Measurement of three-dimensional welding torch orientation for manual arc welding process

Abstract

Torch orientation plays an important role in welding quality control for a manual arc welding process. The detection of the torch orientation can facilitate weld monitoring, welder training, and may also open a door to many other interesting and useful applications. Yet, little research has been done in measuring the torch orientation in the manual arc welding process. This paper introduces a torch orientation measurement scheme that can be conveniently incorporated both in a real manual arc welding process and in a welder training system. The proposed measurement employs a miniature wireless inertial measurement unit (WIMU), which includes a tri-axial accelerometer and a tri-axial gyroscope. A quaternion-based unscented Kalman filter (UKF) has been designed to estimate the three-dimensional (3D) torch orientation, in which the quaternion associated with the orientation is included in the state vector, as is the angular rate measured by the gyroscope. In addition, an auto-nulling procedure has been developed where the WIMU drift and measurement noise are captured and adaptively compensated in-line to ensure the measurement accuracy. The performance of the proposed scheme has been evaluated by simulations and welding experiments with different types of processes and fit-ups. The simulation results show that the inclination (x- and y-axes) of the torch has been accurately measured with a root-mean-square error (RMSE) in the order of 0.3°. The major error obtained in the heading (z-axis) measurement has been reduced significantly by the proposed auto-nulling procedure. Statistics from welding experiments indicate the proposed scheme is able to provide a complete 3D orientation measurement with the RMSE in the order of 3°.

A mobile sensing system for real-time 3D weld pool surface measurement in manual GTAW

“…The calibration procedure for the two IMUs is performed according to previous studies [45]. Using the IMUs, a Kalmanbased algorithm is proposed in [46] to provide an accurate estimation of the orientations of the torch and shield glass.…”

“…The IMUs' positions in the initial setup can be easily measured off-line. In manual GTAW, accurate orientations of the torch and the helmet can be calculated in real-time using the Kalman-based algorithm in [46]. The obtained orientations are denoted as G A1 , and G A2 , respectively.…”

Towards a Uniform Welding Quality: A Novel Weaving Welding Control Algorithm Based on Constant Heat Input

“…Research on robotic autonomous welding shows that guaranteeing weld quality in the groove edge area is more difficult than in the central groove area [ 21 , 22 ]. Variation in the inclination of the welding torch causes this issue, making it difficult to ensure weld quality [ 23 ]. However, the problem is usually overlooked by current robot welding planning methods.…”

Enhancing the accuracy of in-process springback measurements of complex tube bending processes using cost-effective embedded sensors

“…However, only a few applications in manufacturing related to in-process measurements have been reported in the scientific literature. In most cases, IMUs are used as a tracking device, in manual operations such as fastening [19], gas metal arc welding [20], or for in-field ground hole drilling [21]. Alternatively, Vogl et al [22] developed an IMU-based method for diagnostic of cutting machine tool linear axes.…”