A neural network based adaptive robot controller

Morris, Alan S.; Khemaissia, Seddik

doi:10.1007/bf00435721

Cited by 11 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a consequence, the control system with CEP-based design may cause poor tracking performance in the face of large scale uncertainty. To design model-free control algorithm, fuzzy and neural networks based adaptive control design have been attracted by many researchers over the last decade, see for example, [5,6,13,14,16,18,19,23,32,33], to name a few. Many of these methods requires known bound of the fuzzy approximation errors as well as disturbance in order to achieve Lyapunov-stability condition.…”

Section: Introductionmentioning

confidence: 99%

Disturbance rejection in adaptive control for a class of nonlinear mechanical systems

Islam

Liu

2010

2010 IEEE/ASME International Conference on Advanced Intelligent Mechatronics

View full text Add to dashboard Cite

This paper presents hybrid control strategy for robust trajectory tracking control for a class of uncertain nonlinear mechanical systems. The design combines adaptive fuzzy system with robust adaptive control algorithm. Adaptive fuzzy system approximates unknown nonlinear system dynamics while a robustifying adaptive control term is used to cope with uncertainties due to the presence of external disturbance, modeling error and other unmodeled dynamical effects. Using the Lyapunov method, we first develop a stable hybrid controller by assuming that the system output and its derivatives are available for feedback control design. Then, an output feedback form of the position-velocity (state feedback) hybrid controller is proposed where the unknown velocity signal is replaced by the output of a model-free linear estimator. We prove that the tracking error bound under output feedback design can converge asymptotically to the tracking error bound achieved under the state feedback control design. Finally, the proposed method is implemented and evaluated on a 3-DOF Phantom TM medical mechatronics system to demonstrate the theoretical development.

show abstract

Section: Introductionmentioning

confidence: 99%

Disturbance rejection in adaptive control for a class of nonlinear mechanical systems

Islam

Liu

2010

2010 IEEE/ASME International Conference on Advanced Intelligent Mechatronics

View full text Add to dashboard Cite

show abstract

“…INTRODUCTION 20 ear controllers [4, 5, 6, and 49]. Such PD type industrial control system can be traced back to the literature [67,80]. These designs comprise PD control plus gravity compensation and the PD control plus desired gravity compensation term.…”

mentioning

confidence: 99%

Lyapunov-based hybrid control for robust trajectory tracking of robotic manipulators

Islam¹

View full text Add to dashboard Cite

Bibliothèque etArchives Canada Direction du Patrimoine de l'édition 2.2 The desired (dash), output trajectory (radians) (solid) and control inputs (newton-meters) for joints 1 and 2 without using input and output disturbance noise under Theorem 2.1 2.3 The tracking errors for joints 1 and 2 without using input and output disturbance noise under Theorem 2.1 2.4 The desired (dash), output trajectory (radians) (solid) and control inputs (newton-meters) for joints 1 and 2 using output feedback design of Theorem 2.1 with initial conditions 2.5 The tracking errors for joints 1 and 2 using with output feedback design of Theorem 2.1 using with initial conditions 2.6 The disturbance level of w(t) 2.7 The desired (dash), output trajectory (radians) (solid) and control inputs (newton-meters) for joints 1 and 2 using output feedback design of Theorem 2.1 using with input and output disturbance noise w(t) 572.8 The tracking errors of joints 1 and 2 under Theorem 2.1 with the disturbance noise w(t) 59 6 2.9 The desired (dash), output tracking (radians) (solid) and control inputs (newton-meters) for the joint 1 and joint 2 with Theorem 2.1 under dynamics changes 61 2.10 The tracking errors for joints 1 and 2 with Theorem 2.1 under model dynamic changes 62 2.11 The desired (dash), output tracking (radians) (solid) and control inputs (newton-meters) for the joint 1 and joint 2 under Theorem 2.2. 65 2.12 The tracking errors for joint 1 and joint 2 with Theorem 2.2 2.13 The desired (dash), output tracking (radians) (solid) and control inputs (newton-meters) for joints 1 and 2 under output feedback design of Theorem 2.3 without using the input and output disturbance noise 2.14 The tracking errors for joints 1 and 2 with the PD output feedback design of Theorem 2.3 without using the input and output disturbance noise 2.15 The desired (dash), output tracking (radians) (solid) and control inputs (newton-meters) for joints 1 and 2 under output feedback design of Theorem 2.3 using with the input and output disturbance noise w(t) 2.16 The The tracking errors for joint 1 and joint 2 under PD output feedback design with the input and output disturbance noise measurement w(t) 2.17 The tracking errors for joints 1 and 2 under PD and classical adaptive output feedback design under dynamics changes 3.1 The tracking errors (radians) for joint 1 and joint 2 under CAOFB and hybrid adaptive output feedback design with model dynamic changes. The solid-line is for CAOFB design and the red-dash-line is for Theorem 3.3 94 73.2 The tracking errors (radians), the control input (newton-meters) for joint 1 and joint under Theorem 3.1 and Theorem 3.3. The solidline is for Theorem 3.1 and the red-dash-line is for Theorem 3.3. . . 3.3The disturbance noise level d(t) -w(t) (newton-meters)3.4 The tracking errors (radians), the control input (newton-meters) for joint 1 and joint 2 under Theorem 3.1 and Theorem 3.3. The sloidline is for Theorem 3.1 and the red-dash-line is for Theorem 3.3. . . 3.5The tracking errors (radians), the control input (newton-meters) fo...

show abstract

Robotic Agent Control Based on Adaptive Intelligent Algorithm in Ubiquitous Networks

Lee

Hwang

Jang

et al. 2007

Agent and Multi-Agent Systems: Technologies and Applications

View full text Add to dashboard Cite

A neural network based adaptive robot controller

Cited by 11 publications

References 7 publications

Disturbance rejection in adaptive control for a class of nonlinear mechanical systems

Disturbance rejection in adaptive control for a class of nonlinear mechanical systems

Lyapunov-based hybrid control for robust trajectory tracking of robotic manipulators

Robotic Agent Control Based on Adaptive Intelligent Algorithm in Ubiquitous Networks

Contact Info

Product

Resources

About