Feedforward Mode Research Articles

Targeting a Versatile Actuator for EU-DEMO: Novel Control Scheme for Multisource Pellet Injector

Reactor plasma core fueling requires the injection of cryogenic pellets, most probably composed of a mixture of D2 and T2. Likely, pellet injection will be the most important actuator for plasma core density control. Therefore, pellet injection systems must be developed further that are capable of acting as actuator for density control. A novel control scheme is developed based on a centrifuge acceleration system. This scheme considers every available pellet launching slot and compares the current particle flux with the requested one. The response time is within the granularity of the available launching slots, in this case between 7 and 12 ms. First plasma experiments in feedforward mode showed excellent results, providing a good basis for upcoming plasma core density feedback control development activities.

Reliable Early Breast Cancer Detection using Artificial Neural Network for Small Data Set

This paper proposes a breast cancer detection module using Artificial Neural Network for small data set. The developed system consists of hardware and software. Hardware included UWB transceiver and a pair of home- made directional sensor/antenna. The software included a Graphical User Interface (GUI) and k-fold based feed-forward back propagation Neural Network module to detect the tumor existence, size and location along with soft interface between software and hardware. Forward scattering technique is used by placing two sensors diagonally opposite sides of a breast phantom. UWB pulses are transmitted from one side of phantom and received from other side, controlled by the software interface in PC environment. Firstly feed forward backpropagation neural network (FFBNN) is developed. Then, k-fold is combined with developed FFBNN for testing purpose. Four data sets are created where contains 125, 95, 65 and 30 data samples in 1st,2nd,3rd and 4th data set respectively. Collected received signals were then fed into the NN module for training, testing and validation. The process is done for all data sets separately. The system exhibits detection efficiency of tumor existence, location (x, y, z), and size were approximately 87.72%, 87.24%, 83.93% and 80.51% for 1st, 2nd, 3rd and 4th data set respectively. The proposed module is very practical with low-cost and user friendly. The developed breast cancer detection module can be used for large data samples as well as for minimum data samples.

Proximal tubule LPA1 and LPA2 receptors use divergent signaling pathways to additively increase profibrotic cytokine secretion.

Lysophosphatidic acid (LPA) increases platelet-derived growth factor-B (PDGFB) and connective tissue growth factor (CTGF) production and secretion by proximal tubule (PT) cells through LPA2 receptor-Gqα-αvβ6-integrin-mediated activation of transforming growth factor-β1 (TGFB1). LPA2, β6-integrin, PDGFB, and CTGF increase in kidneys after ischemia-reperfusion injury (IRI), coinciding with fibrosis. The TGFB1 receptor antagonist SD-208 prevents increases of β6-integrin, TGFB1-SMAD signaling, and PDGFB/CTGF expression after IRI and ameliorates fibrosis (Geng H, Lan R, Singha PK, Gilchrist A, Weinreb PH, Violette SM, Weinberg JM, Saikumar P, Venkatachalam MA. Am J Pathol 181: 1236-1249, 2012; Geng H, Lan R, Wang G, Siddiqi AR, Naski MC, Brooks AI, Barnes JL, Saikumar P, Weinberg JM, Venkatachalam MA. Am J Pathol 174: 1291-1308, 2009). We report now that LPA1 receptor signaling through epidermal growth factor receptor (EGFR)-ERK1/2-activator protein-1 cooperates with LPA2-dependent TGFB1 signaling to additively increase PDGFB/CTGF production and secretion by PT cells. Conversely, inhibition of both pathways results in greater suppression of PDGFB/CTGF production and secretion and promotes greater PT cellular differentiation than inhibiting one pathway alone. Antagonism of the LPA-generating enzyme autotaxin suppressed signaling through both pathways. After IRI, kidneys showed not only more LPA2, nuclear SMAD2/3, and PDGFB/CTGF but also increased LPA1 and autotaxin proteins, together with enhanced EGFR/ERK1/2 activation. Remarkably, the TGFB1 receptor antagonist SD-208 prevented all of these abnormalities excepting increased LPA2. SD-208 inhibits only one arm of LPA signaling: LPA2-Gqα-αvβ6-integrin-dependent production of active TGFB1 and its receptor-bound downstream effects. Consequently, far-reaching protection by SD-208 against IRI-induced signaling alterations and tubule-interstitial pathology is not fully explained by our data. TGFB1-dependent feedforward modulation of LPA1 signaling is one possibility. SD-208 effects may also involve mitigation of injury caused by IRI-induced TGFB1 signaling in endothelial cells and monocytes. Our results have translational implications for using TGFB1 receptor antagonists, LPA1 and LPA2 inhibitors concurrently, and autotaxin inhibitors in acute kidney injury to prevent the development of chronic kidney disease.

