Model-based Feedback Research Articles

Model-based dynamic feedback control of a planar soft robot: trajectory tracking and interaction with the environment

Leveraging the elastic bodies of soft robots promises to enable the execution of dynamic motions as well as compliant and safe interaction with an unstructured environment. However, the exploitation of these abilities is constrained by the lack of appropriate control strategies. This work tackles for the first time the development of closed-loop dynamic controllers for a continuous soft robot. We present two architectures designed for dynamic trajectory tracking and surface following, respectively. Both controllers are designed to preserve the natural softness of the robot and adapt to interactions with an unstructured environment. The validity of the controllers is proven analytically within the hypotheses of the model. The controllers are evaluated through an extensive series of simulations, and through experiments on a physical soft robot capable of planar motions.

An RBF-ARX Model-Based Variable-Gain Feedback RMPC Algorithm

The RBF-ARX model has been used intensively in modeling and control of nonlinear systems, in which the coefficients of the NARX model are approximated with RBF networks. In this paper, motivated by the fact that the state feedback control policy with variable-gain feedback can support more freedom for the design of RMPCs, we propose an RBF-ARX model-based variable-gain feedback RMPC synthesis method. First, a polytopic state space model construction method is designed, in which the variation rate information of the model parameters is also utilized to improve accuracy of the system model prediction. And then, a robust variable-gain feedback predictive control algorithm is designed to enlarge design freedom and improve control performance. Finally, the verification of the feasibility and effectiveness of our RMPC is conducted on a CSTR process.

Electrorotation of Arbitrarily Shaped Micro-Objects: Modeling and Experiments

In this article, we study the electrorotational behavior of nonspherical micro-objects. We extend a control-oriented model of general dielectrophoresis to incorporate also the hydrodynamics so that we can predict the motion of nonspherical micro-objects in fluidic environments. Such a mathematical (computational) model enables model-based feedback control of a position and orientation of particles by real-time (online) computation of voltages applied to the electrodes. We use the measured data from experiments with electrorotation of an artificial micro-object having a Tetris-like shape to evaluate the performance of the proposed model. We also demonstrate the qualitative difference in behavior from the commonly performed electrorotation of a sphere advocating the necessity for model-based control. Further analysis of the simulation results for other than the experimentally explored scenarios provides additional useful insight into the electrorotational behavior of nonspherical objects.

Enhancing total fracture surface area in naturally fractured unconventional reservoirs via model predictive control

In naturally fractured unconventional reservoirs, naturally present fractures will interact with hydraulic fractures, and divert fracture propagation. Because of complex fracture growth, the ultimate goal of hydraulic fracturing operation in naturally fractured unconventional reservoirs should be changed from achieving the desired fracture geometry to maximizing the total fracture surface area (TFSA) for given fracturing resources, as it will allow more drainage area available for oil recovery. Unfortunately, there are no such techniques available to develop pumping schedules to maximize TFSA for given fracturing resources in naturally fractured unconventional reservoirs. Motivated by this, in this work, we will develop a new model-based pumping schedule by utilizing a recently developed unconventional complex fracture propagation model called Mangrove as our virtual experiment describing complex fracture networks by accounting for the interaction between hydraulic fractures and natural fractures. Initially, using the simulation data from Mangrove, a reduced-order model (ROM) is constructed, which is then used to develop a Kalman filter utilizing the available measurement to estimate the unmeasurable ROM states. Then, we propose a model-based feedback control system to determine the fracturing fluid pumping schedule that maximizes TFSA, which will lead to an enhanced oil production rate from naturally fractured unconventional reservoirs. We demonstrate that by using the proposed control scheme, TFSA can be greatly enhanced which will lead to an oil production rate greater than those of existing pumping schedules which were developed without considering natural fractures.

Model-Based Nonlinear Control of the Dielectric Elastomer Actuator With High Robustness and Precision

Abstract Dielectric elastomers (DEs) is one of the promising artificial muscle for soft robots and flexible devices. As one of the key issues for practical applications, the control of DE actuators remains challenging due to the large actuation, electromechanical coupling, and viscoelastic dissipation. Feedforward control and proportional integral derivative (PID) feedback control are recently studied for the control of DE actuators. The control performance is still limited due to the complex dynamic behavior of DE actuators with both nonlinearities and modeling uncertainties. This paper proposes a model-based feedback control for DE actuator, considering nonlinearity of large deformation, electromechanical coupling, and the modeling uncertainties. A nonlinear motion model is proposed and verified by parameter identification experiments. Based on the nonlinear model, we demonstrate a robust control strategy including nonlinear model compensation and robust feedback to decrease the tracking error. The experimental results verify that the control strategy possesses excellent validity to the DE actuator with improved performance compared to the previous strategy of feedforward and PID feedback control. The system design and control strategy of this paper may guide the future design and application of DE actuators, soft robots, and flexible devices.

