Teaching and learning of deburring robots using neural networks

Author

Liu, Sheng ; Asada, Haruhiko

Author_Institution

Dept. of Mech. Eng., MIT, Cambridge, MA, USA

fYear

1993

fDate

2-6 May 1993

Firstpage

339

Abstract

A method for applying advanced robot adaptive control to manufacturing processes is described. A teaching method for constructing a sensor-based, task-level adaptive control system is described. Adaptive control laws that elucidate human motions are identified and stored in a multi-layer neural network. The resultant task performance is evaluated, and the relationship between the human actions and the performance index is stored in a second neural network. Based on the initial teaching data, the robot begins to perform a task. While performing a task repeatedly, the robot acquires additional data and improves its performance. Errors with respect to the performance index are propagated through the second network to modify the adaptive control law represented by its performance and excel the human operator who has provided the initial teaching data. A proof-of-concept demonstration and simulation are presented

Keywords

adaptive control; feedforward neural nets; industrial robots; learning (artificial intelligence); machining; manufacturing computer control; performance index; adaptive control; deburring robots; learning; manufacturing processes; multilayer neural net; neural networks; performance index; sensor based control; task level control; Adaptive control; Deburring; Education; Educational robots; Humans; Manufacturing processes; Multi-layer neural network; Neural networks; Performance analysis; Robot sensing systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Automation, 1993. Proceedings., 1993 IEEE International Conference on

Conference_Location

Atlanta, GA

Print_ISBN

0-8186-3450-2

Type

conf

DOI

10.1109/ROBOT.1993.292197

Filename

292197