Neural Compensation Research Articles

Enhanced PID control method based on unknown state estimation and neural network-based compensation

PID控制一直是应用最广泛的工业控制技术。但是，基本PID控制算法仅考虑了动态系统输入-输出的外部特性，缺乏对系统内部状态信息的运用，因而在控制非线性耦合动态系统时往往难以获得满意的控制性能。本文在不改变原有PID控制设置的基础上，提出一种通过未知状态估计与神经网络控制补偿的增强PID(En-PID)控制方法，可显著增强原有PID控制的性能。首先，采用扩展卡尔曼滤波技术对未知关键状态进行精确估计，并将状态估计值作为神经网络补偿控制器的输入，从而将基本PID控制算法并未使用的过程状态信息引入到控制输入；其次，采用K均值聚类算法求取历史输入数据的聚类中心，以此作为RBF神经网络补偿器隐含层的中心向量；构建关于控制误差熵的优化性能指标，采用梯度下降算法对神经网络�补偿器输出层权值向量进行优化和修正；最后，将RBF网络求取的补偿控制输入与原有PID控制器的基本控制输入进行综合，并共同作用于实际被控系统，实现高性能控制。理论分析和污水处理过程控制实验验证了所提方法的先进性和实用性。

Smooth Speed Control of Permanent Magnet Synchronous Machine Using Back Propagation Neural Network

Torque ripple is one of the most critical problems in PMSM system. In this paper, a neural network (NN) torque compensator is combined with a conventional extended state observer (ESO)-based active disturbance rejection controller (ADRC) system to suppress the torque ripple at wide machine operation speed range by generating the optimal current reference. The ESO is able to estimate and reject the low-frequency component in the torque ripple, while the remaining disturbances can be learned and compensated by the neural network. Compared with commonly used schemes, the proposed method does not need to analyze the influence of various sources of the torque ripple, such as the cogging torque, non-sinusoidal back-EMF, parameter variations, and unmodeled disturbances. In addition, the simple structure of the neural network helps reduce the computation time and save computer memory. The effectiveness of the proposed neural network compensator with both the rotor position and mechanical angular velocity as inputs is verified in the experiment under different operation speeds.

Atypical connectivity aids conversation in autism

It is well-established that individuals with autism exhibit atypical functional brain connectivity. However, the role this plays in naturalistic social settings has remained unclear. Atypical patterns may reflect core deficits or may instead compensate for deficits and promote adaptive behavior. Distinguishing these possibilities requires measuring the ‘typicality’ of spontaneous behavior and determining how connectivity relates to it. Thirty-nine male participants (19 autism, 20 typically-developed) engaged in 115 spontaneous conversations with an experimenter during fMRI scanning. A classifier algorithm was trained to distinguish participants by diagnosis based on 81 semantic, affective and linguistic dimensions derived from their use of language. The algorithm’s graded likelihood of a participant's group membership (autism vs. typically-developed) was used as a measure of task performance and compared with functional connectivity levels. The algorithm accurately classified participants and its scores correlated with clinician-observed autism signs (ADOS-2). In support of a compensatory role, greater functional connectivity between right inferior frontal cortex and left-hemisphere social communication regions correlated with more typical language behavior, but only for the autism group. We conclude that right inferior frontal functional connectivity increases in autism during communication reflect a neural compensation strategy that can be quantified and tested even without an a priori behavioral standard.

Semantic cognition in healthy ageing: Neural signatures of representation and control mechanisms in naming typical and atypical objects

Effective use of conceptual knowledge engages semantic representation and control processes to access information in a goal-driven manner. Neuropsychological findings of patients presenting either degraded knowledge (e.g., semantic dementia) or disrupted control (e.g., semantic aphasia) converge with neuroimaging evidence from young adults, and delineate the neural segregation of representation and control mechanisms. However, there is still scarce research on the neurofunctional underpinnings of such mechanisms in healthy ageing. To address this, we conducted an fMRI study, wherein young and older adults performed a covert naming task of typical and atypical objects. Three main age-related differences were found. As shown by age group and typicality interactions, older adults exhibited overactivation during naming of atypical (e.g., avocado) relative to typical concepts in brain regions associated to semantic representation, including anterior and medial portions of left temporal lobe (respectively, ATL and MTG). This provides evidence for the reorganization of neural activity in these brain regions contingent to the enrichment of semantic repositories in older ages. The medial orbitofrontal gyrus was also overactivated, indicating that the processing of atypical concepts (relative to typical items) taxes additional control resources in the elderly. Increased activation in the inferior frontal gyrus (IFG) was observed in naming typical items (relative to atypical ones), but only for young adults. This suggests that naming typical items (e.g., strawberry) taxes more on control processes in younger ages, presumably due to the semantic competition set by other items that share multiple features with the target (e.g., raspberry, blackberry, cherry). Together, these results reveal the dynamic nature of semantic control interplaying with conceptual representations as people grow older, by indicating that distinct neural bases uphold semantic performance from young to older ages. These findings may be explained by neural compensation mechanisms coming into play to support neurocognitive changes in healthy ageing.

