Effective recursive parallelotopic bounding for robust output-feedback control

For quasi-linear parameter varying (quasi-LPV) systems with bounded disturbance, a synthesis approach of dynamic output feedback robust model predictive control (OFRMPC) with the consideration of input saturation is investigated. The saturated dynamic output feedback controller is represented by a convex hull involving the actual dynamic output controller and an introduced auxiliary controller. By taking both the actual output feedback controller and the auxiliary controller with a parameter-dependent form, the main optimization problem can be formulated as convex optimization. The consideration of input saturation in the main optimization problem reduces the conservatism of dynamic output feedback controller design. The estimation error set and bounded disturbance are represented by zonotopes and refreshed by zonotopic set-membership estimation. Compared with the previous results, the proposed algorithm can not only guarantee the recursive feasibility of the optimization problem, but also improve the control performance at the cost of higher computational burden. A nonlinear continuous stirred tank reactor (CSTR) example is given to illustrate the effectiveness of the approach.

International audience

ABSTRACTThe output feedback robust model predictive control (MPC), for the linear parameter varying (LPV) system with norm-bounded disturbance, is addressed, where the model parametric matrices are only known to be bounded within a polytope. The previous techniques of norm-bounding technique, quadratic boundedness (QB), dynamic output feedback, and ellipsoid (true-state bound; TSB) refreshment formula for guaranteeing recursive feasibility, are fused into the newly proposed approaches. In the notion of QB, the full Lyapunov matrix is applied for the first time in this context. The single-step dynamic output feedback robust MPC, where the infinite-horizon control moves are parameterised as a dynamic output feedback law, is the main topic of this paper, while the multi-step method is also suggested. In order to strictly guarantee the physical constraints, the outer bound of the true state replaces the true state itself, so tightness of this bound has a major effect on the control performance. In order to tighten the TSB, a procedure for refreshing the real-time ellipsoid based on that of the last sampling instant is given. This paper is conclusive for the past results and far-reaching for the future researches. Two benchmark examples are given to show the effectiveness of the novel results.

For the input and output constraints and uncertainties in batch processes, a 2D output feedback robust constrained model predictive control (MPC) method is designed by combining iterative learning control (ILC), MPC and output feedback. Firstly, an equivalent 2D-FM closed-loop prediction model is established by combining with the proposed output feedback controller. Then an optimization performance index function with terminal constraint is constructed to study its control optimization. According to the designed optimization performance index and Lyapunov stability theory, the feasible MPC problem is obtained by solving the linear matrix inequalities (LMIs). At the same time, the gain of the new output feedback control law is given to ensure that the performance index reaches the minimum upper bound under the constraints of input and output. In order to solve the manual adjustment problem of some parameters in the performance index function, the particle swarm optimization (PSO) algorithm is introduced, and a better solution is found near the controller by using the search optimization method. Finally, taking the injection molding process as an example and comparing with the existing method without using PSO algorithms, it is proved that the above method is more feasible.

Off-line approach to dynamic output feedback robust model predictive control

This paper combines model predictive control (MPC) and set-membership (SM) state estimation techniques for controlling systems subject to hard input and state constraints. Linear systems with unknown but bounded disturbances and partial state information are considered. The adopted approach guarantees that the constraints are satisfied for all the states which are compatible with the available information and for all the disturbances within given bounds. Properties of the proposed MPC-SM algorithm and simulation studies are reported.

Reconfiguration of flow-based networks with back-up components using robust economic MPC

This study proposes a layered hybrid control framework for automation and robotics systems that combines Tube-based Robust MPC (RMPC) as a safety envelope, adaptive parameter estimation to narrow online uncertainty, Sliding Mode Control (SMC) as a high-resilience fail-safe, and Reinforcement Learning (RL) as a performance booster through a warm-start/residual policy scheme that is always constrained by Control Barrier Functions (CBF). The study aims to achieve better tracking performance and computational efficiency without sacrificing safety and constraint compliance. The methodology includes constrained nonlinear plant modeling, design of tube invariance-based RMPC with constraint tightening, RLS/concurrent learning-based adaptive estimator, SMC for infeasibility/model drift conditions, and a CBF-based safety filter formulated as an online small QP. Evaluation is done through ablation studies on multi-domain benchmarks and HIL-style tests, with finite-horizon cost, ITAE, constraint violation rate, and computational latency (average and WCET) metrics. The results show that the RL+CBF configuration reduces the finite-horizon cost by about 15.7% compared to pure RMPC, maintaining a violation rate of ~0.35% (better than RMPC: 0.80%) and significantly below that of unfiltered RL (4.20%). The warm-start RL scheme accelerates the completion time of the MPC solver by ≈15–18% over various prediction horizons and reduces the WCET (26.4 → 23.3 ms), supporting real-time implementation. These findings confirm that the integration of RMPC–Adaptive–SMC–RL(+CBF) effectively bridges model-based optimality, online adaptivity, robust resilience, and learning from data in a single architecture that is safety-certified and feasible for real-world applications demanding high uptime.

