Neural Model Predictive Control Research Articles

Data-Driven Model Predictive Control for Redundant Manipulators With Unknown Model.

The tracking control of redundant manipulators plays a crucial role in robotics research and generally requires accurate knowledge of models of redundant manipulators. When the model information of a redundant manipulator is unknown, the trajectory-tracking control with model-based methods may fail to complete a given task. To this end, this article proposes a data-driven neural dynamics-based model predictive control (NDMPC) algorithm, which consists of a model predictive control (MPC) scheme, a neural dynamics (ND) solver, and a discrete-time Jacobian matrix (DTJM) updating law. With the help of the DTJM updating law, the future output of the model-unknown redundant manipulator is predicted, and the MPC scheme for trajectory tracking is constructed. The ND solver is designed to solve the MPC scheme to generate control input driving the redundant manipulator. The convergence of the proposed data-driven NDMPC algorithm is proven via theoretical analyses, and its feasibility and superiority are demonstrated via simulations and experiments on a redundant manipulator. Under the drive of the proposed algorithm, the redundant manipulator successfully carries out the trajectory-tracking task without the need for its kinematics model.

Kinematic analysis and advanced control of a vectored thruster based on 3RRUR parallel manipulator for micro-size AUVs

Abstract Autonomous underwater vehicles (AUVs) have played a pivotal role in advancing ocean exploration and exploitation. However, traditional AUVs face limitations when executing missions at minimal or near-zero forward velocities due to the ineffectiveness of their control surfaces, considerably constraining their potential applications. To address this challenge, this paper introduces an innovative vectored thruster system based on a 3RRUR parallel manipulator tailored for micro-sized AUVs. The incorporation of a vectored thruster enhances the performance of micro-sized AUVs when operating at minimal and low forward speeds. A comprehensive exploration of the kinematics of the thrust-vectoring mechanism has been undertaken through theoretical analysis and experimental validation. The findings from theoretical analysis and experimental confirmation unequivocally affirm the feasibility of the devised thrust-vectoring mechanism. The precise control of the vector device is studied using Physics-informed Neural Network and Model Predictive Control (PINN-MPC). Through the adoption of this pioneering thrust-vectoring mechanism rooted in the 3RRUR parallel manipulator, AUVs can efficiently and effectively generate the requisite motion for thrust-vectoring propulsion, overcoming the limitations of traditional AUVs and expanding their potential applications across various domains.

Flexible Building Energy Management With Neural ODEs-Based Model Predictive Control

Flexible Building Energy Management With Neural ODEs-Based Model Predictive Control

Enhancing grid stability through predictive control and fuzzy neural networks in flywheel energy storage systems integration

Renewable energy systems, exemplified by solar and wind power, are increasingly integrated into modern power grids to mitigate environmental impact and reduce reliance on fossil fuels. However, the inherent intermittency and unpredictability of renewable energy sources pose challenges to grid stability and reliability. Energy Storage Systems (ESS) offer a promising solution to address these challenges by smoothing out power fluctuations and ensuring a consistent power supply. Among various ESS technologies, Flywheel Energy Storage Systems (FESS) have emerged as a noteworthy contender due to their rapid response times, low operating costs, and extended lifespan. This paper focuses on investigating the operation of a novel unit comprising a solar power system integrated with a Flywheel Energy Storage System (PV-FESS). The aim is to develop an effective control algorithm utilizing adaptive fuzzy neural networks and model predictive control (ANFIS-MPC) to manage power fluctuations stemming from renewable energy sources within the grid. The proposed control strategy aims to optimize the operation of the PV-FESS system by dynamically adjusting the energy absorption or release of the flywheel to maintain grid stability. Simulation studies conducted using Matlab-Simulink/Simcape software validate the efficacy of the ANFIS-MPC algorithm in mitigating abnormal fluctuations from renewable energy sources. The results demonstrate that the PV-FESS system effectively balances power fluctuations, ensuring a stable and reliable power output to the grid.