Reviving the Dynamics of Attacked Reservoir Computers

Physically implemented neural networks are subject to external perturbations and internal variations. Existing works focus on the adversarial attacks but seldom consider attack on the network structure and the corresponding recovery method. Inspired by the biological neural compensation mechanism and the neuromodulation technique in clinical practice, we propose a novel framework of reviving attacked reservoir computers, consisting of several strategies direct at different types of attacks on structure by adjusting only a minor fraction of edges in the reservoir. Numerical experiments demonstrate the efficacy and broad applicability of the framework and reveal inspiring insights into the mechanisms. This work provides a vehicle to improve the robustness of reservoir computers and can be generalized to broader types of neural networks.

Manipulator Trajectory Optimization Using Reinforcement Learning on a Reduced-Order Dynamic Model with Deep Neural Network Compensation

This article reports the construction of an articulated manipulator’s hybrid dynamic model and trajectory planning and optimization of the manipulator using deep reinforcement learning (RL) on the dynamic model. The hybrid model was composed of a physical-based reduced-order dynamic model, linear friction and damping terms, and a deep neural network model to compensate for the nonlinear characteristics of the manipulator. The hybrid model then served as the digital twin of the manipulator for trajectory planning to optimize energy efficiency and operation speed by using RL while taking obstacle avoidance into consideration. The proposed strategy was simulated and experimentally validated. The energy consumption along paths was reduced and the speed was increased so the manipulator could achieve more efficient motion.

Persian Phone Recognition Using Acoustic Landmarks and Neural Network-based variability compensation methods

Persian Phone Recognition Using Acoustic Landmarks and Neural Network-based variability compensation methods

Interval type-2 fuzzy neural network-based adaptive compensation control for omni-directional mobile robot

Interval type-2 fuzzy neural network-based adaptive compensation control for omni-directional mobile robot

APOE Genotype Modifies the Association of Fusiform Gyrus Cerebral Metabolic Rate of Oxygen Consumption and Object Naming Performance.

The apolipoprotein E (APOE) ɛ4 allele confers risk for age and Alzheimer's disease related cognitive decline but the mechanistic link remains poorly understood. Blood oxygenation level dependent (BOLD) response in the fusiform gyrus (FG) during object naming appears greater among APOEɛ4 carriers even in the face of equivalent cognitive performance, suggesting neural compensation. However, BOLD is susceptible to known age and APOE-related vascular changes that could confound its interpretation. To address this limitation, we used calibrated fMRI during an object naming task and a hypercapnic challenge to obtain a more direct measure of neural function - percent change cerebral metabolic rate of oxygen consumption (%ΔCMRO2). Participants were 45 older adults without dementia (28 ɛ4-, 17 ɛ4+) between the ages of 65 and 85. We examined APOE-related differences in %ΔCMRO2 in the FG during object naming and the extent to which APOE modified associations between FG %ΔCMRO2 and object naming accuracy. Exploratory analyses also tested the hypothesis that %ΔCMRO2 is less susceptible to vascular compromise than are measures of %ΔCBF and %ΔBOLD. We observed a modifying role of APOE on associations between FG %ΔCMRO2 and cognition, with ɛ4 carriers (but not non-carriers) demonstrating a positive association between right FG %ΔCMRO2 and object naming accuracy. Results suggest that the relationship between neural function and cognition is altered among older adult APOEɛ4 carriers prior to the onset of dementia, implicating CMRO2 response as a potential mechanism to support cognition in APOE-related AD risk.

Latent trajectories of recent and delayed memory and their predictors: evidence from SHARE.

