Discrete Feedback Research Articles

Single Versus Multiple Events Error Potential Detection in a BCI-Controlled Car Game With Continuous and Discrete Feedback.

This work aimed to find and evaluate a new method for detecting errors in continuous brain-computer interface (BCI) applications. Instead of classifying errors on a single-trial basis, the new method was based on multiple events (MEs) analysis to increase the accuracy of error detection. In a BCI-driven car game, based on motor imagery (MI), discrete events were triggered whenever subjects collided with coins and/or barriers. Coins counted as correct events, whereas barriers were errors. This new method, termed ME method, combined and averaged the classification results of single events (SEs) and determined the correctness of MI trials, which consisted of event sequences instead of SEs. The benefit of this method was evaluated in an offline simulation. In an online experiment, the new method was used to detect erroneous MI trials. Such MI trials were discarded and could be repeated by the users. We found that, even with low SE error potential (ErrP) detection rates, feasible accuracies can be achieved when combining MEs to distinguish erroneous from correct MI trials. Online, all subjects reached higher scores with error detection than without, at the cost of longer times needed for completing the game. Findings suggest that ErrP detection may become a reliable tool for monitoring continuous states in BCI applications when combining MEs. This paper demonstrates a novel technique for detecting errors in online continuous BCI applications, which yields promising results even with low single-trial detection rates.

Efficiency at maximum power of a discrete feedback ratchet

Efficiency at maximum power is found to be of the same order for a feedback ratchet and for its open-loop counterpart. However, feedback increases the output power up to a factor of five. This increase in output power is due to the increase in energy input and the effective entropy reduction obtained as a consequence of feedback. Optimal efficiency at maximum power is reached for time intervals between feedback actions two orders of magnitude smaller than the characteristic time of diffusion over a ratchet period length. The efficiency is computed consistently taking into account the correlation between the control actions. We consider a feedback control protocol for a discrete feedback flashing ratchet, which works against an external load. We maximize the power output optimizing the parameters of the ratchet, the controller, and the external load. The maximum power output is found to be upper bounded, so the attainable extracted power is limited. After, we compute an upper bound for the efficiency of this isothermal feedback ratchet at maximum power output. We make this computation applying recent developments of the thermodynamics of feedback-controlled systems, which give an equation to compute the entropy reduction due to information. However, this equation requires the computation of the probability of each of the possible sequences of the controller's actions. This computation becomes involved when the sequence of the controller's actions is non-Markovian, as is the case in most feedback ratchets. We here introduce an alternative procedure to set strong bounds to the entropy reduction in order to compute its value. In this procedure the bounds are evaluated in a quasi-Markovian limit, which emerge when there are big differences between the stationary probabilities of the system states. These big differences are an effect of the potential strength, which minimizes the departures from the Markovianicity of the sequence of control actions, allowing also to minimize the departures from the optimal performance of the system. This procedure can be applied to other feedback ratchets and, more in general, to other control systems.

Non-Invasive, Temporally Discrete Feedback of Object Contact and Release Improves Grasp Control of Closed-Loop Myoelectric Transradial Prostheses.

Human grasping and manipulation control critically depends on tactile feedback. Without this feedback, the ability for fine control of a prosthesis is limited in upper limb amputees. Although various approaches have been investigated in the past, at present there is no commercially available device able to restore tactile feedback in upper limb amputees. Based on the Discrete Event-driven Sensory feedback Control (DESC) policy we present a device able to deliver short-lasting vibrotactile feedback to transradial amputees using commercially available myoelectric hands. The device (DESC-glove) comprises sensorized thimbles to be placed on the prosthesis digits, a battery-powered electronic board, and vibrating units embedded in an arm-cuff being transiently activated when the prosthesis makes and breaks contact with objects. The consequences of using the DESC-glove were evaluated in a longitudinal study. Five transradial amputees were equipped with the device for one month at home. Through a simple test proposed here for the first time-the virtual eggs test-we demonstrate the effectiveness of the device for prosthetic control in daily life conditions. In the future the device could be easily exploited as an add-on to complement myoelectric prostheses or even embedded in prosthetic sockets to enhance their control by upper limb amputees.