ASV station keeping under wind disturbances using neural network simulation error minimization model predictive control

AbstractStation keeping is an essential maneuver for autonomous surface vehicles (ASVs), mainly when used in confined spaces, to carry out surveys that require the ASV to keep its position or in collaboration with other vehicles where the relative position has an impact over the mission. However, this maneuver can become challenging for classic feedback controllers due to the need for an accurate model of the ASV dynamics and the environmental disturbances. This work proposes a model predictive controller using neural network simulation error minimization (NNSEM–MPC) to accurately predict the dynamics of the ASV under wind disturbances. The performance of the proposed scheme under wind disturbances is tested and compared against other controllers in simulation, using the robotics operating system and the multipurpose simulation environment Gazebo. A set of six tests was conducted by combining two varying wind speeds that are modeled as the Harris spectrum and three wind directions (, , and ). The simulation results clearly show the advantage of the NNSEM–MPC over the following methods: backstepping controller, sliding mode controller, simplified dynamics MPC (SD‐MPC), neural ordinary differential equation MPC (NODE‐MPC), and knowledge‐based NODE MPC. The proposed NNSEM–MPC approach performs better than the rest in five out of the six test conditions, and it is the second best in the remaining test case, reducing the mean position and heading error by at least % ( m) and % (), respectively, across all the test cases. In terms of execution speed, the proposed NNSEM–MPC is at least 36% faster than the rest of the MPC controllers. The field experiments on two different ASV platforms showed that ASVs can effectively keep the station utilizing the proposed method, with a position error as low as m and a heading error as low as within time windows of at least s. This would increase the potential applications of ASVs for launch, recovery, and replenishment in long‐term surveys in collaboration with other autonomous systems.

Development of a compact rotary series elastic actuator with neural network-driven model predictive control implementation

Development of a compact rotary series elastic actuator with neural network-driven model predictive control implementation

Safe path planning and adjustable zonotope-tube model predictive tracking control for autonomous vehicle

The path planning and tracking control problem of autonomous vehicle with model matched uncertainty and external disturbance is studied in this paper. A safe lane change zone is designed by considering lane change time, distance, and driving speed. Meanwhile, the candidate paths are generated using cubic quasi-uniform B-spline curve, and the optimal path is selected based on feasibility, comfort, and efficiency criteria. The proposed planning method ensures an efficient and comfortable lane change path without collision with surrounding vehicles. Model uncertainty and external disturbance may affect lane change tracking control of autonomous vehicle. Hence, a fault-tolerant path tracking controller is implemented by using deep neural networks and adjustable zonotope-tube model predictive control. A deep neural network-based controller is proposed to compensate for model matched uncertainty, thereby enhancing the fault tolerance of the control system. Then, an adjustable zonotope-tube model predictive controller is designed to enclose the actual trajectory within a zonotope-tube with the nominal state as the center and the disturbance set as the radius by using the feedback control to achieve the asymptotic stabilization of the closed-loop system. The effectiveness of this path planning and tracking control method is finally verified by numerical simulation.

Neural Dynamics-Based Model Predictive Control for Mobile Redundant Manipulators With Improved Obstacle Avoidance

Neural Dynamics-Based Model Predictive Control for Mobile Redundant Manipulators With Improved Obstacle Avoidance

Data-Driven Optimal Control for Municipal Solid Waste Incineration Process

Municipal solid waste incineration (MSWI) is a dynamic industrial process involving complex physical and chemical reactions. Due to the uncertain municipal solid waste (MSW) composition and dynamic operation conditions, it is difficult to guarantee optimal operation for the MSWI process. To solve this problem, a data-driven optimal control scheme is proposed with a hierarchical control structure. In the set points optimization stage, an online adaptive fuzzy neural network (OAFNN) is designed to construct the objective functions, including combustion efficiency and NOx emission concentration. An adaptive mutation particle swarm optimization (AMPSO) algorithm is developed to obtain the optimal set points of oxygen content in flue gas. In the control stage, a double long short-term memory neural networks-based model predictive control (DLSTM-MPC) strategy is exploited. A self-organizing long short-term memory (SOLSTM) network is employed to predict the oxygen content in flue gas with a compact structure. To reduce the influences from dynamic disturbances and further improve the tracking control performance, another long short-term memory (LSTM) neural network is established to correct the prediction results. Finally, the set points optimization method and DLSTM-MPC strategy are combined to realize the optimal operation of the MSWI process. The effectiveness of the proposed optimal control scheme is verified by real industrial data. The experimental results demonstrate that the optimal control scheme can achieve promising tracking control performance of oxygen content in flue gas and improve the operational performance of the MSWI process.

