This paper presents a continuous-time decentralized neural control scheme for trajectory tracking of a two degrees of freedom direct drive vertical robotic arm. A decentralized recurrent high-order neural network (RHONN) structure is proposed to identify online, in a series-parallel configuration and using the filtered error learning law, the dynamics of the plant. Based on the RHONN subsystems, a local neural controller is derived via backstepping approach. The effectiveness of the decentralized neural controller is validated on a robotic arm platform, of our own design and unknown parameters, which uses industrial servomotors to drive the joints.

The purpose of this paper is to introduce a novel adaptive neural network-based control scheme for the Furuta pendulum, which is a two degree-of-freedom underactuated system. Adaptation laws for the input and output weights are also provided. The proposed controller is able to guarantee tracking of a reference signal for the arm while the pendulum remains in the upright position. The key aspect of the derivation of the controller is the definition of an output function that depends on the position and velocity errors. The internal and external dynamics are rigorously analyzed, thereby proving the uniform ultimate boundedness of the error trajectories. By using real-time experiments, the new scheme is compared with other control methodologies, therein demonstrating the improved performance of the proposed adaptive algorithm.

In this paper, a class of robot manipulators actuated by direct current (DC) motors and servo amplifiers in current mode is studied. For those robots, it is assumed that the control inputs to the manipulator are the servo amplifier input voltages. A model-based trajectory tracking controller is developed, which renders the closed-loop system to be passive from an auxiliary control input to a filtered error signal. Experimental results support theoretical results.

This paper presents a solution of the trajectory tracking problem for robotic manipulators using a recurrent high order neural network (RHONN) structure to identify the robot arm dynamics, and based on this model a discrete-time control law is derived, which combines block control and the sliding mode techniques. The block control approach is used to design a nonlinear sliding surface such that the resulting sliding mode dynamics is described by a desired linear system. The neural network learning is performed on-line by Kalman filtering. The local controller for each joint uses only local angular position and velocity measurements. The applicability of the proposed control scheme is illustrated via simulations.

This paper presents a discrete-time decentralized control scheme for identification and trajectory tracking of a two degrees of freedom (DOF) robot manipulator. A recurrent high order neural network (RHONN) structure is used to identify the plant model and based on this model, a discrete-time control law is derived, which combines discrete-time block control and sliding modes techniques. The neural network learning is performed online by Kalman filtering. A controller is designed for each joint, using only local angular position and velocity measurements. These simple local joint controllers allow trajectory tracking with reduced computations. The proposed scheme is implemented in real-time to control a two DOF robot manipulator.

In this paper we present a fuzzy self-tuning algorithm to select the Proportional, Integral and Derivative gains of a PID controller according to the actual state of a robotic manipulator. The stability via Lyapunov theory for the closed loop control system is also analyzed and shown that is locally asymptotically stable for a class of gain matrices depending on the manipulator states. This feature increases the potential of the PID control scheme to handle practical constraints in actual robots such as presence of actuators with limited torque capabilities. Experimental results on a two degrees of freedom robot arm shown the usefulness of the proposed approach.

This paper presents the development of a control algorithm, which combines the linear regulator theory and Takagi–Sugeno (T–S) fuzzy modeling. Each local controller consists in a feedback control term plus a trajectory tracking term. The global controller is determined by means of T–S methodology. Then, the proposed algorithm is applied in real time on a robot manipulator. © 2009 Wiley Periodicals, Inc.

This paper proposes a natural saturating extension of the proportional-derivative (PD) with desired gravity compensation (PDdgc) control law for the global regulation of robot manipulators with bounded inputs. Compared with other algorithms previously proposed under the same analytical framework, it proves to be advantageous in several senses. First, it involves a single saturation function at each joint. Second, it does not need to discriminate the terms that shall be bounded, since these are simply all of them included within the only saturation involved at every joint. Experimental results show the effectiveness of the proposed scheme. As far as the authors are aware, no such type of natural saturating extension (i.e., involving only one saturation function at each joint, where all the terms of the controller are embedded) to the bounded input case had been previously proposed for the PDdgc scheme. Furthermore, it is shown how the proposed approach may be conceived within the framework of the energy shaping plus damping injection methodology.