An approach to develop haptic feedback control reference for steering systems using open-loop driving manoeuvres

In this paper, a methodology to capture the model-based haptic feedback control reference for closed-loop steering systems is demonstrated. The parameterisation is based on the measurements of open-loop driving manoeuvres for the inertia-spring-damper-friction reference model. The steady-state and transient manoeuvres are used to identify the model parameters. The reference model is limited to the haptic feedback of driver excitation in the linear vehicle handling range and intended to be used in closed-loop steering control strategies. The model parameters have an intuitive interpretation that allows to be used in both admittance and impedance control setting. The feasibility of the proposed model is demonstrated in a validated simulation environment for electric power assisted steering and on a real hardware for the steer-by-wire force-feedback case.

Lattice hydrodynamic model-based feedback control method with traffic interruption probability

Connected vehicles are expected to become commercially available by the next decade, while traffic interruption is not uncommon in the real traffic environment. In this paper, we propose a feedback control method for lattice hydrodynamic model considering the traffic interruption probability effect. The stability criterion of the new model is explored through linear stability analysis of transfer function. When the stability conditions are not satisfied, a delay feedback controller is used to control the discharging flow to suppress traffic congestion. The impact of gain coefficient and delay time on the performance is discussed. We verify the effectiveness of the devised delay feedback controller by simulations. Results show that the traffic interruption probability effect has a considerable impact on the stability of traffic flow, while the controller is effective in suppressing traffic congestion.

Fuzzy affine model-based output feedback controller design for nonlinear impulsive systems

This paper studies the robust output feedback control problem for a class of nonlinear systems with impulsive effects via T-S fuzzy affine models. By employing a state-input augmentation scheme, some new results on the piecewise affine (PWA) output feedback controller synthesis for T-S fuzzy affine impulsive systems are obtained through piecewise quadratic Lyapunov functions. Sufficient criteria for exponential stability analysis of the resulting closed-loop system are given within a convex optimization setup. A simulation example is shown to verify the effectiveness of the proposed approach.

Modeling and control of cell wall thickness in batch delignification

In the Kraft pulping process, the cell wall thickness (CWT) of wood chips significantly affects paper properties. This work proposes a novel multiscale model by combining the mass continuity and thermal energy balance equations adopted from a modified “extended Purdue model” with a kinetic Monte Carlo algorithm to describe the microscopic events such as the evolution of CWT and Kappa number; otherwise, these microscopic properties of wood chips are not accessible by existing measurement techniques. To reduce the model complexity, a data-driven reduced-order model is developed, which is then used for developing a soft sensing system (i.e., Kalman filter) to estimate the primary measurements (e.g., CWT, Kappa number) by utilizing the secondary measurements (e.g., active effective alkali and dissolved lignin concentrations in the free-liquor phase). Based on this soft sensing system, a model-based feedback control framework is developed to regulate both the CWT and Kappa number of wood chips.

A Model-Based Approach to Detection Limits in Studying Environmental Exposures and Human Fecundity.

Human exposure to persistent environmental pollutants often results in concentrations with a range of values below the laboratory detection limits. Growing evidence suggests that inadequate handling of concentrations below the limit of detection (LOD) may bias assessment of health effects in relation to environmental exposures. We seek to quantify such bias in models focusing on the day-specific probability of pregnancy during the fertile window and propose a model-based approach to reduce such bias. A multivariate skewed generalized t-distribution constrained by the LOD is assumed for the chemical concentrations, which realistically represents the underlying distribution. A latent variable-based framework is used to model fecundibility, which nonlinearly relates conception probability to chemical concentrations, daily intercourses, and other important covariates. The advantages of the proposed approach include the use of multiple chemical concentrations to aid the estimation of left censored chemical exposures, as well as the model-based feedback mechanism for fecundibility outcome to inform the estimations, and an adequate handling of model uncertainty through a joint modeling framework. A Markov chain Monte Carlo sampling algorithm is developed for implementing the Bayesian computations and the logarithm of pseudo-marginal likelihood measure is used for model choices. We conduct simulation studies to demonstrate the performance of the proposed approach and apply the framework to the Longitudinal Investigation of Fertility and the Environment study which evaluates the effects of exposures to environmental pollutants on the probability of pregnancy. We found that p,p'-DDT is negatively associated with the day-specific probability of pregnancy.