Cognitive decline is common in the old age, but some evidence suggests it may already occur during adulthood. Previous studies have linked age, gender, educational attainment, depression, physical activity, and social engagement to better cognitive performance over time. However, most studies have used global measures of cognition, which could mask subtle changes in specific cognitive domains. The aim of this study is to examine trajectories of recent and delayed memory recall from a variable-centered perspective, in order to elucidate the impact of age, gender, educational attainment, depression, physical activity, and social engagement on recent and delayed memory both at initial time and across a 10-year period. The sample was formed by 56,616 adults and older adults that participated in waves 4 to 8 of the Survey of Health, Aging and Retirement in Europe (SHARE). We used latent growth modeling to establish latent recent and delayed memory trajectories, and then tested the effects of the aforementioned covariates on the latent intercept and slopes. Results showed that both recent and delayed recall display a quadratic trajectory of decline. All covariates significantly explained initial levels of immediate and delayed recall, but only a few had statistically significant effects on the slope terms. We discuss differences between present results and those previously reported in studies using a person-centered approach. This study provides evidence of memory decline during adulthood and old adulthood. Further, results provide support for the neural compensation reserve theory.

Temperature Drift Compensation of a MEMS Accelerometer Based on DLSTM and ISSA.

In order to improve the performance of a micro-electro-mechanical system (MEMS) accelerometer, three algorithms for compensating its temperature drift are proposed in this paper, including deep long short-term memory recurrent neural network (DLSTM-RNN, short DLSTM), DLSTM based on sparrow search algorithm (SSA), and DLSTM based on improved SSA (ISSA). Moreover, the piecewise linear approximation (PLA) method is employed in this paper as a comparison to evaluate the impact of the proposed algorithm. First, a temperature experiment is performed to obtain the MEMS accelerometer's temperature drift output (TDO). Then, we propose a real-time compensation model and a linear approximation model for neural network methods compensation and PLA method compensation, respectively. The real-time compensation model is a recursive method based on the TDO at the last moment. The linear approximation model considers the MEMS accelerometer's temperature and TDO as input and output, respectively. Next, the TDO is analyzed and optimized by the real-time compensation model and the three algorithms mentioned before. Moreover, the TDO is also compensated by the linear approximation model and PLA method as a comparison. The compensation results show that the three neural network methods and the PLA method effectively compensate for the temperature drift of the MEMS accelerometer, and the DLSTM + ISSA method achieves the best compensation effect. After compensation by DLSTM + ISSA, the three Allen variance coefficients of the MEMS accelerometer that bias instability, rate random walk, and rate ramp are improved from 5.43×10-4mg, 4.33×10-5mg/s12, 1.18×10-6mg/s to 2.77×10-5mg, 1.14×10-6mg/s12, 2.63×10-8mg/s, respectively, with an increase of 96.68% on average.

Dynamic Modeling and Model-Based Control with Neural Network-Based Compensation of a Five Degrees-of-Freedom Parallel Mechanism

In this paper, a spatial parallel mechanism with five degrees of freedom is studied in order to provide a promising dynamic model for the control design. According to the inverse kinematics of the mechanism, the dynamic model is derived by using the Lagrangian method, and the co-simulation using MSC ADAMS and MATLAB/Simulink is adopted to verify the established dynamic model. Then the pre-trained deep neural network (DNN) is introduced to predict the real-time state of the end-effector of the mechanism. Compared to the traditional Newton’s method, the DNN method reduces the cost of the forward kinematics calculation while ensuring prediction accuracy, which enables the dynamic compensation based on feedback signals. Furthermore, the computed torque control with DNN-based feedback compensation is implemented for the trajectory tracking of the mechanism. The simulations show that, in the most complicated case that involves friction and external disturbance, the proposed controller has better tracking performance. The results indicate the necessity of dynamic modeling in the design of control with high precision.

Neural network adaptive compensation control based on the backstepping method for nonlinear systems with time-varying delays

ABSTRACT Control of nonlinear systems with time-varying delays is a common problem, particularly when the time-varying delays are unknown. This paper presents an adaptive compensation control method based on multiple neural networks. Firstly, considering the nonlinear systems with time-varying delays, a neural network method is adopted to approximate the unknown nonlinear structure with time-varying delays. On this basis, a backstepping controller is designed. Considering the unknown time-varying delays, the control effect of the neural network controller based on the backstepping method is not optimal. To deal with the problem, a double-layer adaptive compensator for the neural network is proposed. Furthermore, for the neural network adaptive compensation controller based on the backstepping method, the stability is also proved by a Lyapunov method. Finally, simulation results demonstrate the effectiveness of the scheme, especially when the nonlinear systems have an unknown time-varying delay.