Feed-forward Modulation Technique for more Accurate Operation of Modular Multilevel Converters

Modular multilevel converters have become the prominent topology for medium- and high-voltage applications. The performance of these converters highly depends on the accuracy of the used modulation approach, for which the capacitor voltage of submodules (SM) are usually assumed to be equal. This article exhibits that ignoring the capacitor voltage differences among SMs adversely affects the system performance. This becomes more obvious the larger the capacitor voltage differences are. Hence, this article proposes a more accurate feedforward modulation approach that takes into account either the instantaneous capacitor voltage value and the real output voltage in the modulation stage. As a result, in applications where larger SM voltage differences are expected, the current distortion and control performance are improved. Particularly, switching–saving approaches benefit from this method as it enables their operation with reduced switching losses without the downsides of increased distortion due to capacitor voltage differences. The proposed approach is analyzed and compared with the nearest-level modulation and with the level-shift PWM. Simulations and experimental validation are presented to confirm the effectiveness of the proposed technique.

Teaching Exploration of Ideological and Political Construction of Control Theory for Mechanical EngineeringCourse Based on Feedback Control

The ideological and political construction of curriculum is a new requirement for the ideological and political work in colleges and universities in the current reform plan of higher education in China. The ideological and political teaching reform of professional courses is an important way to cultivate students' national feelings. This paper discusses the necessity of the ideological and political construction of the course through the teaching exploration of integrating the ideological and political key points into the course of control theory. Based on the basic knowledge of the course Mechanical Control Theory, this paper deals with the ideological and political points as the "interference" factors of the traditional closed-loop teaching mode, and proposes the feed-forward and feedback teaching mode. This teaching mode makes the ideological and political courses with disciplinary characteristics and provides new ideas for solving the practical problems of ideological and political courses; At the same time, the ideological and political elements contained in the knowledge points of mechanical control theory are deeply excavated to explore a feasible ideological and political teaching path of the course "Construction of Control Theory for Mechanical Engineering" based on the principle of feedforward and feedback control.

The thalamus integrates the macrosystems of the brain to facilitate complex, adaptive brain network dynamics.

The human brain is a complex, adaptive system comprised of billions of cells with trillions of connections. The interactions between the elements of the system oppose this seemingly limitless capacity by constraining the system's dynamic repertoire, enforcing distributed neural states that balance integration and differentiation. How this trade-off is mediated by the brain, and how the emergent, distributed neural patterns give rise to cognition and awareness, remains poorly understood. Here, I argue that the thalamus is well-placed to arbitrate the interactions between distributed neural assemblies in the cerebral cortex. Different classes of thalamocortical connections are hypothesized to promote either feed-forward or feedback processing modes in the cerebral cortex. This activity can be conceptualized as emerging dynamically from an evolving attractor landscape, with the relative engagement of distinct distributed circuits providing differing constraints over the manner in which brain state trajectories change over time. In addition, inputs to the distinct thalamic populations from the cerebellum and basal ganglia, respectively, are proposed to differentially shape the attractor landscape, and hence, the temporal evolution of cortical assemblies. The coordinated engagement of these neural macrosystems is then shown to share key characteristics with prominent models of cognition, attention and conscious awareness. In this way, the crucial role of the thalamus in mediating the distributed, multi-scale network organization of the central nervous system can be related to higher brain function.

