Search

Your search keyword '"Sadati, Nasser"' showing total 176 results
Start Over Author "Sadati, Nasser"
176 results on '"Sadati, Nasser"'

1. Fast and Optimal Adaptive Tracking Control: A Novel Meta-Reinforcement Learning via Conditional Generative Adversarial Net

Author

Mahmoudi, Mohammad and Sadati, Nasser

Subjects
Electrical Engineering and Systems Science - Systems and Control
Abstract

The control of nonlinear systems with unknown dynamics has been a significant field of research for many years. This paper presents a novel data-driven optimal adaptive control structure with less control effort and faster adaptation than standard adaptive control counterparts. The proposed control structure utilizes the system's recorded data to increase the speed of adaptation and performance dramatically. In this study, we employ a conditional generative adversarial net (CGAN) as a novel central pattern generator to reproduce the steady-state harmonic pattern of the control signals matching the system's uncertainties over a wide range. We can also use the CGAN architecture as a fault detector. The CGAN provides a low-dimensional latent space of uncertainties. It enables rapid and convenient adaptation when there are many parametric uncertainties, especially for large-scale systems. Then, we introduce a novel meta-reinforcement learning framework to adapt the latent space of CGAN to the system's uncertainties as an optimal direct adaptive controller without any system identifier. Another part of the control structure is a regulator that achieves semi-global asymptotic tracking using the Lyapunov stability analysis. Finally, via some simulations, we evaluate the capabilities of the proposed designs on two dynamical systems, a robot manipulator and a large-scale musculoskeletal structure, in the presence of disturbance and perturbation.

Published
2022

2. Data Driven Safe Gain-Scheduling Control

Author

Modares, Amir, Sadati, Nasser, and Modares, Hamidreza

Subjects
Electrical Engineering and Systems Science - Systems and Control
Abstract

Data-based safe gain-scheduling controllers are presented for discrete-time linear parameter-varying systems (LPV) with polytopic models. First, $\lambda$-contractivity conditions are provided under which safety and stability of the LPV systems are unified through Minkowski functions of the safe sets. Then, to bypass the requirement to identify the system dynamics, a data-based representation of the closed-loop LPV system is provided to directly exploit collected data and construct a safe controller. It is shown that weaker data richness requirements are needed to directly learn a closed-loop safe control policy than to identify the LPV system. The closed-loop data-based representation is leveraged to directly design data-driven gain-scheduling controllers that guarantee $\lambda$-contractiveness, and, thus, invariance of the safe sets. It is also shown that the problem of designing a data-driven gain-scheduling controller for a polyhedral (ellipsoidal) safe set amounts to a linear program (a semi-definite program). A simulation example is provided to show the effectiveness of the presented approach.

Published
2022

3. CASU2Net: Cascaded Unification Network by a Two-step Early Fusion for Fault Detection in Offshore Wind Turbines

Author

Salari, Soorena and Sadati, Nasser

Subjects
Electrical Engineering and Systems Science - Systems and Control, Computer Science - Machine Learning, Electrical Engineering and Systems Science - Signal Processing
Abstract

This paper presents a novel feature fusion-based deep learning model (called CASU2Net) for fault detection in offshore wind turbines. The proposed CASU2Net model benefits of a two-step early fusion to enrich features in the final stage. Moreover, since previous studies did not consider uncertainty while model developing and also predictions, we take advantage of Monte Carlo dropout (MC dropout) to enhance the certainty of the results. To design fault detection model, we use five sensors and a sliding window to exploit the inherent temporal information contained in the raw time-series data obtained from sensors. The proposed model uses the nonlinear relationships among multiple sensor variables and the temporal dependency of each sensor on others which considerably increases the performance of fault detection model. A 10-fold cross-validation approach is used to verify the generalization of the model and evaluate the classification metrics. To evaluate the performance of the model, simulated data from a benchmark floating offshore wind turbine (FOWT) with supervisory control and data acquisition (SCADA) are used. The results illustrate that the proposed model would accurately disclose and classify more than 99% of the faults. Moreover, it is generalizable and can be used to detect faults for different types of systems.

