Session: CIE-22-01 CAPPD: Product and Process Design Automation for Industry 4.0

Paper Number: 68237

Start Time: August 17, 10:00 AM

68237 - Finding Features of Positioning Error for Large Industrial Robots Based on Convolutional Neural Network 

Because most industrial robots are taught using the teaching playback method, they are unsuitable for variable production systems. Although off-line teaching methods have been developed, they have not been put to practical use because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have attempted to calibrate the position and posture, but have not reached a practical level because such methods only consider the joint angle when the robot is stationary and do not consider the features during robot motion. It is presently easy to obtain servo information under numerical control operations owing to the Internet of Things technologies. In this paper, we propose a method for obtaining servo information during robot motion and converting it into images to find features using a convolutional neural network (CNN). A large industrial robot was used. The three-dimensional coordinates of the end-effector were obtained using a laser tracker. The positioning error of the robot was accurately learned by the CNN. We extracted the features of the points where the positioning error was extremely large. By extracting the features of the X-axis positioning error using the CNN, the joint 1 current was a feature. This indicates that the vibration current in joint 1 is a factor of the X-axis positioning error.

Presenting Author: Daiki Kato Doshisha University

Authors:

Daiki Kato Doshisha University
Kenya Yoshitugu Doshisha University
Naoki Maeda Doshisha University
Toshiki Hirogaki Doshisha University
Eiichi Aoyama Doshisha University
Kenichi Takahashi IHI Corporation