Oncogenic mutant RAS signaling activity is rescaled by the ERK/MAPK pathway

Activating mutations in RAS are present in ~ 30% of human tumors, and the resulting aberrations in ERK/MAPK signaling play a central role in oncogenesis. However, the form of these signaling changes is uncertain, with activating RAS mutants linked to both increased and decreased ERK activation in vivo. Rationally targeting the kinase activity of this pathway requires clarification of the quantitative effects of RAS mutations. Here, we use live‐cell imaging in cells expressing only one RAS isoform to quantify ERK activity with a new level of accuracy. We find that despite large differences in their biochemical activity, mutant KRAS isoforms within cells have similar ranges of ERK output. We identify roles for pathway‐level effects, including variation in feedback strength and feedforward modulation of phosphatase activity, that act to rescale pathway sensitivity, ultimately resisting changes in the dynamic range of ERK activity while preserving responsiveness to growth factor stimuli. Our results reconcile seemingly inconsistent reports within the literature and imply that the signaling changes induced by RAS mutations early in oncogenesis are subtle.

Network intrusion detection using multi-architectural modular deep neural network

The exponential growth of computer networks and the adoption of new network-based technologies have made computer security an important challenge. With the emergence of new internet-connected devices, the attack surface is increasing for cyber intruders. Many intrusion detection systems attempt to detect known attacks using signatures in network traffic. In recent years, researchers used several machine learning techniques to detect network attacks without relying on these signatures. These techniques generally suffer from a high false-positive rate which is not acceptable for an industry-ready intrusion detection product. In this paper, we propose a multi-architectural modular deep neural network model to decrease the false-positive rate of anomaly-based intrusion detection systems. Our model consists of a feed-forward module, a stack of restricted Boltzmann machine module, and two recurrent modules, the output weights of these modules are fed to an aggregator module to produce the answer of the model. The experiments are performed using CSE-CIC-IDS2018 dataset, and final models can be used in an IDS for generating alerts or preventing new attacks. The experimental results show improvement in the detection of some types of network attacks with accuracy as high as 100% for network-level attacks compared to related works.

Computational evidence for nonlinear feedforward modulation of fusimotor drive to antagonistic co-contracting muscles

The sensorimotor integration during unconstrained reaching movements in the presence of variable environmental forces remains poorly understood. The objective of this study was to quantify how much the primary afferent activity of muscle spindles can contribute to shaping muscle coactivation patterns during reaching movements with complex dynamics. To achieve this objective, we designed a virtual reality task that guided healthy human participants through a set of planar reaching movements with controlled kinematic and dynamic conditions that were accompanied by variable muscle co-contraction. Next, we approximated the Ia afferent activity using a phenomenological model of the muscle spindle and muscle lengths derived from a musculoskeletal model. The parameters of the spindle model were altered systematically to evaluate the effect of fusimotor drive on the shape of the temporal profile of afferent activity during movement. The experimental and simulated data were analyzed with hierarchical clustering. We found that the pattern of co-activation of agonistic and antagonistic muscles changed based on whether passive forces in each movement played assistive or resistive roles in limb dynamics. The reaching task with assistive limb dynamics was associated with the most muscle co-contraction. In contrast, the simulated Ia afferent profiles were not changing between tasks and they were largely reciprocal with homonymous muscle activity. Simulated physiological changes to the fusimotor drive were not sufficient to reproduce muscle co-contraction. These results largely rule out the static set and α-γ coactivation as the main types of fusimotor drive that transform the monosynaptic Ia afferent feedback into task-dependent co-contraction of antagonistic muscles. We speculate that another type of nonlinear transformation of Ia afferent signals that is independent of signals modulating the activity of α motoneurons is required for Ia afferent-based co-contraction. This transformation could either be applied through a complex nonlinear profile of fusimotor drive that is not yet experimentally observed or through presynaptic inhibition.

