Feedback Control of Robot End-Effector Probable Position Error

1991 ◽  
pp. 1178-1183
Author(s):  
Y. C. Pao ◽  
L. C. Chang
Keyword(s):  
Feedback Control ◽  
Position Error ◽  
End Effector

Task-Space Control Design and Performance Evaluation for a 6DOF Stewart Nanoscale Platform

2008 ◽  
Author(s):  
Chun-Chung Li ◽  
Yung Ting ◽  
Yi-Hung Liu ◽  
Yi-Da Lee ◽  
Chun-Wei Chiu
Keyword(s):  
Feedback Control ◽  
Hybrid Control ◽  
Feedforward Control ◽  
Feedback Controller ◽  
Test Method ◽  
Computation Method ◽  
Task Space ◽  
Force Output ◽  
End Effector ◽  
Control Scheme

A 6DOF Stewart platform using piezoelectric actuators for nanoscale positioning objective is designed. A measurement method that can directly measure the pose (position and orientation) of the end-effector is developed so that task-space on-line control is practicable. The design of a sensor holder for sensor employment, a cuboid with referenced measure points, and the computation method for obtaining the end-effector parameters is introduced. A control scheme combining feedforward and feedback is proposed. The inverse model of a hysteresis model derived by using a dynamic Preisach method is used for the feedforward control. Hybrid control to maintain both the positioning and force output for nano-cutting and nano-assembly applications is designed for the feedback controller. The optimal gain of the feedback controller is searched by using relay feedback test method and genetic algorithm. In experiment, conditions with/without external load employed with feedforward, feedback, and feedforward with feedback control schemes respectively are carried out. Performance of each control scheme verifies the capability of achieving nanoscale precision. The combined feedforward and feedback control scheme is superior to the others for gaining better precision.

Combining Motion Primitives and Image-Based Visual Servo Control

2020 ◽  
Author(s):  
Ghananeel Rotithor ◽  
Ashwin P. Dani
Keyword(s):  
Feedback Control ◽  
Open Loop ◽  
Servo Control ◽  
Visual Servo ◽  
Motion Generation ◽  
End Effector ◽  
Visual Servo Control ◽  
Dynamic Movement ◽  
Motion Primitive ◽  
Dynamic Movement Primitives

Abstract Combining perception feedback control with learning-based open-loop motion generation for the robot’s end-effector control is an attractive solution for many robotic manufacturing tasks. For instance, while performing a peg-in-the-hole or an insertion task when the hole or the recipient part is not visible in the eye-in-the-hand camera, an open-loop learning-based motion primitive method can be used to generate end-effector path. Once the recipient part is in the field of view (FOV), visual servo control can be used to control the motion of the robot. Inspired by such applications, this paper presents a control scheme that switches between Dynamic Movement Primitives (DMPs) and Image-based Visual Servo (IBVS) control combining end-effector control with perception-based feedback control. A simulation result is performed that switches the controller between DMP and IBVS to verify the performance of the proposed control methodology.

scholarly journals Visual end-effector position error compensation for planetary robotics

2007 ◽  
Vol 24 (5) ◽  
pp. 399-420 ◽  
Author(s):  
Max Bajracharya ◽  
Matthew DiCicco ◽  
Paul Backes ◽  
Kevin Nickels
Keyword(s):  
Error Compensation ◽  
Position Error ◽  
End Effector ◽  
Planetary Robotics

Path Planning of Avoiding Obstacle for Fruit-Picking Manipulator Based on Position Information

2011 ◽  
Vol 383-390 ◽  
pp. 1318-1325
Author(s):  
Yu Fei Wang ◽  
Jian Jun Yin ◽  
Shan Feng Chen
Keyword(s):  
Path Planning ◽  
Joint Angle ◽  
Position Error ◽  
Planning Method ◽  
Joint Angles ◽  
Position Information ◽  
End Effector ◽  
Obstacle Avoiding

Motoman SV3X manipulator was used to pick fruit and chosen as research object. Under the instance of known space coordinates of fruit and space position and pose of erect obstacle, a kind of multi-joint angle computation and path planning method for Motoman SV3X was proposed. Firstly, several sets of joint angles were computed according to relative space position between fruit and erect obstacle. And then, one set of joint angle was chosen as terminal angle of path planning. Finally, algorithm was used to plan a obstacle-avoiding path from start angle to terminal angle. The method solved the computation problem of multi-joint angle, and has universality for the problem of avoiding erect obstacle for fruit-picking manipulator. Simulation tests under the environment of Matlab showed that solved joint angles of manipulator can both make Motoman SV3X avoid erect obstacle and end-effector reach preconcerted position with given position error, and validated the correctness of the path planning method.

Sign in / Sign up

Export Citation Format

Share Document