In this study, we experimentally verified a robust control law based on variable structure control and the computed torque method on a laboratory motion system. Variable structure control is characterized by high-frequency switching control, which does not have to be applicable to real motion systems. Therefore, we used a suitable reaching law to create a continuous substitution for the discontinuous nonlinear control law. This continuous substitution caused the loss of control invariance to parametric and signal disturbances, but the control loop remained robust, with bounded position errors. We augmented the variable structure control with the computed torque method, which ensured the feedback linearization of the dynamic system. The proposed design of the control law also considered the velocity constraints of the system. Both components of the control law needed to be considered in the velocity limitation, otherwise, the system constraints would be violated. We designed a velocity limitation algorithm for variable structure control to be compatible with the computed torque method, with regard to the desired performance, with exponential convergence and no overshoot. The control law is shown to satisfy the control objectives, and the results are discussed.
This paper deals with the design of model predictive control (MPC) for a nonlinear model of a Van de Vusse reaction, that has been linearized at specific operating points. Since the conventional MPC formulation requires the full state vector, it is necessary to design a state observer that estimates the unmeasurable states of the system. Due to the fact that the accuracy of the state estimation significantly affects the resulting performance of MPC, the aim of this study is to compare the designed control while considering two different approaches to the state estimation. In particular, the Luenberger state observer and the Kalman filter are studied in this paper. The Luenberger observer represents the deterministic state estimator, that considers a system without significant noise affecting the states and the output, while its design is based on the pole placement method. In contrast, the statistically optimal state estimator is based on the Kalman filter design to estimate the state in the case of stochastic systems. The rest of the paper presents the results of simulations of MPC applied to the Van de Vusse reaction model using both considered state estimators.
This paper presents a robust model predictive control (RMPC) framework tailored for the Van de Vusse reaction (VVR) in a continuous stirred-tank reactor (CSTR), a chemical process characterized by strong nonlinear dynamics, non-minimum phase behavior, and parametric uncertainty. The proposed control strategy employs a polytopic uncertain prediction model combined with linear matrix inequality (LMI)-based synthesis to design an offline static linear state-feedback controller that guarantees robust quadratic stability and closed-loop performance. To enable output reference tracking, an extended Kalman filter (EKF) is integrated for full-state estimation, eliminating the need for output-feedback control and avoiding the complexity of bilinear matrix inequalities (BMIs). The method also includes static optimization of the reactor's operating conditions to improve overall efficiency. Numerical simulations confirm that the proposed approach achieves accurate tracking and robustness with substantially reduced online computational cost. The average solve time is significantly lower than that of nonlinear MPC and constrained linear MPC, demonstrating the method’s suitability for real-time control of uncertain chemical processes with limited computational resources. The presented method offers a practical balance between robustness, performance, and tractability for controlling uncertain nonlinear systems.
The ability to autonomously generate up-to-date 3D models of robotic workcells is critical for advancing smart manufacturing, yet existing Next-Best-View (NBV) methods often rely on paradigms ill-suited for the fixed-base manipulators found in dynamic industrial environments. To address this gap, this paper proposes a novel NBV method for the complete exploration of a 6-DOF robotic arm’s workspace. Our approach integrates collision-based information gain metric, a potential field technique to generate candidate views from exploration frontiers, and a tunable fitness function to balance information gain with motion cost. The method was rigorously tested in three simulated scenarios and validated on a physical industrial robot. Results demonstrate that our approach successfully maps the majority of the workspace in all setups, with a balanced weighting strategy proving most effective for combining exploration speed and path efficiency, a finding confirmed in the real-world experiment. We conclude that our method provides a practical and robust solution for autonomous workspace mapping, offering a flexible, training-free approach that advances the state-of-the-art for on-demand 3D model generation in industrial robotics.