Photonic frequency-octupling scheme for stable microwave generation based on two incoherent optical sources

In this paper, a real-time frequency-octupling photonic generator scheme that can realize stable microwave signal generation by using two incoherent optical sources without the support of phase-locked loops is successfully demonstrated, which proves that the improved feed-forward modulation is a feasible method for reducing phase noise. The principle of the proposed generator is illustrated by theory, simulation, and experiment. A 16GHz stable microwave signal with a bandwidth less than 0.2MHz is generated by using a 2GHz driving signal, and a 64GHz optical millimeter wave signal is obtained by using an 8GHz driving signal. Furthermore, a purity 320GHz terahertz wave signal is obtained by simulation based on the same principle and setup. The results show that the terahertz signal has good transmission characteristic and fine eye diagram in the 40.5km fiber link. In addition, to enhance the flexibility and feasibility of the generator, the feasible range of the modulation index and a signal enhancement method are also explored.

Attentive Relational State Representation in Decentralized Multiagent Reinforcement Learning.

In multiagent reinforcement learning (MARL), it is crucial for each agent to model the relation with its neighbors. Existing approaches usually resort to concatenate the features of multiple neighbors, fixing the size and the identity of the inputs. But these settings are inflexible and unscalable. In this article, we propose an attentive relational encoder (ARE), which is a novel scalable feedforward neural module, to attentionally aggregate an arbitrary-sized neighboring feature set for state representation in the decentralized MARL. The ARE actively selects the relevant information from the neighboring agents and is permutation invariant, computationally efficient, and flexible to interactive multiagent systems. Our method consistently outperforms the latest competing decentralized MARL methods in several multiagent tasks. In particular, it shows strong cooperative performance in challenging StarCraft micromanagement tasks and achieves over a 96% winning rate against the most difficult noncheating built-in artificial intelligence bots.

Multilevel Common-Ground Transformerless Inverter for Photovoltaic Applications

Transformerless inverters have attracted great attention since they became a compact and economic solution for photovoltaic (PV) generation systems. On the other hand, the common-mode (CM) leakage current issue needs to be addressed in order to meet safety requirements. This article proposes a new multilevel common-ground inverter for transformerless systems. The leakage current is eliminated by the electric connection between the grid neutral point and the PV negative terminal. Furthermore, the operation modes of the proposed inverter, as well as the design guidelines, are analyzed in detail. In addition, a feedforward modulation strategy to decouple the output variables from the voltage–capacitor oscillations is proposed. Experimental results are presented to verify the performance and practicability of the proposed inverter in grid-connected operation.

Postural control of arm and fingers through integration of movement commands.

Every movement ends in a period of stillness. Current models assume that commands that hold the limb at a target location do not depend on the commands that moved the limb to that location. Here, we report a surprising relationship between movement and posture in primates: on a within-trial basis, the commands that hold the arm and finger at a target location depend on the mathematical integration of the commands that moved the limb to that location. Following damage to the corticospinal tract, both the move and hold period commands become more variable. However, the hold period commands retain their dependence on the integral of the move period commands. Thus, our data suggest that the postural controller possesses a feedforward module that uses move commands to calculate a component of hold commands. This computation may arise within an unknown subcortical system that integrates cortical commands to stabilize limb posture.

Effects of spastic cerebral palsy on multi-finger coordination during isometric force production tasks.

