The Neural Network Applications to Control of Robot Manipulators

IFAC Proceedings Volumes, 1996

The paper presents a neural network controller design for trajectory tracking for manipulators. Lyapunov' s stability theory was used to design the adaptive law for a computed torque method based artificial neural network controller. A joint space control scheme was used to test this type of controller. The controller was tested through simulation experiments and on the laboratory two D.O.F. SeARA mechanism.

International Journal of Computer Science And Technology, 2013

This technical paper serves to provide a basic study into the field of Neural Networks. We will discuss the various architectures used in neural networks and provide an in-depth discussion into the Back Propagation Algorithm. Lastly, we will apply the back propagation algorithm to an autonomous control system to serve as an automatic guidance system. Neural networks can thus be applied in the control of motion of robot and serve as its automatic guidance system.

IEEE Transactions on Neural Networks, 2002

This paper presents an approach and a systematic design methodology to adaptive motion control based on neural networks (NNs) for high-performance robot manipulators, for which stability conditions and performance evaluation are given. The neurocontroller includes a linear combination of a set of off-line trained NNs (bank of fixed neural networks), and an update law of the linear combination coefficients to adjust robot dynamics and payload uncertain parameters. A procedure is presented to select the learning conditions for each NN in the bank. The proposed scheme, based on fixed NNs, is computationally more efficient than the case of using the learning capabilities of the neural network to be adapted, as that used in feedback architectures that need to propagate back control errors through the model (or network model) to adjust the neurocontroller. A practical stability result for the neurocontrol system is given. That is, we prove that the control error converges asymptotically to a neighborhood of zero, whose size is evaluated and depends on the approximation error of the NN bank and the design parameters of the controller. In addition, a robust adaptive controller to NN learning errors is proposed, using a sign or saturation switching function in the control law, which leads to global asymptotic stability and zero convergence of control errors. Simulation results showing the practical feasibility and performance of the proposed approach to robotics are given.

Advances in intelligent systems and computing, 2017

The paper deals with motion control of autonomous robot. For an autonomous robot the main functionality which is to be implemented is its movement. The robot should be able to move from source to destination successfully avoiding all the obstacles in a known or unknown environment. This paper explains in detail 3 approaches for the motion control: (1) Neural Network where the problem is divided into sub problems FindSpace and FindPath (2) ANFIS (Adaptive Neuro-Fuzzy Inference System) where 6 layers are present (3) Fuzzy Logic along with neural network. Some simulation results are given which shows that Neural Network and fuzzy logic together gives better performance than Neural Network and fuzzy logic alone.

Industrial Electronics …, 1991

The authors present a nonlinear compensator using neural networks for trajectory control of robotic manipulators. The neural networks are not used to learn inverse-dynamics but to compensate nonlinearities of robotic manipulators. The performance of the ...

International Journal of Intelligent Systems and Applications, 2012

In robotic applications and research, inverse kinematics is one of the most important problems in terms of robot kinematics and control. Consequently, finding the solution of Inverse Kinematics in now days is considered as one of the most important problems in robot kinematics and control. As the intricacy of robot manipulator increases, obtaining the mathematical, statistical solutions of inverse kinematics are difficult and computationally expensive. For that reason, now soft-computing based highly intelligent based model applications should be adopted to getting appropriate solution for inverse kinematics. In this paper, a novel application of artificial neural network is used for controlling a robotic manipulator. The proposed methods are based on the establishments of the nonlinear mapping between Cartesian and joint coordinates using multi layer perceptron and functional link artificial neural network.

[Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics, 1992

A neural network robot controller is proposed. A n analysis of the closed loop system stability is given along with simulation results of the proposed approach.

Artificial Intelligence in Real-Time Control 1994, 1995

This peper describes the use of neural networks in diferent domains of robot control. Three robot control problems, relevant to a broad range of robotics applications, are analyzed, with a review of the state of the art and a description of current research by the authors, highlighting the advantages of the use of neural networks with respect to conventional techniques.

International Conference on Robotics and Automation, 2000

In this paper, a neural network controller for constrained,robot manipulators ,is presented. A feed- forward neural network is used to adaptively compensate for the uncertainties in the ,robot dynamics. Training signals are proposed for the feed-forward neural network controller. The neural network weights are tuned on-line, with no off-line learning phase required. It is shown,that the controller is able

This paper presents the design and practical implementation of inverse neural controller which is used to control the operation of six Degree Of Freedom (6DOF) robotic manipulator. An efficient off-line training method has been proposed which is used to train the neural network controller to be used as fed forward controller in the real time applications without need to the on-line training which is time consumption method. All the control algorithms and real time programming had been written with the aid of the MATLAB software.