleton

Energy flexibility quantification of a tropical net-zero office building using physically consistent neural network-based model predictive control

Building energy flexibility plays a critical role in demand-side management for reducing utility costs for building owners and sustainable, reliable, and smart grids. Realizing building energy flexibility in tropical regions requires solar photovoltaics and energy storage systems. However, quantifying the energy flexibility of buildings utilizing such technologies in tropical regions has yet to be explored, and a robust control sequence is needed for this scenario. Hence, this work presents a case study to evaluate the building energy flexibility controls and operations of a net-zero energy office building in Singapore. The case study utilizes a data-driven energy flexibility quantification workflow and employs a novel data-driven model predictive control (MPC) framework based on the physically consistent neural network (PCNN) model to optimize the building energy flexibility. To the best of our knowledge, this is the first instance that PCNN is applied to a mathematical MPC setting, and the stability of the system is formally proved. Three scenarios are evaluated and compared: the default regulated flat tariff, a real-time pricing mechanism, and an on-site battery energy storage system (BESS). Our findings indicate that incorporating real-time pricing into the MPC framework could be more beneficial to leverage building energy flexibility for control decisions than the flat-rate approach. Moreover, adding BESS to the on-site PV generation improved the building self-sufficiency and the PV self-consumption by 17% and 20%, respectively. This integration also addresses model mismatch issues within the MPC framework, thus ensuring a more reliable local energy supply. Future research can leverage the proposed PCNN-MPC framework for different data-driven energy flexibility quantification types.

Using Convolutional Neural Network to Design and Predict the Forces and Kinematic Performance and External Rotation Moment of the Hip Joint in the Pelvis

In order to improve the dynamic and kinematic adaptability of the hip joint, this paper presented a control attitude and kinematics and torque of the hip joint with power based neural network control. The CNN neural network uses input data only from the limb designed by the medical software, and is trained by different natural and artificially altered step patterns of healthy individuals. This type of network has been used for deep learning to realize adaptive speed control, dynamic and motion attitude, as well as prediction of force and torque performance. Detailed movement and torque tests were performed using MIMICS and ANATOMY AND PHYSIOLOGY software, and the obtained data were checked and varied by a healthy person, and finally, the test results showed that the neural network control system was able to control the selection. It has a variable and high speed with proper adaptation in various conditions. Finally, MATLAB software was used to design and predict the data of the problem, and favorable results were obtained.

Application of LSTM neural network based UAV attitude control in attitude estimation

From now on, Unmanned Aerial Vehicle (UAV) technologies have become more mature, and UAV has been skillfully used in various regions under this mature technology. However, during UAV flight, they often encounter interference from external factors such as airflow, wind and temperature changes, which pose challenges to their stability and flight accuracy. This paper proposes an algorithm what is based on LSTM (Long, Short-Term Memory network), aiming that the flight attitude is affected when it is subject to external interference, while still ensuring the robustness of its flight performance in this state. In this paper, the optimization method of anti-jamming adaptive adjustment based on LSTM. In the meantime, a UAV dynamic model and neural network control based on dynamic modeling are established. Then article established the LSTM rule and designed the PID controller of LSTM. And to prove the effectiveness of the designed PID controller by pilot experiments with Matlab / simulated link simulation and flight experiments. Finally, step size response, autonomous tracking, robustness and anti-interference were tested by experimental simulations.

Improving the Quality of Industrial Robot Control Using an Iterative Learning Method with Online Optimal Learning and Intelligent Online Learning Function Parameters

It is inevitable that the characteristics of a robot system change inaccurately or cannot be accurately determined during movement and are affected by external disturbances. There are many adaptive control methods, such as the exact linearization method, sliding control, or neural control, to improve the quality of trajectory tracking for a robot’s motion system. However, those methods require a great deal of computation to solve the constrained nonlinear optimization problem. This article first presents some techniques for determining the online learning function parameters of an intelligent controller, including two circuits: the inner circuit is an uncertain function component estimator to compensate for the robot’s input, and the outer circuit is an iterative learning controller and does not use a mathematical model of the robot with optimal online learning function parameters. The optimal condition is based on the model in the time domain to determine the learning function parameters that change adaptively according to the sum of squared tracking errors of each loop. As for the intelligent online learning function parameters, they closely follow the general model to stabilize the robot system, based on the principle of intelligent estimation of the uncertainty component and total noise. This method is built on Taylor series analysis for the state vector and does not use a mathematical model of the system at all. It allows feedback linearization, as well as intelligent stabilization of the system. This article’s content uses a 2-DOF flat robot implemented on MatlabR2022b software to verify the theory. These findings indicate that superior tracking performance is achievable.

Intra- and interday reliability of sympathetic transduction to blood pressure in young, healthy adults.