Opportunistic Spectrum Access with Discrete Feedback in Unknown and Dynamic Environment：A Multi-agent Learning Approach

This article investigates the problem of opportunistic spectrum access in dynamic environment, in which the signal-to-noise ratio (SNR) is time-varying. Different from existing work on continuous feedback, we consider more practical scenarios in which the transmitter receives an Acknowledgment (ACK) if the received SNR is larger than the required threshold, and otherwise a Non-Acknowledgment (NACK). That is, the feedback is discrete. Several applications with different threshold values are also considered in this work. The channel selection problem is formulated as a non-cooperative game, and subsequently it is proved to be a potential game, which has at least one pure strategy Nash equilibrium. Following this, a multi-agent Q-learning algorithm is proposed to converge to Nash equilibria of the game. Furthermore, opportunistic spectrum access with multiple discrete feedbacks is also investigated. Finally, the simulation results verify that the proposed multi-agent Q-learning algorithm is applicable to both situations with binary feedback and multiple discrete feedbacks.

Total cost of operating an information engine

We study a two-level system controlled in a discrete feedback loop, modeling both the system and the controller in terms of stochastic Markov processes. We find that the extracted work, which is known to be bounded from above by the mutual information acquired during measurement, has to be compensated by an additional energy supply during the measurement process itself, which is bounded by the same mutual information from below. Our results confirm that the total cost of operating an information engine is in full agreement with the conventional second law of thermodynamics. We also consider the efficiency of the information engine as a function of the cycle time and discuss the operating condition for maximal power generation. Moreover, we find that the entropy production of our information engine is maximal for maximal efficiency, in sharp contrast to conventional reversible heat engines.

A Community-Based Approach to Developing a Mobile Device for Measuring Ambient Air Exposure, Location, and Respiratory Health.

In west Eugene (Oregon), community research indicates residents are disproportionately exposed to industrial air pollution and exhibit increased asthma incidence. In Carroll County (Ohio), recent increases in unconventional natural gas drilling sparked air quality concerns. These community concerns led to the development of a prototype mobile device to measure personal chemical exposure, location, and respiratory function. Working directly with the environmental justice (EJ) communities, the prototype was developed to 1) meet the needs of the community and 2) evaluate the use in EJ communities. The prototype was evaluated in three community focus groups (n = 25) to obtain feedback on the prototype and feasibility study design to evaluate the efficacy of the device to address community concerns. Focus groups were recorded and qualitatively analyzed with discrete feedback tabulated for further refinement. The prototype was improved by community feedback resulting in eight alterations/additions to software and instructional materials. Overall, focus group participants were supportive of the device and believed it would be a useful environmental health tool. The use of focus groups ensured that community members were engaged in the research design and development of a novel environmental health tool. We found that community-based research strategies resulted in a refined device as well as relevant research questions, specific to the EJ community needs and concerns.

Discrete and ultradiscrete models for biological rhythms comprising a simple negative feedback loop

Many biological rhythms are generated by negative feedback regulation. Griffith (1968) proved that a negative feedback model with two variables expressed by ordinary differential equations do not generate self-sustained oscillations. Kurosawa et al. (2002) expanded Griffith׳s result to the general type of negative feedback model with two variables. In this paper, we propose discrete and ultradiscrete feedback models with two variables that exhibit self-sustained oscillations. To obtain the model, we applied tropical discretization and ultradiscretization to a continuous model with two variables and then investigated its bifurcation structures and the conditions of parameters for oscillations. We found that when the degradation rate of the variables is lower than their synthesis rate, the proposed models generate oscillations by Neimark–Sacker bifurcation. We further demonstrate that the ultradiscrete model can be reduced to a Boolean system under some conditions.

Proof of Principle of a Brain-Computer Interface Approach to Support Poststroke Arm Rehabilitation in Hospitalized Patients: Design, Acceptability, and Usability