Simultaneous measurement uncertainty reduction and proppant bank height control of hydraulic fracturing

In hydraulic fracturing, higher fracturing fluid injection rates can trigger increased stress, thereby creating more microseismic events; particularly, simultaneously occurring multiple microseismic events can reduce measurement errors. This suggests a new state and output estimation scheme that utilizes the dependence between the fracturing fluid injection rate (i.e., manipulated input) and measurement errors. Motivated by this, we propose a novel control framework for measurement uncertainty reduction while achieving the original control task of proppant bank height control in hydraulic fracturing. Initially, using the simulation data from the high-fidelity model of hydraulic fracturing, a reduced-order model is constructed to design a Kalman filter. Then, a model-based feedback control system is proposed to regulate the uniformity of proppant bank height along the optimal fracture length and achieve accurate state and output estimation by manipulating the fracturing fluid pumping schedule that includes the fracturing fluid injection rate and proppant concentration at the wellbore.

Model-based robust H∞ control of a granulation process using Smith predictor with Reference updating

Model-based feedback control is developed for a continuous granulation process addressing the challenge of time delay and physics-based input-output constraints. The process plant is a multi-input multi-output (MIMO) linear model with time delay. A robust H∞ controller is designed using the mixed sensitivity loop shaping design. A framework has been laid down to insure the robustness of the Smith predictor by incorporating the model mismatch as an additive uncertainty in the predictor’s structure. The control performance and robustness is assessed by simulations for regulation and reference tracking problems. We show significant performance gains by employing a Smith predictor and the technique of reference updating: The control is coping significantly better with time delay, physical constraints and model mismatch. The proposed control approach is more efficient as compared to other widely used methods such as model predictive control (MPC); obtaining a stable behaviour of the response and control effort while forcing them to remain within the desired bounds.

Learning Feedforward Control of a One-Stage Refrigeration System

Refrigeration control is usually realized by means of model-based feedback controllers, which requires high-computational load and time-consuming model identification efforts. The implementation of feedback control requires a compromise between performance and robust stability. Considering these difficulties, an online learning operation controller for one-stage refrigeration cycle is presented, which consists of two components: a model-based feedback component and a learning feedforward component. The feedback controller is utilized to guarantee robustness. Meanwhile, the optimized performance is reached by the learning feedforward controller including a one-hidden-layer structure with B-spline basis functions. The comparison results of benchmark problems validate the effectiveness of this strategy and show that a perfect tracking performance can still be achieved without extensive modeling.

Identification and Accommodation of Control Actuator Faults Using Multi-rate Sampled Measurements

Abstract This work presents a model-based approach for the detection, estimation and accommodation of control actuator faults in dynamic processes controlled using multi-rate sampled state measurements. Initially, a model-based state feedback controller that employs an inter-sample model predictor to compensate for the unavailability of continuous state measurements is designed. Characterizations of both the stability and performance properties of the multi-rate sampled-data closed-loop system are obtained and expressed in terms of the measurement sampling rates, the magnitudes of the faults, and the various predictor and controller design parameters. A parameter estimator that estimates the magnitudes of the faults by minimizing the error between the estimated and sampled states over a moving horizon is then introduced and used to determine the fault or health status of the control actuators at the sampling times. The identification of faults triggers a fault accommodation scheme that adjusts selected predictor and controller design parameters to simultaneously preserve closed-loop stability and optimize a pre-defined performance metric. The implementation of the developed approach is illustrated using a simulated model of a solid oxide fuel cell system subject to both stability and performance-degrading actuator faults. A discussion of some of the intricacies of the multi-rate sampling scheme and how they impact the practical implementation of the fault identification and accommodation schemes is presented.

DEES: A Class of Data-Enabled Robust Feedback Algorithms for Real-Time Optimization

The increasing availability of information-rich data sets offers an invaluable opportunity to complement and improve the performance of existing model-based feedback algorithms. Following this principle, in this paper we present a novel class of Data-Enabled Extremum Seeking (DEES) algorithms for static maps, which make use of current and recorded data in order to solve a convex optimization problem characterized by a variational inequality. The optimization dynamics synergistically combine ideas from concurrent learning and classic neuro-adaptive extremum seeking in order to dispense with the assumption of requiring a persistence of excitation condition in the closed-loop system. Using analytical tools for nonlinear systems we show that for a general class of optimization dynamics it is possible to tune the parameters of the controller to guarantee convergence in finite time to an arbitrarily small neighborhood of the set of optimizers. The results are illustrated in a scalar constrained minimization problem.