Microneurographic recordings of muscle sympathetic nerve activity (MSNA) and the succeeding changes in beat-to-beat blood pressure (i.e., sympathetic transduction) provide important insights into the neural control of the circulation in humans. Despite its widespread use, the reliability of this technique remains unknown. Herein, we assessed the intra- and interday test-retest reliability of signal-averaging sympathetic transduction to blood pressure. Data were analyzed from 15 (9 M/6 F) young, healthy participants who completed two baseline recordings of fibular nerve MSNA separated by 60 min (intraday). The interday reliability was obtained in a subset of participants (n = 13, 9 M/4 F) who completed a follow-up MSNA study. Signal-averaging sympathetic transduction was quantified as peak change in diastolic (DBP) and mean arterial pressure (MAP) following a burst of MSNA. Analyses were also computed considering different MSNA burst sizes (quartiles of normalized MSNA) and burst patterns (singlets, couplets, triplets, and quadruplets+), as well as nonburst responses. Intraclass-correlation coefficients (ICCs) were used as the main reliability measure. Peak changes in MAP [intraday: ICC = 0.76 (0.30-0.92), P = 0.006; interday: ICC = 0.91 (0.63-0.97), P < 0.001] demonstrated very good to excellent reliability. Sympathetic transduction of MSNA burst size displayed moderate to very good reliability, though the reliability of MSNA burst pattern was poor to very good. Nonburst responses revealed poor intraday [ICC = 0.37 (-1.05 to 0.80), P = 0.21], but very good interday [ICC = 0.76 (0.18-0.93), P = 0.01] reliability. Intraday reliability measures were consistently lower than interday reliability. Similar results were obtained using DBP. Collectively, these findings provide evidence that the burst-triggering signal-averaging technique is a reliable measure of sympathetic transduction to blood pressure in young, healthy adults.NEW & NOTEWORTHY We found that signal-averaging sympathetic transduction to blood pressure displayed very good to excellent intra- and interday test-retest reliability in healthy, young adults. Reliability analyses according to muscle sympathetic burst size, burst pattern, and nonburst response were less consistent. Results were similar when using diastolic or mean arterial pressure in the transduction calculation. These findings suggest that the signal-averaging technique can be used with confidence to investigate sympathetic transduction to blood pressure in humans across time.

Sympathetic Neural Control at Rest and During the Cold Pressor Test in Patients With Heart Failure With Preserved Ejection Fraction.

We tested the hypothesis that patients with heart failure with preserved ejection fraction (HFpEF) would have greater muscle sympathetic nerve activity (MSNA) at rest and sympathetic reactivity during a cold pressor test compared with non-heart failure controls. Further, given the importance of the baroreflex modulation of MSNA in the control of blood pressure (BP), we hypothesized that patients with HFpEF would exhibit a reduced sympathetic baroreflex sensitivity. Twenty-eight patients with HFpEF and 44 matched controls (mean±SD: 71±8 versus 70±7 years; 9 men/19 women versus 16 men/28 women) were studied. BP, heart rate, and MSNA (microneurography) were measured during 6 to 10 minutes of supine rest and the 2-minute cold pressor test. Spontaneous sympathetic baroreflex sensitivity was assessed during supine rest. Patients with HFpEF had higher resting MSNA burst frequency (39±14 versus 31±12 bursts/min; P=0.020) and lower sympathetic baroreflex sensitivity (-2.83±0.76 versus -3.57±1.19 bursts/100 heartbeats/mm Hg; P=0.019) than controls, but burst incidence was not different between groups (56±19 versus 50±20 bursts/100 heartbeats; P=0.179). During the cold pressor test, increases in MSNA indices did not differ between groups (P=0.135-0.998), but patients had a smaller increase in diastolic BP (Δ4±6 versus Δ14±11 mm Hg; P<0.001) compared with controls. Despite augmented resting MSNA burst frequency, burst incidence was not significantly different between groups, and sympathetic baroreflex sensitivity was reduced in patients with HFpEF. Furthermore, patients had preserved sympathetic reactivity but attenuated diastolic BP responses during the cold pressor test. These data suggest that, during physiological stress, sympathetic reactivity is intact, but the peripheral pathway for sympathetic vasoconstriction may be impaired in HFpEF.

Command filter-based adaptive neural tracking control of nonlinear systems with multiple actuator constraints and disturbances

In this paper, the adaptive practical finite-time tracking control problem for a class of strictly feedback nonlinear systems with multiple actuator constraints is investigated using backstepping techniques and practical finite-time stability theory. The effects of deadband and saturated nonlinear constraints on the controller design of nonlinear systems are addressed by the equivalent transformation method. The problem of complexity explosion due to the derivatives of virtual control signals is solved by using the virtual control signals as inputs to the command filters and using the outputs of the command filters to perform the corresponding control tasks. An adaptive neural network tracking backstepping control strategy based on the command filter technique and the backstepping design algorithm is proposed by approximating an unknown nonlinear function using a neural network. The control strategy ensures the boundedness of all variables in the closed-loop system, and the output tracking error fluctuates in a small region near the origin. Finally, simulations verify the effectiveness of the control strategy designed in this paper.