Heritability of cognitive abilities and regional brain structures in middle-aged to elderly East Asians.

This study examined the single-nucleotide polymorphism heritability and genetic correlations of cognitive abilities and brain structural measures (regional subcortical volume and cortical thickness) in middle-aged and elderly East Asians (Korean) from the Gwangju Alzheimer's and Related Dementias cohort study. Significant heritability was found in memory function, caudate volume, thickness of the entorhinal cortices, pars opercularis, superior frontal gyri, and transverse temporal gyri. There were 3 significant genetic correlations between (i) the caudate volume and the thickness of the entorhinal cortices, (ii) the thickness of the superior frontal gyri and pars opercularis, and (iii) the thickness of the superior frontal and transverse temporal gyri. This is the first study to describe the heritability and genetic correlations of cognitive and neuroanatomical traits in middle-aged to elderly East Asians. Our results support the previous findings showing that genetic factors play a substantial role in the cognitive and neuroanatomical traits in middle to advanced age. Moreover, by demonstrating shared genetic effects on different brain regions, it gives us a genetic insight into understanding cognitive and brain changes with age, such as aging-related cognitive decline, cortical atrophy, and neural compensation.

SEMG-based Adaptive Cooperative Multi-mode Control of a Soft Elbow Exoskeleton Using Neural Network Compensation.

Soft rehabilitation exoskeletons have gained much attention in recent years, striving to assist the paralyzed individuals restore motor functions. However, it is a challenge to promote human-robot interaction property and satisfy personalized training requirements. This article proposes a soft elbow rehabilitation exoskeleton for the multi-mode training of disabled patients. An adaptive cooperative admittance backstepping control strategy combined with surface electromyography (sEMG)-based joint torque estimation and neural network compensation is developed, which can induce the active participation of patients and guarantee the accomplishment and safety of training. The proposed control scheme can be transformed into four rehabilitation training modes to optimize the cooperative training performance. Experimental studies involving four healthy subjects and four paralyzed subjects are carried out. The average root mean square error and peak error in trajectory tracking test are 3.18° and 5.68°. The active cooperation level can be adjusted via admittance model, ranging from 4.51 °/Nm to 10.99 °/Nm. In cooperative training test, the average training mode value and effort score of healthy subjects (i.e., 1.58 and 1.50) are lower than those of paralyzed subjects (i.e., 2.42 and 3.38), while the average smoothness score and stability score of healthy subjects (i.e., 3.25 and 3.42) are higher than those of paralyzed subjects (i.e., 1.67 and 1.71). The experimental results verify the superiority of proposed control strategy in improving position control performance and satisfying the training requirements of the patients with different hemiplegia degrees and training objectives.

Adaptive Anti-Saturation Prescribed-Time Control for Payload Retrieval of Tethered Space System

This work presents an adaptive anti-saturation prescribed-time control scheme for the stable payload retrieval of a two-body tethered space system following the desired libration state. The tethered payload is retrieved at the analytical speed function corresponding to the desired stable libration state. At the same time, the thrust at the payload is applied with the proposed adaptive anti-saturation prescribed-time control scheme to closely track the desired state to eliminate any deviation from the desired libration state due to external disturbances. The influence of the thrust saturation is minimized by a neural dynamics-based adaptive anti-saturation compensator that quickly drives the control system away from the saturated region. The Lyapunov stability of the proposed control scheme is proved analytically. Numerical simulations validate the effectiveness of the proposed prescribed-time strategy and the adaptive anti-saturation compensator.

Longitudinal trajectories of spectral power during sleep in middle-aged and older adults.