Published
2020

8. Safe Reinforcement Learning via a Model-Free Safety Certifier

Author

Modares, Amir, Sadati, Nasser, Esmaeili, Babak, Yaghmaie, Farnaz Adib, and Modares, Hamidreza

Abstract

This article presents a data-driven safe reinforcement learning (RL) algorithm for discrete-time nonlinear systems. A data-driven safety certifier is designed to intervene with the actions of the RL agent to ensure both safety and stability of its actions. This is in sharp contrast to existing model-based safety certifiers that can result in convergence to an undesired equilibrium point or conservative interventions that jeopardize the performance of the RL agent. To this end, the proposed method directly learns a robust safety certifier while completely bypassing the identification of the system model. The nonlinear system is modeled using linear parameter varying (LPV) systems with polytopic disturbances. To prevent the requirement for learning an explicit model of the LPV system, data-based $\lambda $ -contractivity conditions are first provided for the closed-loop system to enforce robust invariance of a prespecified polyhedral safe set and the system’s asymptotic stability. These conditions are then leveraged to directly learn a robust data-based gain-scheduling controller by solving a convex program. A significant advantage of the proposed direct safe learning over model-based certifiers is that it completely resolves conflicts between safety and stability requirements while assuring convergence to the desired equilibrium point. Data-based safety certification conditions are then provided using Minkowski functions. They are then used to seemingly integrate the learned backup safe gain-scheduling controller with the RL controller. Finally, we provide a simulation example to verify the effectiveness of the proposed approach.

Published
2024
Full Text
View/download PDF

20. Modified maximum entropy fuzzy data association filter

Author

Tafti, Abdolreza Dehghani and Sadati, Nasser

Subjects
Fuzzy algorithms -- Research, Fuzzy logic -- Research, Fuzzy systems -- Research, Entropy (Information theory) -- Research, Electronic data processing -- Methods, Electronic data processing -- Technology application, Fuzzy logic, Technology application, Engineering and manufacturing industries, Science and technology
Abstract

The problem of fuzzy data association for target tracking in a cluttered environment is discussed in this paper. In data association filters based on fuzzy clustering, the association probabilities of tracking filters are reconstructed by utilizing the fuzzy membership degree of the measurement belonging to the target. Clearly in these filters, the fuzzy clustering method has an important role; better approach causes better precision in target tracking. Recently, by using the information theory, the maximum entropy fuzzy data association filter (MEF-DAF), as a fast and efficient algorithm, is introduced in literature. In this paper, by modification of a fuzzy clustering objective function, which is prepared for using in target tracking, a modified maximum entropy fuzzy data association filter (MMEF-DAF) is proposed. The MMEF-DAF has a better performance in case of single and multiple target tracking than MEF-DAF, and the other known algorithms such as probabilistic data association filter and the hybrid fuzzy data association filter. Using Monte Carlo simulations, the superiority of the proposed algorithm in comparison with the previous ones is demonstrated. Simply, less computational cost and suitability for real-time applications are the main advantages of the proposed algorithm. [DOI: 10.1115/1.4000817] Keywords: data/measurement association, target tracking, fuzzy clustering, information theory

Published
2010

21. Adaptive neuro-fuzzy inference system in fuzzy measurement to track association

Author

Tafti, Abdolreza Dehghani and Sadati, Nasser

Subjects
Neural networks -- Research, Fuzzy algorithms -- Research, Fuzzy logic -- Research, Fuzzy systems -- Research, Mensuration -- Methods, Adaptive control -- Research, Electronic data processing -- Methods, Electronic data processing -- Equipment and supplies, Electronic data processing -- Technology application, Neural network, Fuzzy logic, Technology application, Engineering and manufacturing industries, Science and technology
Abstract