This paper extends efficient, robust model predictive control (MPC) approaches for linear parameter varying (LPV) systems to tracking scenarios. A dual mode approach is used and future information about target changes is included in the optimisation tracking problem. The controller guarantees recursive feasibility by adding an artificial target as an extra degree of freedom. Convergence to admissible targets is ensured by constructing a robustly invariant set to track any admissible target. The efficacy of the proposed algorithm is demonstrated by MATLAB simulations.

ABSTRACTIn this work, synthesis of robust distributed model predictive control (MPC) is presented for a class of linear systems subject to structured time-varying uncertainties. By decomposing a global system into smaller dimensional subsystems, a set of distributed MPC controllers, instead of a centralised controller, are designed. To ensure the robust stability of the closed-loop system with respect to model uncertainties, distributed state feedback laws are obtained by solving a min–max optimisation problem. The design of robust distributed MPC is then transformed into solving a minimisation optimisation problem with linear matrix inequality constraints. An iterative online algorithm with adjustable maximum iteration is proposed to coordinate the distributed controllers to achieve a global performance. The simulation results show the effectiveness of the proposed robust distributed MPC algorithm.

A novel robust model predictive control strategy is presented for a class of discrete-time time-varying system with delays. The modified quadratic Lyapunov-Krasovskii functionals for uncertain discrete-time system with both states and input delays is proposed. A robust model predictive control law is derived such that the closed-loop system is asymptotically stable and a dasiaworst-casepsila infinite horizon performance objective is minimized. A feasible state feedback model predictive control law is obtained by using linear matrix inequalities approach. And then, a convex optimization problem is formulated to design the robust model predictive controller which minimizes the upper bound of the closed-loop performance objective. The robust model predictive controller design is illustrated with numerical examples.

A robust dichotomy solution-based model predictive control (DS-MPC) with a voltage observer for a three-phase grid-connected inverter (GCI) is proposed in this paper. The optimal voltage vector can be dynamically selected in the whole space vector circle plane in a very short time due to the voltage vector dichotomy solution algorithm. Therefore, the proposed DS-MPC control has not only fast dynamic performance but also good steady state performance. Meanwhile, the performance of model predictive control (MPC) usually depends on the accuracy of plant model. In weak grid case, the variation of equivalent grid impedance could impact on the performance. To improve the current control robustness, DS-MPC is combined with a grid voltage disturbance observer using the extended Kalman filter (EKF) so that the disturbance of weak grid voltages is effectively suppressed. Moreover, the grid currents are well regulated even with greatly distorted and unbalanced grid voltages. The test results verify the correctness and effectiveness of the proposed DS-MPC with EKF observer under the weak grid.

In this paper, robust constrained model predictive control of uncertain fuzzy affine systems is considered. Based on piecewise quadratic Lyapunov functions, highly efficient robust constrained predictive control approaches are developed so that the closed-loop stability is guaranteed, and the transient control performance is improved even under input or state constraints. Moreover, by using approximate ellipsoid and S -procedure, the solution to the fuzzy affine model based predictive control can be cast as a convex optimization problem subject to some linear matrix inequalities, rather than a nonconvex problem via bilinear matrix inequalities as in conventional fuzzy affine model-based control. The proposed controllers are thus easier for real-time implementation in industry. Simulation results on the fluidized bed combustion plant have demonstrated the performance of the proposed approaches.

This work aims to develop a robust model predictive control (MPC) based on the active disturbance rejection control (ADRC) approach by using a discrete extended disturbance observer (ESO). The proposed technique uses the ADRC approach to lump disturbances and uncertainties into a total disturbance, which is estimated with a discrete ESO and rejected through feedback control. Thus, the effects of the disturbances are attenuated, and a model predictive control is designed based on a canonical model free of uncertainties and disturbances. The proposed control technique is tested through simulation into a robotic autonomous underwater vehicle (AUV). The AUV’s dynamic model is used to compare the performance of a classical MPC and the combined MPC-ADRC. The evaluation results show evidence of the superiority of the MPC-ADRC over the classical MPC under tests of reference tracking, external disturbances rejection, and model uncertainties attenuation.

This article provides a solution to tube-based output feedback robust model predictive control (RMPC) for discrete-time linear parameter varying (LPV) systems with bounded disturbances and noises. The proposed approach synthesizes an offline optimization problem to design a look-up table and an online tube-based output feedback RMPC with tightened constraints and scaled terminal constraint sets. In the offline optimization problem, a sequence of nested robust positively invariant (RPI) sets and robust control invariant (RCI) sets, respectively, for estimation errors and control errors is optimized and stored in the look-up table. In the online optimization problem, real-time control parameters are searched based on the bounds of time-varying estimation error sets. Considering the characteristics of the uncertain scheduling parameter in LPV systems, the online tube-based output feedback RMPC scheme adopts one-step nominal system prediction with scaled terminal constraint sets. The formulated simple and efficient online optimization problem with fewer decision variables and constraints has a lower online computational burden. Recursive feasibility of the optimization problem and robust stability of the controlled LPV system are guaranteed by ensuring that the nominal system converges to the terminal constraint set, and uncertain state trajectories are constrained within robust tubes with the center of the nominal system. A numerical example is given to verify the approach.