Neural-network-based adaptive control of strict-feedback nonlinear systems with actuator faults: Event-triggered communications strategy

This article considers the neural network-based event-triggered adaptive control problem for a class of strict-feedback nonlinear impulsive systems with actuator faults and external disturbances. It is well known that reducing computational resources and carrier loads is a long-standing task in the field of adaptive control. However, the event-triggered scheme used in traditional research still consumes relatively more resources. To solve the problem, the control scheme in this paper only updates the controller and weight parameters at the event-triggered moments. Dedicated efforts are made to certificate the stability of the system through a new Lyapunov theory, in such a way that the system maintains its stability well even under uncertainties. Finally, two emulation examples are exploited to demonstrate the efficacy and applicability of the proposed approach.

Reinforcement learning for distributed transient frequency control with stability and safety guarantees

This paper proposes a reinforcement learning-based approach for optimal transient frequency control in power systems with stability and safety guarantees. Building on Lyapunov stability theory and safety-critical control, we derive sufficient conditions on the distributed controller design that ensure the stability and transient frequency safety of the closed-loop system. Our idea of distributed dynamic budget assignment makes these conditions less conservative than those in recent literature, so that they can impose less stringent restrictions on the search space of control policies. We construct neural network controllers that parameterize such control policies and use reinforcement learning to train an optimal one. Simulations on the IEEE 39-bus network illustrate the guaranteed stability and safety properties of the controller along with its significantly improved optimality.

Green energy-sourced AI-controlled multilevel UPQC parameter selection using football game optimization

The power quality (PQ) has been significantly affected by the integration of intermittent non-conventional sources (NCS) into the local distribution system in addition to the adoption of power electronic technologies to regulate non-linear loads. This article combines the H-bridge cascade five-level unified power quality conditioner (5L-UPQC) with the wind power generation system (WPGS), solar photovoltaic power generation system (SPVGS), and battery storage system (BSS) as an effective approach to address PQ problems. The utilization of the Levenberg–Marquardt backpropagation (LMBP)-trained Artificial neural network controller (ANNC) in the UPQC is recommended for generating appropriate reference signals for the converters. This eliminates the requirement for conventional complex conversions, such as abc, dq0, and αβ. Moreover, the artificial neuro-fuzzy interface system (ANFIS) is recommended for achieving a DC-link balance. Football game optimization (FBGO) is utilized to determine the optimal shunt and series filter characteristics. The major objectives of the proposed system are to reduce the current waveform irregularities, resulting in a decrease in the total harmonic distortion (THD), an enhancement in the power factor (PF), the mitigation of supply voltage imbalances and disturbances, and the maintenance of a steady direct-current link capacitor voltage (DLCV), despite the variations in the load, solar irradiation, and wind velocity. The efficiency of the suggested strategy is assessed using four case studies that involve different loads, variable wind velocities, and source voltage balancing conditions. Based on the simulation studies and obtained results, the suggested method significantly decreases the THD to values of 2.91%, 3.63%, 3.75%, and 3.50%. Additionally, it achieves a power factor of unity, which is considerably lower compared to other multilevel schemes that use the traditional symmetrical reference frame (SRF) and instantaneous reactive power (pq) methods. This design has been executed using the MATLAB/Simulink program.

A novel adaptive neural predefined time control for a class of nonlinear systems

AbstractIn this article, a novel adaptive neural predefined time control scheme is designed for a class of uncertain nonlinear systems. In the design procedure, the unknown smooth nonlinear functions are approximated by the neural networks (NNs). A neural adaptive predefined time controller is proposed based on NNs, adaptive backstepping technology, and predefined time stability theory. The designed control strategy includes a new adaptive updating law which ensures that the derivative of the Lyapunov function satisfies the predefined time stable form. The proposed control scheme ensures that all the signals in the closed‐loop system are predefined time stability. A practical simulation example validates the effectiveness of the designed controller.

Adaptive neural network anti-disturbance control of a cable-driven flexible-joint-based underwater vehicle manipulator system with dead-zone input nonlinearities via multiple neural observers

To achieve the requirements of lightweight, low energy consumption, and low inertia of an underwater vehicle manipulator system, a cable-driven manipulator is installed on the underwater vehicle to form a cable-driven flexible-joint-based underwater vehicle manipulator system (CDFJ–UVMS). The CDFJ–UVMS is a complex nonlinear system subject to model uncertainties, complex marine environment disturbances, and actuator dead-zone nonlinearity. To design track controllers, the CDFJ–UVMS dynamics is divided into two parts: known and unknown. Subsequently, a radial basis function neural network is adopted to approximate the unknown nonlinearity. A neural network performance observer is constructed, whose estimation error is then used to design a novel neural disturbance observer (NDO) to estimate the total disturbance. Finally, an adaptive neural network control method is proposed for the CDFJ–UVMS based on the NDO, neural network compensator, and neural performance observer. The stability of the closed-loop system is analyzed using the Lyapunov method. The proposed control algorithm is applied to a CDFJ–UVMS with two cable-driven joints and compared with other control methods to show the effectiveness of the proposed control algorithm.