This paper addresses the set-point control of robot manipulators with friction where avoiding saturation of the actuators is a major issue. The original contribution is a novel direct fuzzy control system dealing with both practical constraints in mechanical manipulators: saturation and friction. The control system is made by taking advantage of input-output properties of the so-called sectorial fuzzy controllers. When friction is considered, we prove, via Lyapunov theory, that the steady state position errors owing to static friction are inside of a global attractor, which can be arbitrarily reduced. In case of absence of friction, the closed-loop system becomes globally asymptotic stable. In both cases, the important theoretical and practical feature of maintaining the control actions always within prescribed limits according to the actuator torque capabilities is guaranteed. Experimental evaluation of the proposed direct fuzzy control system on a nonlinear direct-drive robot arm is presented to validate its effectiveness.

This paper shows that fuzzy control systems satisfying sectorial properties are effective for motion tracking control of robot manipulators. We propose a controller whose structure is composed by a sectorial fuzzy controller plus a full nonlinear robot dynamics compensation, in such a way that this structure leads to a very simple closed-loop system represented by an autonomous nonlinear differential equation. We demonstrate via Lyapunov theory, that the closed-loop system is globally asymptotically stable. Experimental results show the feasibility of the proposed controller.

This paper deals with applications of fuzzy logic systems to motion control of robot manipulators. In the proposed application, the fuzzy logic system plays the role of a tuner of the robot controller gains. This idea is evoked for the gains of the so-called PD+ control scheme and we demonstrate, by taking into account the full nonlinear and multivariable nature of the robot dynamics, that the overall closed loop system is globally asymptotically stable. Experimental results on a two degrees of freedom direct-drive arm show the usefulness of the proposed control approach.

Computed-torque control is a well-known motion control strategy for manipulators which ensures global asymptotic stability for fixed symmetric positive definite (proportional and derivative) gain matrices. In this paper, we show that global asymptotic stability also holds for a class of gain matrices depending on the manipulator state. This feature increases the potential of the computed-torque control scheme to handle practical constraint in actual robots such as presence of friction in the joints and actuators with limited torque capabilities. We illustrate this potential by means of a fuzzy self-tuning algorithm to select the proportional and derivative gains according to the actual tracking position error. Experiments on a two degrees of freedom robot arm show the usefulness of the proposed approach.

One of the simplest and naturally appealing adaptive tracking controllers for robot manipulators is the so-called adaptive PD controller with computed feedforward. Since its introduction, in the middle of eighties, to the best of author knowledge, only local convergence analyses have been presented. In this paper, based on some properties of the robot dynamics, which deal with continuously differentiable increasing saturating functions, we present the global convergence analysis of this adaptive tracking controller.

Computed-torque control is a well known motion control strategy for manipulators which ensures global asymptotic stability for fixed symmetric positive definite (proportional and derivative) gain matrices. We show that the global asymptotic stability attribute also holds for a class of gain matrices depending on the manipulators state. This feature increases the potential of the computed-torque scheme to handle practical constraints in actual robots such as presence of friction in the joints and actuators with limited torque capabilities. We illustrate this potential by means of a fuzzy self-tuning algorithm to select the proportional and derivative gains according to the actual tracking position error. Experiments on a two degrees of freedom robot arm shown the usefulness of the proposed approach.

In this paper we propose a class of nonlinear PID global regulators for manipulators which do not include their dynamics in the central law. The structure of this class of global regulators is based on a linear PD feedback, plus an integral action driven by a nonlinear function of the joint position error, and the addition of the gradient of a C/sup 1/ artificial potential function satisfying some typical features required by the energy shaping methodology. By using Lyapunov's direct method and LaSalle's invariance principle we provide simple explicit conditions on the regulators gains to ensure global asymptotic stability.

In this paper we present in a simple and unified framework, the asymptotic stability analysis for energy shaping plus damping injection based set-point controllers for rigid and elastic joint robot manipulators by involving passivity theorems. Such an analysis is made for both cases: when position and velocity measurements are available for feedback purposes, and when only position measurements are available. To this end we use a particular case of a theorem introduced by Byrnes et al. (1991) and we present an adaptation, for asymptotic stability purposes, of a passivity theorem given by Canudas de Wit et al. (1996) and Khalil (1996).

In this paper, we propose a simple controller: the saturated P-saturated D controller with gravity compensation to solve the global regulation problem under actuator torque constraints. The performance of the proposed controller is illustrated via simulations on a two degree-of-freedom robot.

Proposes a systematic methodology, based on the energy shaping framework, to find strict Lyapunov functions for a class of global regulators for robot manipulators. The class of controller is described by control laws composed by a gradient of an artificial potential energy plus a linear velocity feedback. The authors provide explicit sufficient conditions on the artificial potential energy which allows them to obtain in a straightforward manner a strict Lyapunov function. Based on this technique the authors establish strict Lyapunov functions for well known regulators and propose a new one.