Robotic weld groove scanning for large tubular T-joints using a line laser sensor

Cibicik, Andrej; Njaastad, Eirik B.; Tingelstad, Lars; Egeland, Olav

doi:10.1007/s00170-022-08941-7

Cited by 8 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The actual groove angle, denoted as α, is the sum of α1 and α2. The coordinate system for the intersecting structure of circular pipes and the welding groove section are illustrated in Figure 1 and 2 [17] . The polar coordinates of any point (X1, Y1, Z1) on the branch pipe weld seam are denoted as (r, δ), which can be determined through spatial analytic geometry [18][19] .…”

Section: Actual Groove Modelmentioning

confidence: 99%

See 1 more Smart Citation

Bead Layout and Error Rectification of Groove Weld at Intersecting Structures

Chen,

Guo,

Zhou

et al. 2024

Preprint

View full text Add to dashboard Cite

The welding of intersecting structures requires multi-layer multi-pass welding. Manual welding of intersecting curves is being replaced by robotic welding due to its low efficiency and poor consistency. To enhance weld accuracy in intersecting structures and optimize multi-layer multi-pass planning in robot welding of large components, a realistic groove model is devised to address discrepancies between theoretical and actual groove surfaces. This include determining the size of the actual error angle, and the modeling of the actual groove in the intersecting structure. A method for groove error correction is then proposed based on this model. Firstly, the total and the error areas of the actual groove are calculated. Subsequently, the errors for each layer of the groove are computed from the total error area of the groove. Error correction for the groove layout is implemented layer by layer, resulting in an algorithm for multi-layer multi-pass groove layout correction. Additionally, a fitting function that closely approximates the groove profile is derived. This procedure finalizes the groove error correction, and through MATLAB simulation, the corrected intersecting structure groove is modeled, with results aligning with the error correction algorithm. Finally, further simulation and experimentation validate the feasibility, accuracy and effectiveness of the error correction algorithm for multi-layer multi-pass weld planning the actual intersecting structure groove, offering a theoretical foundation for the intricate weld planning required in robotic welding of intersecting components.

show abstract

Section: Actual Groove Modelmentioning

confidence: 99%

“…The combined area of error for the front v-groove layer can be determined by intersecting Equation (17) through Equation ( 18):…”

Section: Total Area Errormentioning

confidence: 99%

Bead Layout and Error Rectification of Groove Weld at Intersecting Structures

Chen,

Guo,

Zhou

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The performance of the feature extraction method was evaluated using real-world data collected in [7] and [8]. The method was implemented using Python and the deep learning framework Pytorch and was evaluated in terms of accuracy and execution time compared to the results obtained by the algorithm used in [7], [8].…”

Section: Experimental Study and Comparisonmentioning

confidence: 99%

“…This dataset was collected by intentionally creating conditions for high noise levels. The second set [8], collected from an actual large welded tubular joint, contained over 650 point clouds with less noise, see Fig. 1 and Fig.…”

Section: Experimental Study and Comparisonmentioning

confidence: 99%

CNN-based Feature Extraction for Robotic Laser Scanning of Weld Grooves in Tubular T-joints

Mjolhus

Cibicik

Njaastad

et al. 2022

2022 Sixth IEEE International Conference on Robotic Computing (IRC)

Self Cite

View full text Add to dashboard Cite

This paper presents an algorithm for feature point extraction from scanning data of large tubular T-joints (a subtype of a TKY joint). Extracting such feature points is a vital step for robot path generation in robotic welding. Therefore, fast and reliable feature point extraction is necessary for developing adaptive robotic welding solutions. The algorithm is based on a Convolutional Neural Network (CNN) for detecting feature points in a scanned weld groove, where the scans are done using a laser profile scanner. To facilitate fast and efficient training, we propose a methodology for generating synthetic training data in the computer graphics software Blender using realistic physical properties of objects. Further, an iterative feature point correction procedure is implemented to improve initial feature point results. The algorithm's performance was validated using a real-world dataset acquired from a large tubular T-joint.

show abstract

“…As it was previously mentioned, reflections are a common problem for image sensors in general, and laser line scanners in particular, when images are captured of reflective metal surfaces. Several published strategies could potentially be used to minimize the reflections in a captured image by means of: a matte coat or sensor position planning [27]- [29], stereo cameras, or fringe projection [30], [31], lens-mounted polarizer filter, or color separation [4], [32], Time of Flight, High Dynamic Range, or motion contrast sensors [33]- [35], as well as image-based methods, or machine learning [36], [37].…”

Section: Introductionmentioning

confidence: 99%

Polarization Image Laser Line Extraction Methods for Reflective Metal Surfaces

2022

Self Cite

View full text Add to dashboard Cite

In this work, we propose a novel pipeline method for laser line extraction from images with a polarization image sensor. The proposed method is specially developed for strong laser beam reflections from metal surfaces. For the pre-processing stage, we propose a demosaicing algorithm for color polarizer filter array (CPFA) sensors. This can be implemented by using either one quarter or full resolution of the sensor. In addition, we propose two methods for optimizing the information available in a 12-channel color polarization image: The first method, is based on the minimum linearly polarized irradiance, and the second method, is based on the linear polarization intensity. These preprocessing, and optimization methods are combined with laser line extraction methods. The laser line extraction is done with either the Polarized Finite Impulse Response (FIR) Center Of Gravity (COG), where the laser line coordinates are computed from the filtered laser intensity distribution, or with the Polarized FIR-Peak, where the laser line coordinates are calculated from the first derivative of the filtered laser signal. The performance of the proposed algorithms is studied experimentally using a laser line scanner assembly, made of a polarization camera, and a laser line projector operating in the blue wavelength range.

show abstract

Robotic weld groove scanning for large tubular T-joints using a line laser sensor

Cited by 8 publications

References 28 publications

Bead Layout and Error Rectification of Groove Weld at Intersecting Structures

Bead Layout and Error Rectification of Groove Weld at Intersecting Structures

CNN-based Feature Extraction for Robotic Laser Scanning of Weld Grooves in Tubular T-joints

Polarization Image Laser Line Extraction Methods for Reflective Metal Surfaces

Contact Info

Product

Resources

About