Age-related changes in sleep appear to contribute to cognitive aging and dementia. However, most of the current understanding of sleep across the lifespan is based on cross-sectional evidence. Using data from the Sleep Heart Health Study, we investigated longitudinal changes in sleep micro-architecture, focusing on whether such age-related changes are experienced uniformly across individuals. Participants were 2,202 adults (ageBaseline=62.40±10.38, 55.36% female, 87.92% White) who completed home polysomnography assessment at two study visits, which were 5.23years apart (range: 4-7years). We analyzed NREM and REM spectral power density for each 0.5Hz frequency bin, including slow oscillation (0.5-1Hz), delta (1-4Hz), theta (4-8Hz), alpha (8-12Hz), sigma (12-15Hz), and beta-1 (15-20Hz) bands. Longitudinal comparisons showed a 5-year decline in NREM delta (p<.001) and NREM sigma power density (p<.001) as well as a 5-year increase in theta power density during NREM (p=.001) and power density for all frequency bands during REM sleep (ps<0.05). In contrast to the notion that sleep declines linearly with advancing age, longitudinal trajectories varied considerably across individuals. Within individuals, the 5-year changes in NREM and REM power density were strongly correlated (slow oscillation: r=0.46; delta: r=0.67; theta r=0.78; alpha r=0.66; sigma: r=0.71; beta-1: r=0.73; ps<0.001). The convergence in the longitudinal trajectories of NREM and REM activity may reflect age-related neural de-differentiation and/or compensation processes. Future research should investigate the neurocognitive implications of longitudinal changes in sleep micro-architecture and test whether interventions for improving key sleep micro-architecture features (such as NREM delta and sigma activity) also benefit cognition over time.

Satellite Attitude Identification and Prediction Based on Neural Network Compensation

This paper proposed a new attitude determination method for low-orbit spacecraft. The attitude prediction accuracy is greatly improved by adding the unmodeled environmental torque to the dynamic equation. Specifically, the environmental torque extraction algorithm based on extended Kalman filter and series extended state observer is introduced, and the unmodeled part of dynamic is identified through the inverse dynamic model. Then, the collected data are analyzed and trained by a backpropagation neural network, resulting in an attitude-torque mapping network with compensation ability. The simulation results show that the proposed feedback attitude prediction algorithm can outperform standard methods and provide a high accurate picture of prediction and reliability with discontinuous measurement.

Heat Flux Identification of Aircraft Structure with Artificial Neural Network Compensation

Aerodynamic heating measurement of reusable hypersonic vehicles has always been an important aspect of hypersonic vehicle design. In this paper, a mechanistic-model-based heat flux identification method with artificial neural network (ANN) compensation is established to determine the spatially distributed heat flux of the aircraft structure. A one-dimensional heat conduction model is used to estimate heat flux by a robust and efficient algorithm integrating Tikhonov regularization with Levenberg–Marquardt method. The one-dimensional estimated heat flux has large errors for not considering multidimensional heat conduction effects. The proposed mechanistic-model-based method is then utilized to compensate the multidimensional heat conduction by ANN. The performance of the proposed method will be assessed by the determination of the heat flux of a two-dimensional plate and aircraft structure. Results show that compared with the one-dimensional inversion results, ANN compensation method can significantly improve the accuracy of estimated heat flux and is also applicable for larger levels of heat flux. The proposed compensation method is an effective technique to identify the nonuniform surface heat flux of multidimensional structures.

HE[formula omitted]LM-AD: Hierarchical and efficient attitude determination framework with adaptive error compensation module based on ELM network

Satellite attitude determination, as a crucial pre-processing technology in remote sensing earth observation, is significantly associated with the geometric accuracy of satellite stereo images. To compensate for the observation errors in raw attitude data for achieving accurate results, the conventional Kalman filter-based and filter-smoothing combination frameworks provide insufficient access to statistical features of data, resulting in inferior practical system performance. To conquer these drawbacks, a hierarchical and efficient attitude determination framework is proposed called HE2LM-AD, which consists of a simplified adaptive Kalman filtering module, a neural network-based system error compensation module, and a weighted attitude smoothing module. To be more specific, an adaptive noise covariance matrix update is exploited in the Extended Kalman Filter (EKF) iterations, and a simplification scheme based on Cholesky decomposition is employed to obtain coarse filter outputs efficiently. Following that, a dedicated Extreme Learning Machine (ELM)-based error compensation network is designed to capture the pattern of observation errors, providing robust compensation at lower overhead. To further pursue high-performance attitude determination results, we fused the outcomes of the aforementioned two components according to the weights achieved from adaptive learning. Experiment results on in-orbit satellite attitude simulation datasets demonstrate and validate the feasibility and effectiveness of our framework.