Model-based strategies for sensor fault accommodation in uncertain dynamic processes with multi-rate sampled measurements

This work presents model-based strategies for the accommodation of sensor faults in dynamic processes with multi-rate sampled state measurements. The developed strategies account explicitly for both closed-loop stability and performance considerations. Initially, a model-based feedback control system, in which a model predictor compensates for the unavailability of state measurements between sampling times, is designed. The stability and performance characteristics of the multi-rate sampled-data closed-loop system are analyzed and explicitly characterized in terms of the state sampling rates, the fault parameters, and the various process, model and controller design parameters. The resulting characterizations provide insight into the robustness and margins of tolerable faults that can be accommodated, and are used to develop both stability-based and performance-based fault accommodation schemes that aim to maintain closed-loop stability and minimize the degradation of closed-loop performance in the presence of sensor faults. Two types of sensor faults are considered. These include faults that manifest themselves as improper sensor readings, as well as faults that cause drift in the sensor sampling rate. The first type of fault introduces errors in the model state updates at the sampling times, while the second type of fault alters the rate at which the sensors sample the process states. The developed methods are illustrated through a case study involving a cascade of two non-isothermal continuous-stirred tank reactors with plant-model mismatch and access to full state measurements that are sampled at different rates. The case study gives some insight into the effects of sensor faults on the stability and performance of the sampled-data state feedback control system, and how these effects can be mitigated through use of fault-tolerant control.

Biological pest control using a model-based robust feedback.

Biological control is the artificial manipulation of natural enemies of a pest for its regulation to densities below a threshold for economic damage. The authors address the biological control of a class of pest population models using a model-based robust feedback approach. The proposed control framework is based on a recursive cascade control scheme exploiting the chained form of pest population models and the use of virtual inputs. The robust feedback is formulated considering the non-linear model uncertainties via a simple and intuitive control design. Numerical results on three pest biological control problems show that the proposed model-based robust feedback can regulate the pest population at the desired reference via the manipulation of a biological control action despite model uncertainties.

Spacecraft attitude control using model-based disturbance feedback control strategy

In this paper, the performance of the spacecraft attitude control system is enhanced using model-based disturbance feedback control (DFC) strategy in the presence of disturbance. This control strategy is applied to a single-axis spacecraft attitude control with thruster, reaction wheel, and magnetic torqrod actuators, separately. An anti-windup observer-based modified PI-D is utilized for each actuation system as a main controller. The controller gains are tuned using genetic algorithm when the time average of absolute value of pointing error is chosen as an objective function. The performance of DFC with the modified PI-D controller is investigated under disturbance and model uncertainties. The numerical simulation shows that the DFC strategy can reject disturbance effect and improve the pointing error for the three mentioned actuators; however, for a very large value of external disturbance, a critical value for uncertainty is observed for the thruster lag at which the pointing error is suddenly increased. For this critical value, the control system cannot tolerate any longer lag uncertainty in comparison with the two other actuation systems. Increasing the value of disturbance decreases the tolerable value of uncertainty in the thruster lag.

The need for prediction in feedback control of a mixing layer

We present the first application of a Smith predictor in the model-based feedback control of a convectively unstable flow. The two-dimensional incompressible mixing layer is chosen as a plant and integrated with a finite-element Navier–Stokes solver. The flow is actuated by a modulation of the inflow mimicking a pulsed jet actuator. The state is monitored by a downstream velocity sensor. The control goal is to command the fluctuation level at the sensor location. As an additional challenge, the incoming shear-flow profile is perturbed by the small noise level. The starting point of this study is a standard low-order control without prediction. A Smith predictor is included to compensate for the convective delay from actuation to sensing. The closed-loop results with a Smith predictor are distinctly better than without it. The design of the linear robust controllers is performed with an open-loop control. We conjecture that predictors may augment feedback control for a large class of convectively unstable flows with significant time-delay from actuation to sensing.

Model-Based Compensation of Dynamic Errors in Measuring Machines and Machine Tools

The demand for fast automated inspection of all manufactured components in industrial production is rising due to increasing quality demands and high-speed machining technologies. Modern surface metrology sensors are capable of implementing these tasks if they can be embedded in fast and accurate measuring machines. However, the common approach to achieve high positioning accuracy in such machines is abiding by very conservative limits in acceleration, and thus, velocity. We propose an active control system including a model-based compensation system for dynamic errors to increase acceleration and velocity. A novel optical sensor permitting the direct measurement of the actual tool-center-point (TCP) position is used for system identification of the required dynamic model. The position of a LED point source is detected with high subpixel precision by replicating the spot several times on the camera chip and averaging over the center of gravities of all spots. A model for the largest dynamic error component of a small multiaxis measuring machine is derived and the sensor is used to identify the model parameters. A model-based feedforward and feedback control system reducing the dynamic tracking error is designed. This control systems allows machine motions with both high accuracy and large velocity. The achieved performance is evaluated using TCP tracking error measurements obtained with the novel optical sensor. This model-based compensation system can also be applied to machine tools.