Adaptive model predictive control based on neural networks for Hover attitude tracking with input saturation and unknown disturbances

The anti-disturbance control of quadrotor attitude tracking under saturation constraints is a difficult problem. In this paper, a neural network-based model predictive controller for quadrotor systems with input saturation and external disturbances is developed. The unmodeled dynamics and external disturbances of the system are simplified to the disturbance superimposed on the nominal system, and the gradient descent neural networks are used to complete the estimation and compensation of the disturbance. The adaptive model predictive controller is designed based on the nominal system. The disturbance value estimated by the neural network adaptively adjusts the control constraints in the model predictive controller. The robustness and anti-disturbance of the designed controller are analyzed. The experiments show that, compared to the robust model predictive control, the algorithm proposed in this paper reduces the steady-state mean errors of the yaw, pitch, and roll attitude channels of the Hover system. Specifically, the algorithm results in a decrease of 2.622%, 2.292%, and 1.192% without external disturbances and 2.056%, 4.17%, and 0.956% with outside disturbances. Experimental results confirm the effectiveness of the proposed method.

Data-driven torque and pitch control of wind turbines via reinforcement learning

This paper addresses the torque and pitch control problems of wind turbines. The main contribution of this work is the development of an innovative reinforcement learning (RL)-based control method targeting wind turbine applications. Our RL-based control framework synergistically combines the advantages of deep neural networks (DNNs) and model predictive control (MPC) technologies. The proposed control strategy is data-driven, adapting to real-time changes in system dynamics and enhancing control performance and robustness. Additionally, the incorporation of an MPC structure within our design improves learning efficiency and reduces the high computational complexity typically found in deep RL algorithms. Specifically, a DNN is designed to approximate the wind turbine dynamics based on a continuously updated dataset composed of state and action measurements taken at specified sampling intervals. The real-time control policy is generated by integrating the online trained DNN into an MPC architecture. The proposed method iteratively updates the DNN and control policy in real-time to optimize performance. As a primary result of this work, the proposed method demonstrates superior robustness and control performance compared to commonly-employed MPC and other baseline wind turbine controllers in the presence of uncertainties and unexpected actuator faults. This effectiveness is showcased through simulations with a high-fidelity wind turbine simulator.

Neural Koopman operator-assisted model predictive control of an Organic Rankine Cycle

This study deals with the Neural Koopman operator-assisted model predictive control of an Organic Rankine Cycle (ORC). The modeling of the evaporator and condenser is accomplished by utilizing the Finite Control Volume (FCV) method, and the modeling of the turbine and pump is fulfilled with pseudo-steady state equations. Afterward, the identification of the neural Koopman model is carried out with the input–output data obtained from the developed ORC model. The identification process resulted in 0.01121 and 0.01119 training and validation losses, respectively, which are sufficiently acceptable values. A linear MPC design is proposed to control the system in the Koopman-identified linear state space. Four different controllers are designed, each having different objectives, and their performances are compared with each other. The results reveal that the second law controller comes up with the most desirable exergy destruction rate of 2349 W, which is 1.44 % lower than the worst performer entropy generation controller. Moreover, the first and second law controllers result in the highest overall first and second law efficiencies, 0.14258 and 0.42945, respectively, through the simulation time. It is realized that the performance of the controllers is in line with their assigned objectives, and the proposed controller effectively regulates the system.

Nonlinear Control of Fouling in Polyethylene Reactors.

Fouling formation in reactor vessels poses a serious threat to the safe operation of the industrial low-density polyethylene (LDPE) polymerization. Fouling also degrades the polymer quality and causes productivity loss to some extent. In this work, neural Wiener model predictive control (NWMPC) is introduced to address the fouling concern. In addition, a soft sensor model is used to activate the fouling-defouling (F-D) mechanism when the fouling surpasses the thickness limit to prevent vessel overheating. NWMPC is proven to be fast, stable, and robust under various control scenarios. The use of a soft sensor model in conjunction with NWMPC enables the online monitoring and controlling of the F-D processes. When comparison is made with a state space (SSMPC) utilizing only the linear block, NWMPC is found to be able to control the LDPE grade with quicker grade transition and lower resource consumption.