The main issue in a surveillance environment is the target tracking. The most important concern in this problem is the association of the various measurements with the existing target tracks. The fuzzy c-means data association (FCMDA) algorithm, based on the fuzzy c-means (FCM) algorithm, is an efficient solution for the problem of measurement to track association in a multisensor multitarget environment. It has a high accuracy in measurement to track association when targets are far from each other. However, its accuracy remains low when targets are close to one another. The FCMDA algorithm performance is usually lost in this environment, especially when measurement noise is high. In the FCMDA algorithm, the association between measurements and tracks is determined using an optimal membership function derived from the FCM algorithm for the fixed predicted state of targets. The prediction of the target state deviates from its correct value based on updating the tracker/filter with the wrong associated measurement. Consequently, the wrong association can take place using a deviated prediction of target state in the FCMDA algorithm. In this paper, to overcome this shortcoming of the FCMDA algorithm, the predicted state of every target in a surveillance environment is compensated for the effect of wrong associated measurement by an adaptive neurofuzzy inference system (ANFIS). An ANFIS has both the advantages of expert knowledge of a fuzzy inference system and the learning capability of neural networks. So a trained ANFIS is able to compensate the effect of a wrong associated measurement on the prediction of target state. Using the compensated prediction of target state in the FCMDA algorithm can always save the performance of the FCMDA algorithm and extend its domain of usage in practical applications. The simulation results demonstrate that considerable improvements in terms of accuracy and performance are achieved by using the compensated prediction of target state in the FCMDA algorithm. [DOI: 10.1115/1.4000663] Keywords: measurement to track association, fuzzy c-means data association (FCMDA) algorithm, adaptive neurofuzzy inference system (ANFIS)

Published
2010

27. Low-order H[infinity] controller design for an active suspension system via LMIs

Author

Amirifar, Ramin and Sadati, Nasser

Subjects
Robust statistics -- Analysis, Electric controllers -- Design and construction, Electronic control -- Design and construction, Linear models (Statistics) -- Analysis, Business, Computers, Electronics, Electronics and electrical industries
Abstract

An application of a new controller order reduction technique with stability and performance preservation based on linear matrix inequality optimization to an active suspension system is presented. In this technique, the rank of the residue matrix of a proper rational approximation of a high-order H[infinity] controller subject to the H[infinity]-norm of a frequency-weighted error between the approximated controller and the high-order H[infinity] controller is minimized. However, because solving this matrix rank minimization problem is very difficult, the rank objective function is replaced with a nuclear-norm that can be reduced to a semidefinite program so that it can be solved efficiently. Application to the active suspension system of the Automatic Laboratory of Grenoble provides a fourth-order controller. The experimental results show that the control specifications are met to a large extent. Index Terms--Active suspension system, linear matrix inequalities (LMIs), order reduction, robust control.

Published
2006

33. Hierarchical optimal control of a binary distillation column

Author

Ramezani, Mohammad Hossein and Sadati, Nasser

Subjects
large-scale systems, optimal control, Distillation Column, Interaction-Prediction Approach, hierarchical control
Abstract

This paper presents a two-level optimal control for the binary distillationcolumn. From the control point of view, the binary distillation column is ahigh-order system consisting of several interconnected subsystems (trays) thataim to maximize output purity. Controlling the system in a hierarchical mannernot only decreases complexity and solution time but also can improve thecontrol structure's reliability. In this study, the distillation column is decomposedinto NT second-order subsystems, each representing one stage (one tray).Accordingly, the cost function is decomposed, and thus, the overall problem isconverted into NT lower-order subproblems where each tray has its own controller.A two-level interaction prediction approach provides optimal control forthe overall system. The task of the first level is to solve the subproblems using thepredicted values of the coordination parameters, whereas the second level actsas a coordinator to update the coordination parameters. Simulation results showthe capability and efficacy of the proposed two-level control method in findingthe optimal solution with less complexity and lower solution time than those ofthe centralized method.

Published
2019

Catalog

Books, media, physical & digital resources
See catalog results