In this study, we quantified changes in finger interdependence (enslaving), multi-finger synergies, and feedforward modulation of synergy properties (i.e., anticipatory synergy adjustment) during single- and multi-finger force production tasks in individuals with cerebral palsy (CP). Spastic diplegic CP and healthy control subjects performed sets of finger force production tasks by each of the hands, including maximal force production and submaximal quick pulse force production in an isometric condition. The framework of the uncontrolled manifold hypothesis was used to quantify the indices of multi-finger synergies and the anticipatory synergy adjustment (ASA). The CP group showed lower maximal forces and higher indices of finger interdependence (enslaving), while the indices of multi-finger synergies stabilizing total finger forces during stable force production were not different significantly compared to the controls. Further, the time of ASA for the CP group was not delayed. The CP group showed a significantly less drop in the synergy indices during the anticipatory and quick pulse phase compared to the control group, which was accompanied by larger co-contraction indices of the forearm muscles. These findings suggest that the function of assembling motor synergies for stable force production is not affected by CP, while the ability to modulate synergy properties may be impaired with CP partially due to spasticity. The spasticity presumably hampers the purposeful feedforward destabilization of the performance. The results suggest that quantification of multi-digit synergies may provide an alternative tool for quantitative assessment of impaired coordination in the CP individuals.

Optimizing optogenetic stimulation protocols in auditory corticofugal neurons based on closed-loop spike feedback

Objective. Optogenetics provides a means to probe functional connections between brain areas. By activating a set of presynaptic neurons and recording the activity from a downstream brain area, one can establish the sign and strength of a feedforward connection. One challenge is that there are virtually limitless patterns that can be used to stimulate a presynaptic brain area. Functional influences on downstream brain areas can depend not just on whether presynaptic neurons were activated, but how they were activated. Corticofugal axons from the auditory cortex (ACtx) heavily innervate the auditory tectum, the inferior colliculus (IC). Here, we sought to determine whether different modes of corticocollicular activation could titrate the strength of feedforward modulation of sound processing in IC neurons. Approach. We used multi-channel electrophysiology and optogenetics to record from multiple regions of the IC in awake head-fixed mice while optogenetically stimulating ACtx neurons expressing Chronos, an ultra-fast channelrhodopsin. To identify cortical activation patterns associated with the strongest effects on IC firing rates, we employed a closed-loop evolutionary optimization procedure that tailored the voltage command signal sent to the laser based on spike feedback from single IC neurons. Main results. Within minutes, our evolutionary search procedure converged on ACtx stimulation configurations that produced more effective and widespread enhancement of IC unit activity than generic activation parameters. Cortical modulation of midbrain spiking was bi-directional, as the evolutionary search procedure could be programmed to converge on activation patterns that either suppressed or enhanced sound-evoked IC firing rate. Significance. This study introduces a closed-loop optimization procedure to probe functional connections between brain areas. Our findings demonstrate that the influence of descending feedback projections on subcortical sensory processing can vary both in sign and degree depending on how cortical neurons are activated in time.

Feedforward Modulation for the Neutral-Point-Clamped Converter With Confined Capacitor Voltage Ripples and Reduced Switching Power Losses

This article presents a new modulation technique with feedforward compensation for the three-phase three-level neutral-point-clamped converter. With the proposed technique, the capacitor voltage ripples are allowed to vary within certain limits. This enables an optimized design of the converter since the maximum capacitor voltages are predefined. Furthermore, the proposed modulation technique is able to reduce switching power losses compared to modulation techniques that eliminate capacitor voltage ripples completely. The proposed technique is therefore a tradeoff solution between aforementioned techniques and traditional modulation techniques where the capacitor voltage ripples are not limited. In the proposed technique, if the capacitor voltages are within the tolerable specified range, all the phases switch in two consecutive voltage levels in steady state. When the capacitor voltages go beyond the specified limits, one of the three phases is chosen to switch in three levels to inject an appropriate neutral point current for capacitor voltage balance. As the capacitor voltages are allowed to oscillate, the modulation technique is provided with feedforward compensation to avoid producing low-frequency distortion on the output voltages of the converter. Experimental results are presented operating under different loading conditions.