Data-Driven Neural Predictors-Based Robust MPC for Power Converters

This article proposes a novel data-driven neural predictors-based robust finite control-set model predictive control (FCS-MPC) methodology for power converters, which aims to enhance the robustness of the control system and the flexibility of multiple control objectives. To be specific, the data-driven predictors-based neural network solution, which has a good potential to estimate the unknown nonlinear system dynamics by deploying real-time and historical data, is incorporated into the proposed design with a cost function derived intuitively from Lyapunov’s control theory. The key feature of this work is that the undesired effects of the uncertainties and the complex multiobjective optimization procedure can be explicitly dealt with, without involving accurate modeling information and weighting factor combinations, while guaranteeing adaptability to unknown nonlinear system dynamics, model variations, and environment changes. Finally, the stability analysis is given, and the effectiveness of the proposed FCS-MPC methodology is verified analytically and confirmed by comprehensive results that prove the theoretical investigations for active front-end modular multilevel converter.

Neural Network Based Model Predictive Control for a Quadrotor UAV

A dynamic model that considers both linear and complex nonlinear effects extensively benefits the model-based controller development. However, predicting a detailed aerodynamic model with good accuracy for unmanned aerial vehicles (UAVs) is challenging due to their irregular shape and low Reynolds number behavior. This work proposes an approach to model the full translational dynamics of a quadrotor UAV by a feedforward neural network, which is adopted as the prediction model in a model predictive controller (MPC) for precise position control. The raw flight data are collected by tracking various pre-designed trajectories with PX4 autopilot. The neural network model is trained to predict the linear accelerations from the flight log. The neural network-based model predictive controller is then implemented with the automatic control and dynamic optimization toolkit (ACADO) to achieve real-time online optimization. Software in the loop (SITL) simulation and indoor flight experiments are conducted to verify the controller performance. The results indicate that the proposed controller leads to a 40% reduction in the average trajectory tracking error compared to the traditional PID controller.

Higher Order Modeling of Reactor Regulating System and Nonlinear Neural Model Predictive Controller Design for a Nuclear Power Generating Station

In the existing instrumentation and control system of an operating Pressurized Heavy Water Reactor (PHWR) based nuclear power plant, conventional controllers are used to control the reactor power. A new idea of Nonlinear Neural Model Predictive Controller (NNMPC) is introduced in this research work. The new 17th order nonlinear higher order model of Reactor Regulating System (RRS) is developed under different plant operating modes and various parametric conditions in Single Input Multi Output (SIMO) configuration with special emphasis on Helium Control Valve Dynamics (HCVD) and Coupled Nonlinear Iodine and Xenon Dynamics (CNIXD). The SIMO RRS model is developed based on first principle. The 17th order model is reduced to 9th order lower dynamic model using Balanced Truncation Method (BTM). The Reduced Order SIMO RRS (RO-SIMO-RRS) model is programmed, simulated and validated in SIMULINK environment. The plant Neural SIMO RRS (N-SIMO-RRS) model is developed using innovative data generated from RO-SIMO-RRS simulations. The plant neural N-SIMO-RRS model is optimized using Levenberg-Marquardt Algorithm (LMA). Using the identified N-SIMO-RRS model, the Nonlinear Neural Model Predictive Controller (NNMPC) is designed, trained, verified, validated, and finally optimized using the backtracking technique in the SIMULINK environment. The optimized results are obtained from designed closed loop RRS and found within the acceptable design limits. The performance of the proposed closed loop RRS is also tested in reference tracking mode with excellent fast tractability near the optimal target demanded power level.

