Métodos de filtragem, estimação e controle adaptativo indireto aplicados a sistemas de teleoperação bilateral

Abstract

In this thesis, filtering, estimation and indirect adaptive control algorithms applied to bilateral teleoperation systems are proposed. The objective of this work was to identify and to implement indirect adaptive control in a slave master system. In a growing scenario of remote access to environments and equipment, as in industry 4.0, there is a need to quantify and characterize dynamic system parameters in order to improve the performance of control systems.In this manuscript three contributions regarding the teleoperated systems estimation and control are proposed. First, the Extended Kalman Filter (EKF) algorithm was combined with the Recursive Least Squares State-Variable Filter (RLSSVF), giving rise to the Hybrid Algorithm (AH1). The AH1 was able to estimate the parameters of a continuous time variant system in the state space form. In the second contribution, the Kalman Filter (KF) algorithm was combined with the Recursive Least Squares Estimator with Forgetting Factor Ú (RLSEFFÚ), originating the Hybrid Algorithm (AH2). AH2 was able to estimate specific parameters of a continuous time variant state space system. In the last contribution, the Method for Parameter and Delay Time Estimation (MEPTA) was combined with the Pole Allocation Method, giving rise to the Hybrid Algorithm for Identification and Adaptive Control (AHICA). Thus, AHICA was able to perform indirect adaptive control of a process with discrete time variant parameters. Such control system meets transient and stationary constraints, as well as the setpoint tracking.The proposed methods presented a good performance, especially regarding the algorithms coupling, what can be verified through the simulated results. Such methods can be applied to similar problems from new mathematical arrangements and computational constructions.

Engenharia de controle e automação: estudos fundamentais

“…Assim, torna-se importante a compreensão de métodos de identificação de sistemas, relacionados à área da modelagem matemática [5]. A utilização de técnicas de identificação de sistemas dinâmicos permite modelar a dinâmica do processo com pouco ou nenhum conhecimento prévio sobre a planta [6].…”

“…Por outro lado, um método de identificação é dito online, se esse executa simultaneamente a operação do processo a ser identificado, ou seja, é atualizado recursivamente a cada nova amostra [8]. Neste sentido, métodos online possuem característica essencial de identificar mudanças paramétricas em tempo real ao longo da operação e controle do processo [9]. consiste de um sistema seguidor de modelo, dessa forma se estabeleceu um modelo que descreve o comportamento dinâmico esperado do sistema de malha fechada.…”

“…Para que o algoritmo atualize as variações paramétricas com maior capacidade, o algoritmo do MQR digital pode ser reescrito ao aplicar um fator de esquecimento, o chamado algoritmo dos Mínimos Quadrados Recurso com Fator de Esquecimento (MQR-FE) [6].…”

“…Figures 8,9,10,and 11 present the robust residuals when this kind of fault occurs. As a result of the isolation scheme applied, the residuals generated by the UIO correspondent to the rotational velocity sensor, are insensitive to that fault, see figure11.…”

Modeling, simulation, motion trajectory planning and nonlinear control in the joint space of the manipulator robot SCARA T3 401SS manufacturer Epson

“…In the article [4], we presented the modeling of a PUMA robot with six degrees of freedom and the construction of an application for trajectory planning. In another article [5], the modeling, simulation, and trajectory control of a SCARA robot were shown. There are other articles, in which the authors have devoted themselves to laying the mathematical foundations of manipulative robots [6]…”