ObjectiveTo evaluate the feasibility of brain-computer interface (BCI)-assisted motor imagery training to support hand/arm motor rehabilitation after stroke during hospitalization. DesignProof-of-principle study. SettingNeurorehabilitation hospital. ParticipantsConvenience sample of patients (N=8) with new-onset arm plegia or paresis caused by unilateral stroke. InterventionsThe BCI-based intervention was administered as an “add-on” to usual care and lasted 4 weeks. Under the supervision of a therapist, patients were asked to practice motor imagery of their affected hand and received as a discrete feedback the movements of a “virtual” hand superimposed on their own. Such a BCI-based device was installed in a rehabilitation hospital ward. Main Outcome MeasuresFollowing a user-centered design, we assessed system usability in terms of motivation, satisfaction (by means of visual analog scales), and workload (National Aeronautics and Space Administration–Task Load Index). The usability of the BCI-based system was also evaluated by 15 therapists who participated in a focus group. ResultsAll patients successfully accomplished the BCI training. Significant positive correlations were found between satisfaction and motivation (P=.001, r=.393). BCI performance correlated with interest (P=.027, r=.257) and motivation (P=.012, r=.289). During the focus group, professionals positively acknowledged the opportunity offered by BCI-assisted training to measure patients' adherence to rehabilitation. ConclusionsAn ecological BCI-based device to assist motor imagery practice was found to be feasible as an add-on intervention and tolerable by patients who were exposed to the system in the rehabilitation environment.

Learning behaviors via human-delivered discrete feedback: modeling implicit feedback strategies to speed up learning

For real-world applications, virtual agents must be able to learn new behaviors from non-technical users. Positive and negative feedback are an intuitive way to train new behaviors, and existing work has presented algorithms for learning from such feedback. That work, however, treats feedback as numeric reward to be maximized, and assumes that all trainers provide feedback in the same way. In this work, we show that users can provide feedback in many different ways, which we describe as "training strategies." Specifically, users may not always give explicit feedback in response to an action, and may be more likely to provide explicit reward than explicit punishment, or vice versa, such that the lack of feedback itself conveys information about the behavior. We present a probabilistic model of trainer feedback that describes how a trainer chooses to provide explicit reward and/or explicit punishment and, based on this model, develop two novel learning algorithms (SABL and I-SABL) which take trainer strategy into account, and can therefore learn from cases where no feedback is provided. Through online user studies we demonstrate that these algorithms can learn with less feedback than algorithms based on a numerical interpretation of feedback. Furthermore, we conduct an empirical analysis of the training strategies employed by users, and of factors that can affect their choice of strategy.

Second laws for an information driven current through a spin valve.

We propose a physically realizable Maxwell's demon device using a spin valve interacting unitarily for a short time with electrons placed on a tape of quantum dots, which is thermodynamically equivalent to the device introduced by Mandal and Jarzynski [D. Mandal and C. Jarzynski, Proc. Natl. Acad. Sci. USA 109, 11641 (2012)]. The model is exactly solvable and we show that it can be equivalently interpreted as a Brownian ratchet demon. We then consider a measurement-based discrete feedback scheme, which produces identical system dynamics, but possesses a different second law inequality. We show that the second law for discrete feedback control can provide a smaller, equal, or larger bound on the maximum extractable work as compared to the second law involving the tape of bits. Finally, we derive an effective master equation governing the system evolution for Poisson distributed bits on the tape (or measurement times, respectively) and we show that its associated entropy production rate contains the same physical statement as the second law involving the tape of bits.

Discrete feedforward and feedback optimal tracking control for offshore steel jacket platforms

This paper is concerned with discrete feedforward and feedback optimal tracking control schemes for an offshore steel jacket platform subject to irregular wave force. By discretizing a dynamic model of the offshore steel jacket platform system, a discrete feedforward and feedback optimal tracking controller is developed first to attenuate the wave-induced vibration of the offshore platform. Then, for the case of the offshore platform with control delays, a discrete feedforward and feedback optimal tracking controller with memory is presented. The controllers can be designed by solving an algebraic Riccati equation and a Stein equation, respectively. It is found through simulation results that compared with the classic state feedback optimal tracking control schemes, the proposed control schemes are more efficient in attenuating the vibration of the offshore platform. In addition, compared with the feedforward and feedback optimal control schemes, the proposed control schemes require less control cost.

Discrete vs. continuous surface electromyographic interface control.

Over 50% of the 273,000 individuals with spinal cord injuries in the US have cervical injuries and are therefore unable to operate a keyboard and mouse with their hands. In this experiment, we compared two systems using surface electromyography (sEMG) recorded from facial muscles to control an onscreen keyboard. Both systems used five sEMG sensors to capture muscle activity during five distinct facial gestures that then mapped to five cursor commands: move left, move right, move up, move down, and click. One system used a discrete movement and feedback algorithm, in which the user would make one quick facial gesture, causing a corresponding discrete movement to an adjacent button. The other system was continuously updated and allowed the user to move in any 360(o) direction smoothly. Information transfer rates (ITRs) in bits per minute were high for both systems. Users of the continuous system showed significantly higher ITRs (average of 68.5; p <; 0.02) compared to users of the discrete system (average of 54.3 bits/min).