Magnetically Suspended Control Sensitive Gyroscope Rotor High-Precision Deflection Decoupling Method using Quantum Neural Network and Fractional-Order Terminal Sliding Mode Control

To achieve high-precision deflection control of a Magnetically Suspended Control and Sensitive Gyroscope rotor under high dynamic conditions, a deflection decoupling method using Quantum Radial Basis Function Neural Network and fractional-order terminal sliding mode control is proposed. The convergence speed and time complexity of the neural network controller limit the control accuracy and stability of rotor deflection under high-bandwidth conditions. To solve the problem, a quantum-computing-based structure optimization method for the Radial Basis Function Neural Network is proposed for the first time, where the input and the center of hidden layer basis function of the neural network are quantum-coded, and quantum rotation gates are designed to replace the Gaussian function. The parallel characteristic of quantum computing is utilized to reduce the time complexity and improve the convergence speed of the neural network. On top of that, in order to further address the issue of input jitter, a fractional-order terminal sliding mode controller based on the Quantum Radial Basis Function Neural Network is designed, the fractional-order differential sliding mode surface and the fractional-order convergence law are proposed to reduce the input jitter and achieve finite-time convergence of the controller, and the Quantum Radial Basis Function Neural Network is used to approximate the residual coupling and external disturbances of the system, resulting in improving the rotor deflection control accuracy. The semi-physical simulation experiments demonstrate the effectiveness and superiority of the proposed method.

Prefrontal cortex activation while walking did not change but gait speed improved after a randomized physical therapy intervention

BackgroundHigher prefrontal cortex (PFC) activation while walking may indicate reduced gait automaticity.AimWe examine whether PFC activation during walking improves after training in older adults at risk for mobility disability.MethodsForty-two adults aged ≥ 65 participated in a randomized clinical trial (NCT026637780) of a 12-week timing and coordination physical therapy intervention to improve walking (n = 20 intervention, n = 22 active control). PFC activation was measured by functional near-infrared spectroscopy (fNIRS) during four walking tasks over 15 m, each repeated 4 times: even surface walking, uneven surface walking, even dual-task, uneven dual-task; dual-task was reciting every other letter of the alphabet while walking. Gait speed and rate of correct letter generation were recorded. Linear mixed models tested between arm differences in change of fNIRS, gait speed, and letter generation from baseline to follow-up (12-week, 24-week, and 36-week).ResultsIntervention arms were similar in mean age (74.3 vs. 77.0) and baseline gait speed (0.96 vs. 0.93 m/s). Of 24 comparisons of between arm differences in the fNIRS signals, only two were significant which were not supported by differences at other follow-up times or on other tasks. Gait speed, particularly during dual-task conditions, and correct letter generation did improve post-intervention but improvements did not differ by arm.Discussion and ConclusionsAfter training, PFC activation during walking generally did not improve and did not differ by intervention arm. Improvements in gait speed without increased PFC activation may point toward more efficient neural control of walking.

Accelerating neurotechnology development using an Agile methodology.

Novel bioelectronic medical devices that target neural control of visceral organs (e.g., liver, gut, spleen) or inflammatory reflex pathways are innovative class III medical devices like implantable cardiac pacemakers that are lifesaving and life-sustaining medical devices. Bringing innovative neurotechnologies early into the market and the hands of treatment providers would benefit a large population of patients inflicted with autonomic and chronic immune disorders. Medical device manufacturers and software developers widely use the Waterfall methodology to implement design controls through verification and validation. In the Waterfall methodology, after identifying user needs, a functional unit is fabricated following the verification loop (design, build, and verify) and then validated against user needs. Considerable time can lapse in building, verifying, and validating the product because this methodology has limitations for adjusting to unanticipated changes. The time lost in device development can cause significant delays in final production, increase costs, and may even result in the abandonment of the device development. Software developers have successfully implemented an Agile methodology that overcomes these limitations in developing medical software. However, Agile methodology is not routinely used to develop medical devices with implantable hardware because of the increased regulatory burden of the need to conduct animal and human studies. Here, we provide the pros and cons of the Waterfall methodology and make a case for adopting the Agile methodology in developing medical devices with physical components. We utilize a peripheral nerve interface as an example device to illustrate the use of the Agile approach to develop neurotechnologies.