Neural-Network Based Learning Control of Flexible Mechanism With Application to a Single-Link Flexible Arm

1994 ◽  
Vol 116 (4) ◽  
pp. 792-795 ◽  
Author(s):  
Kazuhiko Takahashi ◽  
Ichiro Yamada
Keyword(s):  
Neural Network ◽  
Angular Position ◽  
Target System ◽  
Space Representation ◽  
Robot Arm ◽  
Flexible Robot ◽  
Neural Network Controller ◽  
Flexible Arm ◽  
Single Link ◽  
Flexible Mechanism

This paper shows the effectiveness of a neural-network controller for controlling a flexible mechanism such as a flexible robot arm. An adaptive-type direct neural controller is formulated using state-space representation of the dynamics of the target system. The characteristics of the controller are experimentally investigated by using it to control the tip angular position of a single-link flexible arm.

Comparison of models for a single-link flexible robot ARM

1994 ◽  
Vol 27 (14) ◽  
pp. 415-420
Author(s):  
E. Bove ◽  
S. Nicosia ◽  
M. Simonelli
Keyword(s):  
Robot Arm ◽  
Flexible Robot ◽  
Single Link

Dynamic Finite Element Analysis of the Position Control System of a Two-Link Horizontal Flexible Robot

1990 ◽  
Vol 2 (2) ◽  
pp. 83-90
Author(s):  
Hiroyuki Kojima ◽  
Keyword(s):  
Finite Element ◽  
Control System ◽  
Position Control ◽  
Velocity Curve ◽  
Element Formulation ◽  
Robot Arm ◽  
Flexible Robot ◽  
Element Analysis ◽  
Flexible Arm ◽  
Position Control System

In this paper, a finite element formulation method for a horizontal flexible robot arm with two links is first presented. In the analysis, the kinetic energy of the flexible arm is represented in brief compared with previous methods, and the matrix equation of motion in consideration of the nonlinear forces, such as the Coriolis force, is derived by the finite element method and the variational theorem. Then, the state equation of the mechatronics system consisting of the flexible arm and the position control system is obtained. Secondly, numerical simulations in the case of applying path control based on the trapezoidal velocity curve are carried out by use of the Wilson-<I>θ</I> method, and the effects of the bending rigidity and the shape of the trapezoidal velocity curve on the dynamic characteristics of the mechatronics system are demonstrated.

scholarly journals Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method

Electronics ◽  
2021 ◽  
Vol 10 (7) ◽  
pp. 767
Author(s):  
Taekgyu Lee ◽  
Yeonsik Kang
Keyword(s):  
Neural Network ◽  
Model Predictive Control ◽  
Predictive Control ◽  
Nonlinear Model ◽  
Deep Neural Network ◽  
Nonlinear Model Predictive Control ◽  
Control Technique ◽  
Computational Time ◽  
Target System ◽  
Neural Network Controller

Nonlinear model predictive control (NMPC) is based on a numerical optimization method considering the target system dynamics as constraints. This optimization process requires large amount of computation power and the computation time is often unpredictable which may cause the control update rate to overrun. Therefore, the performance must be carefully balanced against the computational time. To solve the computation problem, we propose a data-based control technique based on a deep neural network (DNN). The DNN is trained with closed-loop driving data of an NMPC. The proposed "DNN control technique based on NMPC driving data" achieves control characteristics comparable to those of a well-tuned NMPC within a reasonable computation period, which is verified with an experimental scaled-car platform and realistic numerical simulations.

Sign in / Sign up

Export Citation Format

Share Document