Humans can integrate feedback of discrete events in their sensorimotor control of a robotic hand.

Providing functionally effective sensory feedback to users of prosthetics is a largely unsolved challenge. Traditional solutions require high band-widths for providing feedback for the control of manipulation and yet have been largely unsuccessful. In this study, we have explored a strategy that relies on temporally discrete sensory feedback that is technically simple to provide. According to the Discrete Event-driven Sensory feedback Control (DESC) policy, motor tasks in humans are organized in phases delimited by means of sensory encoded discrete mechanical events. To explore the applicability of DESC for control, we designed a paradigm in which healthy humans operated an artificial robot hand to lift and replace an instrumented object, a task that can readily be learned and mastered under visual control. Assuming that the central nervous system of humans naturally organizes motor tasks based on a strategy akin to DESC, we delivered short-lasting vibrotactile feedback related to events that are known to forcefully affect progression of the grasp-lift-and-hold task. After training, we determined whether the artificial feedback had been integrated with the sensorimotor control by introducing short delays and we indeed observed that the participants significantly delayed subsequent phases of the task. This study thus gives support to the DESC policy hypothesis. Moreover, it demonstrates that humans can integrate temporally discrete sensory feedback while controlling an artificial hand and invites further studies in which inexpensive, noninvasive technology could be used in clever ways to provide physiologically appropriate sensory feedback in upper limb prosthetics with much lower band-width requirements than with traditional solutions.

Applied State Space Feedback Control of a Deep Drawing Process

This paper presents results from application of discrete state spacebased feedback control of the blank holder force (BHF) magnitude and distribution in the drawing ofa rectangular workpiece. By means of monitoring the flange draw-in during the punch stroke, the controller makes continuousadjustments to the BHF magnitude and distribution by individuallycontrolling the pressure of a hydraulic fluid in a number of cavitiesembedded in the blank holder plate. The adjustments are conducted onlineduring the punch stroke to minimize the deviation of the achieved flange draw-in from a predetermined reference.

Discrete and continuous auditory feedback based on pitch and spatial lateralization for human-machine-interface control

The purpose of this study was to investigate auditory-motor learning via discrete and continuous auditory feedback using pitch and spatial lateralization. Sixteen subjects used facial surface electromyography (sEMG) to control a human-machine-interface (HMI). The output of the HMI was a lateralized harmonic tone. The fundamental frequency and lateralization (left-right ear) changed with the sum and the difference of the bilateral muscle signals. For eight participants, these changes were continuous, whereas the other eight participants received discrete feedback, in which the frequency of the tone was one of nine possible semitones (from midi note #64 to #75) and the lateralization was either left, center or right. Participants trained over three days. A mixed-models analysis of variance showed a significant effect of learning over sessions and a trend for increased performance for the group utilizing discrete feedback. Overall, information transfer rates using this purely auditory feedback averaged 38.5 bit/min by day 3, which is similar to results from similar systems utilizing visual feedback. These results show that with minimal training, auditory feedback can provide usable HMI control.

Supporting driver headway choice: The effects of discrete headway feedback when following headway instructions

With specific headway instructions drivers are not able to attain the exact headways as instructed. In this study, the effects of discrete headway feedback (and the direction of headway adjustment) on headway accuracy for drivers carrying out time headway instructions were assessed experimentally. Two groups of each 10 participants (one receiving headway feedback; one control) carried out headway instructions in a driving simulator; increasing and decreasing their headway to a target headway of 2 s at speeds of 50, 80, and 100 km/h. The difference between the instructed and chosen headway was a measure for headway accuracy. The feedback group heard a sound signal at the moment that they crossed the distance of the instructed headway. Unsupported participants showed no significant difference in headway accuracy when increasing or decreasing headways. Discrete headway feedback had varying effects on headway choice accuracy. When participants decreased their headway, feedback led to higher accuracy. When increasing their headway, feedback led to a lower accuracy, compared to no headway feedback. Support did not affect driver's performance in maintaining the chosen headway. The present results suggest that (a) in its current form discrete headway feedback is not sufficient to improve the overall accuracy of chosen headways when carrying out headway instructions; (b) the effect of discrete headway feedback depends on the direction of headway adjustment.