Motion control of linear induction motor using self-recurrent wavelet neural network trained by model predictive controller

Due to end effects phenomena that cause a decrease of air-gap flux and thrust force, obtaining a precise velocity for a linear induction motor (LIM) has become a significant challenge. This study suggests implementing a novel controller based on a self-recurrent wavelet neural network (SRWNN) and model predictive controller (MPC) to regulate the velocity and thrust force of LIM. The MPC was used to train the SRWNN in this study. The ultimate goal of employing such a control approach in neural network training is to reduce the degree of uncertainty caused by changes in motor parameters and load disturbance. The indirect field-oriented control (IFOC) approach was used to investigate velocity and flux control under varied loading circumstances. Furthermore, to supply the required LIM stator voltage, a SVPWM dependent voltage source inverter was used in this work. To ensure reliable performance, the suggested system combines the benefits of neural networks with the MPC method, resulting in a versatile controller with a basic construction that is easy to accomplish. The MATLAB package is utilized to simulates and outputs LIM responses. The results confirm that the proposed method, which efficiently controls the velocity and thrust force of the LIM, can cope with changes in load force disruption and motor parameters.

Physics-informed Neural Networks-based Model Predictive Control for Multi-link Manipulators

We discuss nonlinear model predictive control (MPC) for multi-body dynamics via physics-informed machine learning methods. In more detail, we use a physics-informed neural networks (PINNs)-based MPC to solve a tracking problem for a complex mechanical system, a multi-link manipulator. PINNs are a promising tool to approximate (partial) differential equations but are not suited for control tasks in their original form since they are not designed to handle variable control actions or variable initial values. We thus follow the strategy of Antonelo et al. (arXiv:2104.02556, 2021) by enhancing PINNs with adding control actions and initial conditions as additional network inputs. Subsequently, the high-dimensional input space is reduced via a sampling strategy and a zero-hold assumption. This strategy enables the controller design based on a PINN as an approximation of the underlying system dynamics. The additional benefit is that the sensitivities are easily computed via automatic differentiation, thus leading to efficient gradient-based algorithms for the underlying optimal control problem.

A neural network-based model predictive controller for displacement tracking of piezoelectric actuator with feedback delays

Piezoelectric actuators are widely used in micro/nanoscale robotic manipulators. Due to its hysteresis and dynamic-related nonlinearity, accurate displacement tracking control of piezoelectric actuator is challenging. Besides, in some low-cost practical systems with low sampling rate, transmission delay causes mismatches between feedback and real displacement, further increasing the challenge in tracking control. In this article, a neural network-based model predictive controller (MPC) is proposed for precise tracking control of piezoelectric actuator’s displacement in situation where feedback is slow and delayed. The prediction model is based on a nonlinear-autoregressive-moving-average-with-exogenous-inputs framework, which outputs entire prediction horizon of future displacement in a single time, and is fulfilled by a multilayer feedforward neural network. An extended Kalman filter-based estimation for displacement is introduced to relieve the influence of feedback delays so as to improve dynamic performance of the controller. Another neural network is trained to provide initial values for MPC to reduce computation costs and improve performance in dynamic tracking. In a series of tracking experiments, the effectiveness of proposed controller is verified.

Low‐density polyethylene tubular reactor control using neural Wiener model predictive control

AbstractLow‐density polyethylene (LDPE) is a valued commodity plastic with versatile applications. However, controlling the LDPE polymerization reactor is a challenging task due to its nonlinear behavior and multivariable process and operates in a board operating region. This work aims to develop and explore the neural Wiener MPC (NWMPC) application in controlling the LDPE tubular reactor process. The motivation for NWMPC development originates from its advantage in terms of lower development effort, time, resource, and computational costs compared to nonlinear MPC (NMPC) using the first principle model (FPM). In order to develop the LDPE tubular reactor model, the Aspen Plus and Aspen Dynamic software are used to simulate the process in steady‐state and dynamic form. The neural Wiener (NW) model identification is performed using state space and neural network model identification using Matlab software. The identification result shows the ability of the NW model to identify the nonlinear LDPE tubular reactor process successfully. Furthermore, the NWMPC has outperformed the state space MPC (SSMPC) in controlling grade transition, feed pressure loss, feed impurity disturbance, and heat of polymerization change during the online closed‐loop performance tests. These results signify the ability of NWMPC to handle practical LDPE tubular reactor control scenarios.