Robust Attentional Aggregation of Deep Feature Sets for Multi-view 3D Reconstruction

We study the problem of recovering an underlying 3D shape from a set of images. Existing learning based approaches usually resort to recurrent neural nets, e.g., GRU, or intuitive pooling operations, e.g., max/mean poolings, to fuse multiple deep features encoded from input images. However, GRU based approaches are unable to consistently estimate 3D shapes given different permutations of the same set of input images as the recurrent unit is permutation variant. It is also unlikely to refine the 3D shape given more images due to the long-term memory loss of GRU. Commonly used pooling approaches are limited to capturing partial information, e.g., max/mean values, ignoring other valuable features. In this paper, we present a new feed-forward neural module, named AttSets, together with a dedicated training algorithm, named FASet, to attentively aggregate an arbitrarily sized deep feature set for multi-view 3D reconstruction. The AttSets module is permutation invariant, computationally efficient and flexible to implement, while the FASet algorithm enables the AttSets based network to be remarkably robust and generalize to an arbitrary number of input images. We thoroughly evaluate FASet and the properties of AttSets on multiple large public datasets. Extensive experiments show that AttSets together with FASet algorithm significantly outperforms existing aggregation approaches.

A Large Measurable Range Capacitance-to-Digital Converter for Smart Humidity Sensors.

This study aims to propose a capacitance-to-digital converter (CDC) based on a third-order cascade of integrators with a feed-forward (CIFF) incremental sigma-delta modulator for smart humidity sensor application. Disguised zoom-in technology was proposed to enlarge the measurable range of the CDC. The input range of the CDC was 0–388 pF. The proposed CDC was realized using 0.18 μm complementary metal-oxide-semiconductor technology. Results show that the CDC performs a 13-bit capacitance-to-digital conversion in 0.8 ms. The analog system consumes 169.7 μA from a 1.8 V supply, which corresponds to a figure of merit (FOM) of 3.0 nJ/step. The proposed CDC was combined with a HS1101 humidity sensor to demonstrate its incorporation in an overall system design. The resolution was 0.7% relative humidity (RH) over a range of 30%–90% RH.

A reinforcing HNF4-SMAD4 feed-forward module stabilizes enterocyte identity.

BMP/SMAD signaling is a crucial regulator of intestinal differentiation1–4. However, the molecular underpinnings of the BMP pathway in this context are unknown. Here, we characterize the mechanism by which BMP/SMAD signaling drives enterocyte differentiation. We establish that the transcription factor HNF4A acts redundantly with an intestine-restricted HNF4 paralog, HNF4G, to activate enhancer chromatin and upregulate the majority of transcripts enriched in the differentiated epithelium; cells fail to differentiate upon double knockout of both HNF4 paralogs. Furthermore, we show that SMAD4 and HNF4 function via a reinforcing feed-forward loop, activating each other’s expression and co-binding to regulatory elements of differentiation genes. This feed-forward regulatory module promotes and stabilizes enterocyte cell identity; disruption of the HNF4-SMAD4 module results in loss of enterocyte fate in favor of progenitor and secretory cell lineages. This intersection of signaling and transcriptional control provides a framework to understand regenerative tissue homeostasis, particularly in tissues with inherent cellular plasticity5.

Mixed local motion planning and tracking control framework for autonomous vehicles based on model predictive control

This study proposes a novel mixed motion planning and tracking (MPT) control framework for autonomous vehicles (AVs) based on model predictive control (MPC), which is made up of an MPC-based longitudinal motion planning module, a feed-forward longitudinal motion tracking module, and an MPC-based integrated lateral motion planning and tracking module. First, given the global reference path and the surroundings information obtained from onboard devices and V2X network, the longitudinal motion planning based on a vehicle kinematics model is applied to determine the local target path, the desired longitudinal acceleration, and velocity considering the longitudinal safety priority. Then, based on the planned target path and longitudinal velocity, the integrated lateral MPT module based on a 2 degree-of-freedom vehicle model is developed to determine the optimal steering angle while satisfying the multiple kinematics and dynamics constraints. Finally, based on the desired longitudinal acceleration and the steering angle, the longitudinal forces of tires are determined. More importantly, co-simulations under several typical scenarios between MATLAB/Simulink and CarSim are conducted, and the results demonstrate excellent performance of the proposed mixed framework in both planning and tracking and also its real-time implementation.