Frame Rate-based Discrete Visual Feedback Pose Regulation: A Passivity Approach

This paper studies a visual feedback 3D pose regulation problem explicitly handling camera frame rates. Although numerous works have already tackled vision-based estimation/control problems by focusing on the limitation of measured output (2D visual information), almost all of them have assumed that visual measurements of a camera are continuously available. However, camera frame rates and image processing time are often non-negligible compared with other computation time. In view of this fact, we newly propose a discrete visual feedback pose regulation law under the situation that visual measurements are sampled due to frame rates. We first give a sufficient condition of frame rates to achieve a desired relative pose to a static target object. Then, the tracking performance for a moving target is analyzed via the notion of ultimate boundedness. The present analysis provides a guideline for the design of estimation/control gains. We finally show the significance and validity of this work through 3D simulation.

Platelet count control in immune thrombocytopenic purpura patient: optimum romiplostim dose profile

Patients with immune thrombocytopenic purpura (ITP), a disease characterized by abnormally low platelet count, are susceptible to excessive bleeding as a direct consequence. While the problem of low platelet count can be addressed fundamentally either by slowing down the rate of platelet destruction or by increasing platelet production, or both, one of the more effective means of treating ITP patients is to increase platelet production with romiplostim. However, current romiplostim treatment strategies tend to produce undesirable responses where platelet counts oscillate between dangerously low values and extremely high peaks, as a direct consequence of the complex nonlinear dynamics associated with platelet count regulation. The objective of this study is to determine the optimum dose profile of romiplostim for a specific ITP patient required to maintain a platelet count of 70×109/L. Using clinical data of the specific patient's platelet count obtained in response to a series of subcutaneously applied doses of romiplostim, a standard pharmacokinetics/pharmacodynamics (PKPD) model was developed, validated, and analyzed to obtain insight into the patient's physiological characteristics. The model was subsequently used to investigate the performance of three control strategies: “fixed dose” open-loop control, “variable dose” discrete PI feedback control, and “variable dose” model-based open-loop optimal control. The control strategies were implemented for weekly and bi-weekly treatment regimens. With both treatment frequencies, the fixed dose open-loop control strategy resulted in unacceptable sustained oscillating platelet count. PI feedback control and model-based optimal open-loop control led to stable platelet count profiles after approximately 50 days but only for weekly injections. In summary, a stable platelet count is more likely to be achieved consistently in the specific patient with weekly treatments. Bi-weekly treatments are less effective because, as we show, fundamental pharmaceutical characteristics of romiplostim make oscillations in platelet count unavoidable at this treatment frequency. The results show that model-based decisions determined using patient-specific mathematical models are potentially useful for designing better treatment regimens for ITP patients.

Stability and Neimark-Sacker bifurcation analysis for a discrete single genetic negative feedback autoregulatory system with delay

In this paper, we deal with a discrete single genetic negative feedback autoregulatory system with delay by using Euler method. Choosing the delay \(\tau \) as the bifurcation parameter and analyzing the associated characteristic equation corresponding to the unique positive fixed point, it is found that the stability of the positive equilibrium and Neimark-Sacker bifurcation may occur when \(\tau \) crosses some critical values. Then the explicit formula which determines the stability, direction, and other properties of bifurcating periodic solution is derived by using the center manifold theorem and normal form theory. Finally, in order to illustrate our theoretical analysis, numerical simulations are also included in the end.

Stable active running of a planar biped robot using Poincare map control

This work formulates the active limit cycles of bipedal running gaits for a compliant leg structure as the fixed point of an active Poincare map. Two types of proposed controllers stabilize the Poincare map around its active fixed point. The first one is a discrete linear state feedback controller designed with appropriate pole placement. The discrete-time control first uses purely constant torques during stance and flight phase, then discretizes each phase into smaller constant-torque intervals. The other controller is an invariant manifold based chaos controller: a generalized Ott, Grebogi and Yorke controller having a linear form and a nonlinear form. Both controllers can stabilize active running gaits on either even or sloped terrains. The efficiency of these controllers for bipedal running applications are compared and discussed.